A loader load history prediction method based on simulation data driving

Abstract

The present application relates to the technical field of data simulation analysis, and more particularly to a loader load history prediction method based on simulation data driving. The method comprises the following steps: establishing a loading lifting capacity optimization mechanics relationship model based on the structural parameters of a skidder; establishing a multi-domain loader oil circuit-dynamics simulation coupling model; performing a loader lifting mechanics collaborative optimization processing on the multi-domain loader oil circuit-dynamics simulation coupling model through the loading lifting capacity optimization mechanics relationship model to obtain a loader working condition multi-domain collaborative simulation model. The loader working condition multi-source load data is transmitted to the loader working condition multi-domain collaborative simulation model to analyze the loader multi-domain collaborative simulation load history characteristic data; a precision optimization load history prediction mapping relationship of the loader multi-domain collaboration is established based on the loader multi-domain collaborative simulation load history characteristic data to generate an optimized loader multi-domain collaborative load history prediction model. The present application realizes accurate loader load history prediction.

Description

Technical Field

[0001] This invention relates to the field of data simulation and analysis technology, and in particular to a method for predicting the load history of loaders based on simulation data. Background Technology

[0002] Loaders, as core mechanical equipment in engineering, are widely used in mining, infrastructure construction, and material handling. In actual operation, loaders often need to perform lifting, traveling, and loading / unloading operations under complex and variable working conditions, exhibiting highly nonlinear and dynamically changing stress environments. Due to the complexity of these conditions, the load history generated by loaders during operation is uncertain and random, which not only affects the loader's operating efficiency but also directly relates to its load operation stability. However, existing research on loader load history mainly relies on single theoretical mechanical modeling methods. These methods are often based on idealized assumptions and cannot accurately reflect the dynamic response of the hydraulic circuit, multibody dynamics, and the coupling effect of energy consumption and temperature during actual operation. Their prediction accuracy is limited, and they cannot meet the requirements for efficient lifting under heavy loads, high unloading heights, and large unloading distances. Furthermore, they cannot effectively address the issue of excessively high hydraulic circuit loads during the load history. Summary of the Invention

[0003] Based on this, the present invention provides a load history prediction method for loaders driven by simulation data to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a load history prediction method for loaders based on simulation data includes the following steps:

[0005] Step S1: Obtain the structural parameters of the skid steer loader; establish a mechanical mapping relationship analysis for loading and lifting capacity optimization based on the structural parameters of the skid steer loader to obtain a mechanical relationship model for loading and lifting capacity optimization;

[0006] Step S2: Establish a multi-domain loader hydraulic circuit-dynamics simulation coupled model; perform loader lifting mechanics co-optimization processing on the multi-domain loader hydraulic circuit-dynamics simulation coupled model through the loading lifting capacity optimization mechanical relationship model to obtain a multi-domain co-simulation model of loader working conditions;

[0007] Step S3: Obtain multi-source load data of the loader; transmit the multi-source load data of the loader to the multi-domain collaborative simulation model of the loader to perform load history feature analysis of the multi-domain collaborative simulation of the loader, so as to obtain load history feature data of the multi-domain collaborative simulation of the loader; establish the accuracy optimization load history prediction mapping relationship of the multi-domain collaborative simulation of the loader based on the load history feature data of the multi-domain collaborative simulation of the loader, and generate an optimized load history prediction model of the multi-domain collaborative simulation of the loader.

[0008] Step S4: Obtain the orientation signal parameters of the electric control handle; perform load monitoring and processing on the loader based on the orientation signal parameters of the electric control handle to obtain the load data of the loader; transmit the load data of the loader to the optimized multi-domain collaborative load history prediction model of the loader for load history prediction processing to obtain the load history prediction data of the loader; execute the intelligent auxiliary control operation of the loader load history based on the load history prediction data of the loader.

[0009] Furthermore, step S1 includes the following steps:

[0010] Step S11: Obtain the structural parameters of the skid steer loader;

[0011] Step S12: Perform multi-source structural type abstraction processing on the skid steer loader structural parameters to obtain multi-source skid steer loader structural abstract data;

[0012] Step S13: Based on the abstract data of the multi-source skid steer loader structure, perform data analysis and processing on the loader lifting characteristics and bucket trajectory to obtain the loader lifting characteristic data and bucket trajectory data, and extract the boom parameters and corresponding hinge point parameters of the skid steer loader structure from the abstract data of the multi-source skid steer loader structure.

[0013] Step S14: Perform topological analysis of the boom force characteristics based on the boom rod parameters and corresponding hinge point parameters of the skid steer loader structure to obtain the boom force characteristics topological data. Based on the boom force characteristics topological data, perform boom mechanical relationship modeling for loader lifting to obtain the loader lifting boom mechanical relationship model.

[0014] Step S15: Based on the loader lifting characteristic data and the loader lifting boom mechanical relationship model, perform a minimum lifting force characteristic analysis of the boom with maximum lifting capacity to obtain lifting capacity-boom lifting force characteristic data, and design a loading lifting optimization objective function based on the lifting capacity-boom lifting force characteristic data.

[0015] Step S16: Map the bucket trajectory data as loading and lifting constraints and the loading and lifting optimization objective function as loading and lifting optimization objective conditions to the loader lifting boom mechanical relationship model to perform mechanical mapping relationship analysis for loading and lifting capacity optimization, so as to obtain the loading and lifting capacity optimization mechanical relationship model.

[0016] Furthermore, step S2 includes the following steps:

[0017] Step S21: Obtain the loader's walking hydraulic parameters and corresponding loader walking dynamics data, as well as the loader's working hydraulic parameters and corresponding loader working dynamics data;

[0018] Step S22: Perform dynamic simulation analysis of the hydraulic circuit based on the loader's walking hydraulic parameters and working hydraulic parameters to obtain the loader's hydraulic circuit simulation model;

[0019] Step S23: Perform multibody dynamics simulation analysis of the loader based on the loader's walking dynamics data and working dynamics data to obtain the loader's multibody dynamics simulation model;

[0020] Step S24: Based on the hydraulic circuit component data of the loader and the multibody dynamics component data of the loader, perform multi-domain simulation model coupling processing to obtain a multi-domain loader hydraulic circuit-dynamics simulation coupled model;

[0021] Step S25: The lifting mechanics of the multi-domain loader hydraulic circuit-dynamic simulation coupling model is optimized by using the loading lifting capacity optimization mechanical relationship model to obtain the multi-domain collaborative simulation model of the loader working condition.

[0022] Furthermore, step S22 includes the following steps:

[0023] The hydraulic circuit components of the loader are analyzed based on the hydraulic parameters of the loader's travel and working to obtain the data of the hydraulic circuit components. The hydraulic circuit structure model of the loader is then established based on the data of the hydraulic circuit components. Based on the hydraulic parameters of the loader's travel and working, the hydraulic circuit structure model of the loader is subjected to dynamic simulation analysis to obtain the simulation model of the loader's hydraulic circuit.

[0024] Furthermore, step S23 includes the following steps:

[0025] Based on the loader's walking dynamics data and working dynamics data, the loader's drive components are analyzed to obtain the loader's drive component data. A multi-body dynamics structural model of the loader is then established using the loader's drive component data. Based on the loader's walking dynamics data and working dynamics data, a multi-body dynamics simulation analysis of the loader's multi-body dynamics structural model is performed to obtain a loader multi-body dynamics simulation model.

[0026] Furthermore, step S24 includes the following steps:

[0027] Based on the hydraulic circuit component data and multibody dynamics component data of the loader, the correlation characteristics of the hydraulic circuit and multibody dynamics components of the loader are analyzed to generate the correlation characteristic data of the hydraulic circuit and dynamics components of the loader. Then, the simulation model coupling processing of the hydraulic circuit simulation model and the multibody dynamics simulation model of the loader is performed through the correlation characteristic data of the hydraulic circuit and dynamics components of the loader to obtain the hydraulic circuit and dynamics simulation coupling model of the loader.

[0028] Furthermore, step S3 includes the following steps:

[0029] Step S31: Obtain multi-source load data of the loader;

[0030] Step S32: Transmit the multi-source load data of the loader's working condition to the multi-domain collaborative simulation model of the loader's working condition to perform load history characteristic analysis of the multi-domain collaborative simulation of the loader, so as to obtain the load history characteristic data of the multi-domain collaborative simulation of the loader.

[0031] Step S33: Perform load history change gradient feature analysis on the load history feature data of the multi-domain collaborative simulation of the loader, generate the load history change gradient feature data of the collaborative simulation, and establish the load history prediction mapping relationship of the multi-domain collaboration of the loader through the load history change gradient feature data of the collaborative simulation to obtain the load history prediction model of the multi-domain collaboration of the loader.

[0032] Step S34: Based on the multi-domain collaborative simulation load history characteristic data of the loader, perform energy consumption loss impact characteristic analysis on the simulation load history, and generate energy consumption loss impact characteristic data on the simulation load history.

[0033] Step S35: Extract and process the oil circuit temperature anomaly of the simulated load history based on the load history-energy consumption characteristic data of the loader collaborative simulation to obtain the simulated load history oil circuit temperature anomaly data. Analyze the simulation drive adjustment coefficient of the load history temperature protection feedback based on the preset temperature self-feedback protection engine to generate the temperature protection feedback simulation drive adjustment coefficient.

[0034] Step S36: Based on the temperature protection feedback simulation drive adjustment coefficient, adjust the energy consumption and loss influence characteristics of the simulated load history energy consumption and loss feature data to generate adjusted simulated load history energy consumption and loss influence feature data.

[0035] Step S37: Optimize the accuracy of the load history prediction mapping relationship of energy consumption loss in the multi-domain collaborative load history prediction model of the loader by using the characteristic data of energy consumption loss influence of the adjusted simulated load history, so as to obtain an optimized multi-domain collaborative load history prediction model of the loader.

[0036] Furthermore, step S34 includes the following steps:

[0037] Step S341: Based on the load history characteristic data of the multi-domain co-simulation of the loader, perform nonlinear mapping relationship analysis between the load history and energy consumption characteristics of the loader co-simulation, and generate load history-energy consumption characteristic data of the loader co-simulation.

[0038] Step S342: Extract the first simulation load history-energy consumption characteristic data and the second simulation load history-energy consumption characteristic data based on the loader collaborative simulation load history-energy consumption characteristic data;

[0039] Step S343: Analyze the energy consumption loss impact characteristics of the simulated load history using the first simulated load history-energy consumption characteristic data and the second simulated load history-energy consumption characteristic data, and generate simulated load history energy consumption loss impact characteristic data.

[0040] Furthermore, in step S342, the energy consumption value of the first simulated load history-energy consumption characteristic data is greater than the energy consumption value of the second simulated load history-energy consumption characteristic data, and / or, the time series interval of the first simulated load history-energy consumption characteristic data is greater than the time series interval of the second simulated load history-energy consumption characteristic data.

[0041] Furthermore, step S4 includes the following steps:

[0042] Step S41: Obtain the orientation signal parameters of the electric control handle;

[0043] Step S42: Perform load monitoring and processing on the loader operating condition based on the orientation signal parameters of the electric control handle to obtain loader operating condition load data. The loader operating condition load data includes load data under normal conditions and load data under extreme conditions. The load data under extreme conditions includes rated load data under straight driving conditions, rated load data under stationary full-circuit turning conditions, and loading and unloading conditions.

[0044] Step S43: Transmit the loader's working condition load data to the optimized loader multi-domain collaborative load history prediction model for loader load history prediction processing, including: when extreme condition load data of the loader is identified in the loader's working condition load data, the extreme condition load history accuracy of the loader's extreme condition load data is optimized by adjusting the energy consumption and loss influence feature data of the simulated load history built into the optimized loader multi-domain collaborative load history prediction model, and the extreme condition load history prediction data of the loader is output; when extreme condition load data of the loader is identified in the loader's working condition load data, the normal condition load data of the loader is output, and the normal condition load history prediction data of the loader is output; the extreme condition load history prediction data of the loader and the normal condition load history prediction data of the loader are integrated to obtain the loader load history prediction data.

[0045] Step S44: Execute intelligent auxiliary control of loader load history using loader load history prediction data.

[0046] The beneficial effects of this application are as follows: By performing multi-source abstract modeling and mechanical relationship analysis on the structural parameters of skid steer loaders, this invention can scientifically optimize the lifting capacity of the loader under different structural types and constraints. It not only considers geometric constraints such as boom parameters, hinge point parameters, and bucket trajectory, but also establishes a mapping relationship between lifting capacity and boom lifting force through topological force characteristic modeling. By introducing a lifting optimization objective function and trajectory constraints, it achieves quantitative and target-oriented design of lifting capacity, thereby accurately reflecting the lifting performance characteristics of skid steer loaders under different working conditions, better adapting to complex structures and diverse working conditions, improving the accuracy of early-stage modeling for load history prediction, and providing reliable mechanical optimization model support for subsequent multi-domain coupled simulation and prediction. By establishing a multi-domain loader hydraulic circuit-dynamics simulation coupled model, and introducing a mechanical relationship optimization model for collaborative optimization, it achieves deep integration and dynamic interaction between the hydraulic system and the dynamics system. This method effectively solves the problem of the separation between hydraulic simulation and dynamics simulation in traditional methods, enabling synchronous iterative calculation of oil circuit pressure, flow response, and multibody dynamic load changes, thus obtaining simulation results that more closely resemble actual working conditions. Simultaneously, the co-optimization process dynamically adjusts the correlation characteristics between the oil circuit and dynamics models, ensuring that the model retains both the transient response characteristics of the hydraulic system and the nonlinear dynamic characteristics of the mechanical structure, significantly improving the accuracy and stability of multi-domain simulation. The resulting multi-domain co-simulation model of the loader not only realistically reflects the load history changes under different working conditions but also provides a highly reliable simulation foundation for load prediction and intelligent control. By acquiring multi-source working condition load data and inputting it into the multi-domain co-simulation model for history feature analysis, the dynamic load evolution law of the loader can be accurately captured under multiple working conditions. In addition to considering multi-dimensional factors such as load gradient changes, energy consumption, and abnormal oil circuit temperature during load feature extraction, a temperature protection feedback adjustment mechanism is introduced, enabling the load prediction model to dynamically adapt to changes in complex operating environments and extreme conditions. In particular, the joint analysis of load history gradient and energy consumption impact characteristics significantly improves the accuracy and robustness of load prediction, avoiding error accumulation caused by single-feature modeling. Continuous optimization of the prediction model through self-feedback adjustment ensures not only high accuracy but also valuable data support for energy-saving control, fault warning, and durability design. By acquiring the electric control handle's orientation signal parameters in real time and combining them with an optimized multi-domain collaborative load history prediction model, online prediction and intelligent auxiliary control of the loader under different working conditions can be achieved. The prediction process distinguishes between normal and extreme working conditions, and an optimization mechanism for energy consumption impact characteristics is introduced for extreme working conditions, ensuring that the prediction results cover the entire working condition range and possess high reliability.By integrating prediction data under both normal and extreme conditions, a complete load history prediction result can be output, which can then drive the loader's auxiliary control strategy. This design not only allows for real-time adjustment of the loader's operating status, preventing equipment damage caused by overload or abnormal operation, but also plays a significant role in energy saving, improved operational efficiency, and extended equipment lifespan, achieving a closed-loop integration of load prediction and actual control.

[0047] Therefore, the load history prediction method for loaders based on simulation data proposed in this invention achieves high-precision prediction of the load history of loaders under complex operating conditions by introducing loader structural parameter modeling, multi-domain coupled simulation of hydraulic circuits and multibody dynamics, and collaborative analysis driven by multi-source working condition data. Compared with traditional single theoretical mechanics modeling methods, this invention can comprehensively consider the dynamic response of hydraulic circuits, the nonlinear characteristics of multibody dynamics, and the coupling effect between energy consumption and temperature, effectively overcoming the deviations caused by idealized assumptions. By constructing a mechanical relationship model for optimizing lifting capacity, this invention can meet the operational requirements of loaders for high-efficiency lifting of heavy loads, high unloading height, and large unloading distance while ensuring the accuracy of load prediction. Through the analysis of the nonlinear mapping relationship between simulated load history and energy consumption characteristics, combined with a temperature protection feedback mechanism, active adjustment and protection against excessively high temperatures in the hydraulic circuit are achieved, thereby improving the reliability and stability of loaders under complex operating conditions. Overall, this invention not only improves the accuracy and adaptability of loader load history prediction but also provides effective technical support for overall machine performance optimization, energy consumption control, and intelligent regulation. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the steps of a load history prediction method for loaders based on simulation data driven by the present invention.

[0049] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2.

[0050] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3.

[0051] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0053] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. Functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0054] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for predicting the load history of a loader based on simulation data. In the embodiments of this invention, please refer to... Figure 1 The diagram shown is a flowchart illustrating the steps of a load history prediction method for loaders based on simulation data, according to the present invention. The load history prediction method for loaders based on simulation data includes the following steps:

[0055] Step S1: Obtain the structural parameters of the skid steer loader; establish a mechanical mapping relationship analysis for loading and lifting capacity optimization based on the structural parameters of the skid steer loader to obtain a mechanical relationship model for loading and lifting capacity optimization;

[0056] In this embodiment of the invention, the structural parameters of the skid steer loader obtained from the database are systematically geometrically parameterized. The three-dimensional geometric topology of the body, boom, tie rod, hinge point, bucket, and connecting components is clearly defined, and each geometric component is parameterized using rigid rods, elastic rods, or beam elements. The hinge point position is represented using a local coordinate system, and a coordinate transformation matrix is ​​used to form the overall kinematic chain. Based on the above parameterized model, static / dynamic equilibrium equations for the boom forces are established, and the equilibrium expressions for hinge point torque, shear force, and axial force are listed. The stiffness matrix and damping matrix of the boom-bucket system are derived using the principle of virtual work or the Lagrange equation. Furthermore, the bucket trajectory constraint is added to the lifting mechanics model in the form of equality constraints, and the trajectory curve is parameterized as a time function and substituted into the dynamic equations as a motion boundary condition. Then, based on two objectives, max[min(F)], namely maximum lifting capacity and minimum boom lifting force, a convex / non-convex optimization objective function is constructed and constraints (mechanical limit of hinge point, geometric interference limit, trajectory boundary) are introduced. A numerical solution method is used to couple the mechanical equations and the optimization objectives. During the solution process, sensitivity analysis is used to calculate the sensitivity matrix of structural parameters to lifting capacity, so as to map the structural parameters to the lifting capacity output. Finally, a mechanical relationship model for lifting capacity optimization is derived in the form of a mapping relationship function. This model accepts structural parameter vectors and trajectory parameter vectors as inputs and outputs the boom internal force distribution, ultimate lifting force, and minimum lifting force characteristics for subsequent simulation coupling and control constraints.

[0057] Step S2: Establish a multi-domain loader hydraulic circuit-dynamics simulation coupled model; perform loader lifting mechanics co-optimization processing on the multi-domain loader hydraulic circuit-dynamics simulation coupled model through the loading lifting capacity optimization mechanical relationship model to obtain a multi-domain co-simulation model of loader working conditions;

[0058] In this embodiment of the invention, the hydraulic system is modeled at the component level. Port-type physical models are established for components such as pumps, proportional relief valves, directional valves, actuators, pipelines, and accumulators. Port variables are defined as pipeline pressure, flow rate, and volume. Valve port characteristic curves, the geometric relationship of cylinder volume change with displacement, and flow resistance relationships are given. The transient equations of the hydraulic circuit are derived using the mass-energy conservation equation and the compressibility model. A multibody dynamics model is established for the loader's rigid body and linkage mechanism. Rigid body dynamics equations, joint constraint equations, and contact models are used to describe the vehicle body motion, boom motion, and interaction with the bucket ground. The time distribution of external loads and inertial moments is given. Then, following the port coupling concept, a force-displacement / velocity interface is established between the hydraulic circuit model and the multibody dynamics model: the thrust generated by the pressure inside the cylinder is input to the multibody model, and the actuator displacement and velocity are used as boundary conditions for the hydraulic circuit model. The two domains are solved implicitly in each simulation step until the coupling residual is lower than a preset convergence threshold. Next, the lifting capacity optimization mechanical relationship model is embedded into a coupled simulation loop in the form of parameter constraints and objective functions. The lifting capacity objective is used to iteratively adjust the valve opening strategy, pump flow distribution, and cylinder bore ratio. The adjustment process uses gradient-based parameter updates or a solver based on the quasi-Newton method to ensure that the coupled model maintains numerical stability while satisfying the mechanical objective. The final multi-domain collaborative simulation model of the loader's working conditions includes: a hydraulic transient sub-model, a multibody dynamics sub-model, a two-domain coupling interface, and a lifting mechanical constraint module. It outputs time-domain load, valve position / flow curves, cylinder pressure curves, and boom mechanical parameters.

[0059] Step S3: Obtain multi-source load data of the loader; transmit the multi-source load data of the loader to the multi-domain collaborative simulation model of the loader to perform load history feature analysis of the multi-domain collaborative simulation of the loader, so as to obtain load history feature data of the multi-domain collaborative simulation of the loader; establish the accuracy optimization load history prediction mapping relationship of the multi-domain collaborative simulation of the loader based on the load history feature data of the multi-domain collaborative simulation of the loader, and generate an optimized load history prediction model of the multi-domain collaborative simulation of the loader.

[0060] In this embodiment of the invention, the types and synchronization schemes of multi-source load data are defined. Data types include: bucket contact force history, boom hinge torque, hydraulic cylinder pressure timing, vehicle acceleration / angular velocity, ground condition segments (hardness, slope), and oil temperature and ambient temperature. A unified high-precision timestamp is used as the time reference, and timing alignment is achieved through cross-correlation. Signal preprocessing is performed: band-limited filtering and baseline drift correction are applied to high-frequency noise. Segmented windows are used to extract time-domain features (peak value, RMS value, rise / fall slope), frequency-domain features (short-time Fourier or wavelet packet energy distribution), and statistical features (kurtosis, skewness) from the history signal. Simultaneously, load categories and operating modes are labeled for each work condition segment. Then, the preprocessed multi-source load data is input into a multi-domain collaborative simulation model. Segmented simulations are performed according to the work condition sequence, and the results are compared with the actual measured history. The simulation-measured residual timing is calculated, and a load history feature difference matrix is ​​generated accordingly. Based on the difference matrix, a load history prediction mapping relationship is constructed: a parameter estimation framework is used to solve the mapping model parameters by least squares or minimization with regularization terms to reduce the mean square error between simulation output and actual measurement; during the mapping training process, energy consumption-history nonlinear mapping decomposition is introduced, and energy consumption sensitive segments are modeled separately to extract the time series of the first and second types of energy consumption features and form energy consumption factors; the prediction mapping is cross-validated and the accuracy optimization coefficient is obtained through gradient correction, and finally an optimized multi-domain collaborative load history prediction model for loaders is formed. This model takes the preprocessed feature vector as input and outputs the load history curve within the future time window and carries the energy consumption influencing factor information.

[0061] Step S4: Obtain the orientation signal parameters of the electric control handle; perform load monitoring and processing on the loader based on the orientation signal parameters of the electric control handle to obtain the load data of the loader; transmit the load data of the loader to the optimized multi-domain collaborative load history prediction model of the loader for load history prediction processing to obtain the load history prediction data of the loader; execute the intelligent auxiliary control operation of the loader load history based on the load history prediction data of the loader.

[0062] In this embodiment of the invention, an analytical framework for the orientation signal parameters of the electronic control handle is defined, including the forward and backward swing angles, left and right deflection angles, pedal opening, and time differential information of speed. The signals are then denoised and zero-point corrected, and Kalman filtering or adaptive filtering is used to smooth transient jitter. Based on the processed handle orientation parameters, a set of working condition identification rules is constructed. This set employs a hybrid judgment logic based on threshold and temporal pattern matching to identify normal and extreme working conditions (rated load for straight-line driving, rated load for full-circuit turning in place, and loading / unloading conditions). Upon triggering identification, a working condition label is generated, carrying an associated time window. Then, the identified working condition load data is transmitted according to the time window to an optimized multi-domain collaborative load history prediction model for the loader. After receiving the handle drive characteristics, the prediction model outputs short-term and medium-term load history prediction curves. When the prediction result is an extreme working condition, the prediction model internally calls the temperature protection feedback coefficient and energy consumption adjustment coefficient to optimize the accuracy of the prediction history, generating temperature protection trigger suggestions and valve position limit suggestions. Intelligent auxiliary control operations are performed based on load history prediction data. This control includes: issuing cylinder pressure buffer commands based on the predicted load curve, adjusting the pump output curve through the power distribution module, implementing critical amplitude limiting commands on the engine torque curve, and activating oil circuit temperature self-feedback protection logic under high load conditions such as turning in place. All control commands are sent to the vehicle control unit in real-time control protocol format and the feedback after closed-loop execution is recorded for subsequent online model calibration.

[0063] Furthermore, step S1 includes the following steps:

[0064] Step S11: Obtain the structural parameters of the skid steer loader;

[0065] In this embodiment of the invention, the structural parameters of the vehicle and the working mechanism are systematically defined. First, a global coordinate system is established on the vehicle body base, and parameter tables are created for the vehicle body, boom, connecting rods, hinge points, bucket, hydraulic cylinders, and connecting parts using this coordinate system as a reference. The parameter tables include geometric dimensional parameters (length, cross-sectional dimensions, relative coordinates of hinge points), material mechanical parameters (elastic modulus, Poisson's ratio, density), connection characteristic parameters (hinge stiffness, friction coefficient, installation clearance), center of mass and moment of inertia information, and manufacturing tolerance representation. Geometric parameter capture is performed according to the measurement process: point cloud measurements are performed on key components using a 3D scanning device; component boundaries are fitted using point cloud data, and geometric feature vectors are extracted from the boundaries; multiple measurements are taken at the hinge point positions and averaged to reduce measurement noise, and measurement uncertainties are recorded for subsequent sensitivity analysis. Mass and inertia parameters are estimated using a piecewise integration method; local centers of mass are calculated based on the cross-sectional profile, and the overall inertia is synthesized. All parameters are archived using a unified dimension, a unified reference surface, and a timestamp encoding, and verification items are established: geometric consistency checks are performed on geometric closed-loop constraints, extreme values ​​of motion range, and articulated interference; material parameters are verified through material certificates and compared with static loading tests to complete parameter confirmation. The output is a structural parameter vector and parameter confidence intervals, which are used in subsequent abstraction and mechanical modeling stages.

[0066] Step S12: Perform multi-source structural type abstraction processing on the skid steer loader structural parameters to obtain multi-source skid steer loader structural abstract data;

[0067] In this embodiment of the invention, the acquired structural parameter set is processed in a modular way. First, the multi-source structure is labeled according to functional units: four types of units are divided: vehicle platform, lifting mechanism, linkage mechanism, and actuator. A parameterized template library is established for each unit. The parameterized template adopts a hybrid representation method of rod system and beam system: the components are approximated as one-dimensional beam elements or rigid rod elements, and the cross-sections are given inertial and elastic properties. Hinges are represented by nodes and the constraint degrees of freedom are recorded. Isomorphic mapping is performed on the structural types from heterogeneous sources. The original geometry and component numbers are mapped to the template node and edge numbering scheme using topological mapping rules, thereby obtaining a unified abstract representation format. The continuous geometric quantities in the template are represented by basis function expansion. The basis function coefficients are used to describe the curvature and cross-sectional changes, realizing dimensionality reduction from specific configurations to low-dimensional parameter vectors. A versioned template index is established and each template is labeled with an applicable boundary condition set and limit value to distinguish between standard and reinforced components. The abstract output is a multi-source skid steer loader structure abstract dataset, including template ID, node coordinate sequence, edge attribute matrix, constraint matrix and basis function coefficients, providing a unified input interface for subsequent lifting characteristics and trajectory analysis.

[0068] Step S13: Based on the abstract data of the multi-source skid steer loader structure, perform data analysis and processing on the loader lifting characteristics and bucket trajectory to obtain the loader lifting characteristic data and bucket trajectory data, and extract the boom parameters and corresponding hinge point parameters of the skid steer loader structure from the abstract data of the multi-source skid steer loader structure.

[0069] In this embodiment of the invention, two parallel processing flows are applied to the abstracted parameters: the first is lifting characteristic analysis, establishing kinematic mapping relationships and deriving mechanical boundary behavior; the second is bucket trajectory extraction, forming a time-pose parameterized curve. Kinematic mapping is achieved through closed-chain solving: the coordinates of template nodes and hinge points are substituted into the kinematic constraint equations, and a numerical iterative algorithm is used to solve for the analytical or numerical solutions of joint angles with lifting angles, forming a joint angle time-series function. Based on the kinematic results, combined with component inertia and external load distribution, a force time-series equation set is established, and the torque and axial force at each hinge point under typical operating angles are obtained through static calculation, thereby generating a lifting characteristic data vector, including torque curves, stiffness variation curves, and dynamic response time constants under typical angles. The bucket trajectory is obtained through trajectory recognition and curve fitting: the pose sequence of the bucket relative to the ground is extracted from the working condition sample, the curve is smoothed using spline functions, and the trajectory is represented by time parameters; trajectory constraints include the bucket-ground contact angle, the bucket penetration depth threshold, and the soil clearing path boundary. The structural abstract data is used to extract boom parameters (boom length, cross-sectional distribution, end connection type) and hinge parameters (coordinates, constraint degrees of freedom, friction stiffness). These parameters are then backfilled into the mechanical modeling module through a mapping table and numbering consistency rules.

[0070] Step S14: Perform topological analysis of the boom force characteristics based on the boom rod parameters and corresponding hinge point parameters of the skid steer loader structure to obtain the boom force characteristics topological data. Based on the boom force characteristics topological data, perform boom mechanical relationship modeling for loader lifting to obtain the loader lifting boom mechanical relationship model.

[0071] In this embodiment of the invention, the boom parameters of the skid steer loader structure are abstracted into a mechanical topology network with nodes and edges. Nodes represent hinge points or connection points, and edges represent beam or rod elements, carrying cross-sectional mechanical properties and length information. Path analysis is performed on the topology network to determine the main force channels: the force transmission path from the bucket contact point to the vehicle support point is calculated, and key force nodes are identified. Based on the node-edge topology, the overall stiffness matrix and damping matrix are assembled, and the finite element approximation method is used to refine the mesh of key edges to solve for local stress concentration areas. The boom force analysis includes static solutions, quasi-static load cases, and frequency domain modal analysis: the static solution solves for hinge point forces and cross-sectional internal forces through equilibrium equations; the quasi-static load case calculates the force history of the boom at each point on the trajectory in discrete time steps; modal analysis obtains the boom's free mode shape and natural frequency to evaluate the dynamic response. The model includes material nonlinearity and contact nonlinearity treatment; the contact uses an elastoplastic contact model and a numerical contact algorithm to handle the switching term of contact forces. Based on the above analysis, a boom mechanics model is established. This model is driven by the bucket load timing and hydraulic cylinder input, and outputs the hinge point torque, rod internal force distribution and global displacement response. The model is expressed as a parameterized mapping function or a state-space equation set for use in lifting capacity optimization mapping.

[0072] Step S15: Based on the loader lifting characteristic data and the loader lifting boom mechanical relationship model, perform a minimum lifting force characteristic analysis of the boom with maximum lifting capacity to obtain lifting capacity-boom lifting force characteristic data, and design a loading lifting optimization objective function based on the lifting capacity-boom lifting force characteristic data.

[0073] In this embodiment of the invention, after obtaining the lifting characteristic data of the loader and the mechanical relationship model of the loader lifting boom, an extreme value analysis is performed on the maximum lifting capacity problem. A lifting force index is defined: under given displacement or trajectory constraints, the minimum static support force required by each hydraulic cylinder and hinge point to bring the bucket to a specified position is calculated. A local minimization algorithm is used to solve for this minimum support force at each discrete trajectory point. To fully characterize the process features, time-series statistics are performed on the lifting force within the trajectory time window, extracting the maximum value, mean value, lifting rate, and duration distribution. Based on this, an optimization objective function is constructed. This objective function is a multi-objective scalar expression, including a load-bearing margin term, a structural strength safety margin term, and an energy consumption sensitivity term. The objective function is presented in the form of a weighted linear combination and a penalty function term. Constraints include the upper limit of hinge point stress, member deflection limit, geometric interference constraint, and stability constraint (including the center of gravity projection not exceeding the support baseline). A constraint-based nonlinear programming numerical solution strategy is used to solve the objective function, guiding parameter updates through a sensitivity matrix to reduce the objective value. The output is a lifting capacity-boom lifting force feature dataset. This dataset contains the optimal solution trajectory, corresponding lifting force timing, and constraint boundary triggering records recorded during the objective function solution process, providing training and calibration samples for subsequent mapping relationships.

[0074] Step S16: Map the bucket trajectory data as loading and lifting constraints and the loading and lifting optimization objective function as loading and lifting optimization objective conditions to the loader lifting boom mechanical relationship model to perform mechanical mapping relationship analysis for loading and lifting capacity optimization, so as to obtain the loading and lifting capacity optimization mechanical relationship model.

[0075] In this embodiment of the invention, the trajectory parameterization of the bucket trajectory data is substituted as an equality constraint into the established boom mechanics model. The trajectory constraint is expressed in the time domain as a third-order continuity constraint of position, velocity, and acceleration, and is applied in parallel during numerical solution using Lagrange multipliers or penalty functions. The scalarized expression provided by the loading and lifting optimization objective function is embedded into the solution framework of the mechanics model, forming a constrained optimal control / optimal design problem. The numerical solution process is as follows: the time window is discretized, a global sparse Jacobian matrix of discrete dynamic equations and constraint equations is constructed, and sequential quadratic programming or Newton's iteration method with Jacobian correction is used for solution. During the iteration process, the control variables (initial hydraulic cylinder pressure, geometric ratio, valve flow distribution strategy) are judged and corrected according to the convergence criterion. To ensure the stability of the solution, incremental loading and arc length tracking strategies are used for nonlinear terms to cross possible critical points. After solving, the output is a mechanical relationship model for optimizing the lifting capacity. This model is a family of mapping functions that accepts structural parameters, trajectory parameters, and operational input vectors, and outputs the optimal lifting force distribution, the ultimate lifting capacity value, and a set of control suggestions for online applications. Finally, the output is validated through multi-condition simulations. Cross-condition residual analysis is used to evaluate the model's generalization ability and record the error distribution spectrum for subsequent model updates.

[0076] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment, step S2 includes the following steps:

[0077] Step S21: Obtain the loader's walking hydraulic parameters and corresponding loader walking dynamics data, as well as the loader's working hydraulic parameters and corresponding loader working dynamics data;

[0078] In this embodiment of the invention, each component of the traveling hydraulic system and the working hydraulic system is identified and a parameter list is established. These parameters include: the pump's transient flow-pressure characteristic curve, the pump spool valve position-flow coupling curve, the valve core flow channel area and valve port pressure drop function, the cylinder working stroke and piston area, pipeline length and equivalent damping, the volume of joints and valves, the accumulator capacity and pre-charge pressure, the hydraulic oil thermodynamic and viscosity parameters (bulk modulus, specific heat, viscosity as a function of temperature), and sealing and friction loss model parameters. Simultaneously, a measurement matrix is ​​established for the traveling dynamics and working dynamics. This measurement matrix includes: the total mass and moment of inertia of the vehicle body, the power transmission characteristics of the drive axle, wheel radius, drive gear ratio, vehicle body suspension stiffness and damping, the mass matrix of the boom and bucket, the hinge point damping and friction matrix, and a description of ground conditions (bearing modulus, friction coefficient, slope distribution). The parameter acquisition process combines fixed-point measurements on the test bench with measurements under actual field conditions: the static and quasi-static flow-pressure responses measured on the test bench are used to establish component-level characteristic curves; dynamic measurements employ synchronous sampling using high-frequency pressure sensors, displacement encoders, accelerometers, and torque sensors, with each sampling point timestamped using a high-precision time base for subsequent time alignment. All measurement signals undergo bandpass filtering, depolarization, and wavelet denoising before frequency and time domain feature extraction; measurement uncertainty is obtained through repeated experiments and encoded in the form of a covariance matrix for model parameter identification and sensitivity analysis. The output consists of parameter vectors and a dynamic time series sample set, with confidence intervals and constraint boundaries established for the parameter vectors for subsequent simulation model calibration.

[0079] Step S22: Perform dynamic simulation analysis of the hydraulic circuit based on the loader's walking hydraulic parameters and working hydraulic parameters to obtain the loader's hydraulic circuit simulation model;

[0080] In this embodiment of the invention, the hydraulic system is expressed using a port-based physical modeling method. Component modeling includes: establishing momentum and energy equations for the pump model based on the nonlinear relationship between speed, flow rate, and pressure; establishing transient flow equations for proportional and directional valves based on valve core position, flow channel function, and throttling equations; using distributed parameter approximation or a finite-segment concentrated volume-damping model for pipelines to represent fluid inertia and compressibility; representing coupled dynamics for the cylinders using intracavity pressure, piston motion equations, and inflow / outflow equations; and providing the evolution relationship of oil temperature with power loss and heat dissipation through energy conservation. Numerical integration employs an implicit time-stepping scheme to handle the rigidity of the system: applying Newton-Raphson iteration to the nonlinear equations, and using a sparse matrix structure to store the linearized Jacobian matrix to improve solution efficiency; and using an event detection mechanism to apply state transitions when valve switching or contact events occur, and performing local arc-length extension at the transition point to ensure convergence. The parameters of nonlinear components are calibrated using least-squares identification and Kalman filtering incremental adjustment. The identification objective is to minimize the mean square error between the simulated pressure / flow time series and the bench measurement time series. Residual autocorrelation analysis is used to verify whether the model error exhibits a white noise distribution. The output is an oil circuit simulation model, which provides system node pressure, pipeline flow, pump operating power, oil temperature time series, and local pressure drop distribution, for coupling with the multibody dynamics model.

[0081] Step S23: Perform multibody dynamics simulation analysis of the loader based on the loader's walking dynamics data and working dynamics data to obtain the loader's multibody dynamics simulation model;

[0082] In this embodiment of the invention, a multibody dynamics simulation model of a loader based on walking dynamics and working dynamics data is constructed and verified. The dynamics model consists of rigid bodies and variable cross-section beam elements, defining the vehicle body, boom, connecting rods, bucket, axle, and wheels as basic mass-rigid body subsystems. A joint constraint matrix is ​​constructed to describe the articulated degrees of freedom and restricted rotations. Ground interaction adopts a segmented contact model and a tire-ground interaction model. Tire mechanics uses elastoplastic contact theory or the Bekker foundation model to characterize the contact reaction force. The bucket-material contact is represented by a distributed contact pressure term and a frictional sliding model to represent the entry and exit processes. The dynamic equations are derived into a system of differential algebraic equations (DAE) using Lagrange or variational methods. Lagrange multipliers are introduced into the constraints to solve for the reaction forces. The time-domain integration uses an implicit difference scheme to ensure numerical stability under high-stiffness contact conditions. Constraint drift is corrected through constraint stabilization techniques (such as Baumgarte or coordinate projection methods) to maintain constraint consistency. Model calibration involves comparing key kinematic quantities (bucket pose timing, boom angular velocity, and vehicle acceleration) with field measurements and updating mass and inertial parameters using least-squares estimation. Model outputs include displacement / velocity / acceleration timing, joint torques, contact force distribution, and dissipated power, providing piston displacement and desired torque boundary conditions for the hydraulic circuit simulation model to support multi-domain coupling.

[0083] Step S24: Based on the hydraulic circuit component data of the loader and the multibody dynamics component data of the loader, perform multi-domain simulation model coupling processing to obtain a multi-domain loader hydraulic circuit-dynamics simulation coupled model;

[0084] In this embodiment of the invention, the multi-domain simulation coupling processing and interface definition of hydraulic circuit component data and multibody dynamics component data are implemented. The coupling adopts a bidirectional strong coupling strategy, with interface variables defined as cylinder cavity pressure and piston displacement / velocity. The coupling algorithm iterates within sub-steps in each global time step to converge the interface residuals. Specifically, at the beginning of the time step, the multibody model provides the predicted piston displacement value to the hydraulic circuit model; the hydraulic circuit model solves for the cavity pressure and feeds it back to the multibody model; subsequently, the multibody model updates the forces and recalculates the piston displacement. Both processes are executed in sub-iterations through Newton iteration or quasi-Newton interface Jacobian estimation and converge when the residual norm is below a preset threshold. To accelerate the coupling process and ensure energy consistency, energy conservation checks and relaxation factor adjustments are used at the interface. When encountering continuous switching events (such as bucket contact with the ground or valve jumps), the event handling mechanism is activated, and the solution across critical points is completed using the arc-length method. To address the difficulty of explicitly calculating the interface Jacobian, an approximation Jacobian or finite difference method is used to estimate the interface sensitivity. This is combined with Aitken acceleration or local fitting of the Jacobian to reduce the number of sub-iterations. The coupled model is deployed in a parallel computing framework using a domain decomposition strategy. Inter-domain communication is achieved through tightly coupled message passing. The interface residuals, energy inputs and outputs, and sub-domain convergence history of each iteration are recorded to serve as the basis for subsequent optimization and reliability analysis.

[0085] Step S25: The lifting mechanics of the multi-domain loader hydraulic circuit-dynamic simulation coupling model is optimized by using the loading lifting capacity optimization mechanical relationship model to obtain the multi-domain collaborative simulation model of the loader working condition.

[0086] In this embodiment of the invention, a lifting mechanics co-optimization process and working condition model generation for a multi-domain coupled model are performed based on a loading lifting capacity optimization mechanics relationship model. First, the lifting capacity optimization objective is injected into the coupled solution process in a constraint-objective function dual form: the objective function includes a weighted combination of minimizing lifting force, maximizing structural strength margin, and minimizing energy consumption; constraints include upper limits for cavity pressure, upper limits for member stress, geometric interference, and stability boundaries. Optimization employs a direct method nested between outer-layer parameter updates and inner-layer coupled simulations: the outer layer updates design variables (such as valve equivalent flow path parameters, pump flow distribution strategies, initial hydraulic cylinder ratios, and control parameters) using sequential quadratic programming or gradient descent. Each design variable test triggers the inner-layer multi-domain coupled simulation to evaluate the objective function value and constraint violations. To improve gradient calculation efficiency, sensitivity analysis and adjoint methods are used to obtain the derivative of the objective with respect to design variables for the dynamic system. The adjoint equation calculates sensitivity during backpropagation and couples with the outer optimizer. For operating conditions exhibiting discontinuous behavior, multiple firing methods or direct discretization of the time domain are employed and used as constraint points in the optimization process to ensure trajectory consistency and constraint satisfaction. After optimization, a multi-domain co-simulation model of the loader's operating conditions is output. This model includes: a calibrated hydraulic-dynamic coupling sub-model, optimal or quasi-optimal control and geometric parameter sets, and performance index curves (lifting capacity, oil temperature rise, energy consumption spectrum) varying with operating conditions. To ensure engineering usability, cross-operating-condition robustness verification is performed, recording the index distribution under a set of representative ground conditions, operating cycles, and ambient temperatures. The verification results are used to form the final multi-domain co-simulation model for archiving and subsequent online calibration reference.

[0087] Furthermore, step S22 includes the following steps:

[0088] The hydraulic circuit components of the loader are analyzed based on the hydraulic parameters of the loader's travel and working to obtain the data of the hydraulic circuit components. The hydraulic circuit structure model of the loader is then established based on the data of the hydraulic circuit components. Based on the hydraulic parameters of the loader's travel and working, the hydraulic circuit structure model of the loader is subjected to dynamic simulation analysis to obtain the simulation model of the loader's hydraulic circuit.

[0089] In this embodiment of the invention, the hydraulic circuit components of the loader are analyzed and a hydraulic circuit structure model is established. First, physical models are created for key components in the traveling and working hydraulic systems, such as pumps, valves, cylinders, pipelines, accumulators, and joints. Each component is mathematically represented to accurately describe its flow rate, pressure, and energy transfer characteristics. For pump components, a nonlinear momentum equation is established based on the pump speed-flow rate-pressure relationship. For valve components, a throttling equation is established based on the valve core displacement and flow channel area function, and a pressure drop constraint is introduced. For cylinder components, a dynamic coupling relationship is established based on the cavity pressure equation and piston motion equation. For pipeline components, fluid inertia and compressibility characteristics are established using distributed parameters or a finite-segment concentrated volume-damping model. For joints and interfaces, volume changes and minor leakage losses are considered. During the analysis, the geometric dimensions, physical parameters, friction and damping coefficients of each component are recorded using a component parameter matrix. Boundary constraints are set for the flow rate, pressure, speed, and temperature range to ensure consistency between the model and engineering reality. Based on the component data, a comprehensive hydraulic circuit structure model is established using a topology connection method. This model clarifies the flow paths and branch nodes of the pump-valve-cylinder-pipeline system, establishes pressure difference and flow constraints between nodes, and achieves a closed-loop physical description of the hydraulic circuit. Component analysis also includes the analysis and scalar matrixing of pressure response characteristics, flow regulation characteristics, and pump-valve linkage effects, providing a complete set of input data structures and boundary conditions for subsequent dynamic simulation. Based on the hydraulic circuit structure model obtained from component analysis, the hydraulic equations are discretized for time-domain dynamic simulation. The pressure in each cylinder cavity varies with the inflow and outflow rates. The piston motion equation is coupled with the flow equation to form a nonlinear differential equation system, and the valve core displacement and pressure response form transient coupling constraints. The pipeline uses a piecewise volume and damping model to establish transient flow equations, and the pump's transient flow is constrained by load pressure and speed. The dynamic solution employs an implicit time integration method to ensure stable convergence of the system's rigidity characteristics and rapid pressure fluctuations during the solution process. During the solution process, an iterative solution mechanism is introduced for the nonlinear parameters of the components. At each time step, pressure, flow rate, and valve core position are solved through Newton-Raphson iteration. The convergence condition of the iteration is limited by the norm of the nodal pressure residuals and the flow balance error. For transient discontinuous events such as valve opening and closing and cylinder contact limits, an event detection mechanism is introduced. When an event occurs, the time step is locally adjusted, and the arc length method or local iteration is used to ensure the continuity of the system state. The dynamic simulation output includes the pressure change of each cylinder chamber over time, the transient flow rate at the pump outlet, the valve throttling pressure drop, the pressure distribution of pipeline nodes, and the transient power consumption curve of the system. It can be used for subsequent multi-domain coupling calculations with the multibody dynamics model to achieve accurate prediction of the hydraulic circuit dynamic response under loader operating conditions.

[0090] Furthermore, step S23 includes the following steps:

[0091] Based on the loader's walking dynamics data and working dynamics data, the loader's drive components are analyzed to obtain the loader's drive component data. A multi-body dynamics structural model of the loader is then established using the loader's drive component data. Based on the loader's walking dynamics data and working dynamics data, a multi-body dynamics simulation analysis of the loader's multi-body dynamics structural model is performed to obtain a loader multi-body dynamics simulation model.

[0092] In this embodiment of the invention, key drive components of the loader's walking and working systems are physically modeled, including the engine, gearbox, drive shaft, wheel hub, tracks or tires, articulation mechanisms, boom and bucket mechanical drive parts, as well as various connecting parts and support components. A rigid or flexible body dynamic model is established for each component, defining its mass, inertia tensor, center of mass position, and geometric boundary constraints. Constraint relationships are established for each hinge and joint, including rotational constraints, translational constraints, and combined motion constraints, considering frictional torque, damping characteristics, and contact stiffness. Based on the loader's walking dynamics data, the wheel-ground or track-ground interaction force, traction force, and slippage effect are calculated and mapped to the drive component constraint equations. Based on the loader's working dynamics data, the load distribution, torque action, and acceleration response of the boom and bucket under different working conditions are analyzed, obtaining the stress and motion response characteristics of each component under working loads. Through the above analysis, a complete set of loader drive component parameters is generated, including component mass distribution, stiffness matrix, damping coefficient, joint degrees of freedom, and boundary constraints, providing an accurate physical basis for multibody dynamics structural modeling. A holistic multibody dynamics structural model is established through component connection topology, forming multibody coupled system equations, including linear and nonlinear constraints, clarifying the degree-of-freedom distribution, force transmission paths, and energy flow, thus achieving a global description of dynamic behavior. Based on the multibody dynamics structural model obtained from the analysis of driving components, kinematic and dynamic equations are established, including the Newton-Euler equations and the Lagrange equations, covering the dynamic responses of rigid and flexible bodies. Walking dynamics analysis includes wheel / track drive wheel speeds, wheel-to-wheel torque distribution, changes in ground friction and contact reaction forces, and the transient effects of acceleration, deceleration, steering, and slippage on the overall vehicle dynamics. Working dynamics analysis includes the torque response, joint loads, and inertial force distribution of the boom and bucket under different lifting angles and load conditions, as well as the mechanical response of the coupling between hydraulic drive and mechanical transmission. The dynamic equations are solved using implicit integration methods, employing a block-based iterative approach to solve the multi-degree-of-freedom coupled equations, ensuring numerical stability under the system's rigidity. During the solution process, factors such as joint friction, damping, nonlinear stiffness, and transient responses caused by changes in contact conditions are considered. The acceleration, velocity, position, and torque of the system are tracked and recorded in the time domain. The final output is a multibody dynamics simulation result, including the overall machine motion trajectory, the time-varying angles of the boom and bucket, joint reaction forces, load distribution on the drive wheels or tracks, and force curves of structural components. This forms a complete multibody dynamics simulation model of the loader, which is used for multi-domain coupling with the hydraulic circuit simulation model to achieve accurate prediction of the loader's load history and operational condition analysis.

[0093] Furthermore, step S24 includes the following steps:

[0094] Based on the hydraulic circuit component data and multibody dynamics component data of the loader, the correlation characteristics of the hydraulic circuit and multibody dynamics components of the loader are analyzed to generate the correlation characteristic data of the hydraulic circuit and dynamics components of the loader. Then, the simulation model coupling processing of the hydraulic circuit simulation model and the multibody dynamics simulation model of the loader is performed through the correlation characteristic data of the hydraulic circuit and dynamics components of the loader to obtain the hydraulic circuit and dynamics simulation coupling model of the loader.

[0095] In this embodiment of the invention, structural feature comparison and analysis are performed on hydraulic circuit component data and multibody dynamics component data, including the dynamic and kinematic correspondences between hydraulic cylinders, pumps, valves, pipelines, and energy storage elements and boom, bucket, chassis, steering system, wheels / tracks, and joint components. By analyzing the mapping relationship between the stroke, cylinder diameter, and hydraulic pressure changes of the hydraulic cylinders and the angular velocity, angular acceleration, displacement, and stress state of the multibody components, hydraulic circuit-dynamics component correlation feature data is generated. This feature data includes the contribution of hydraulic output pressure to boom lifting torque, the influence of hydraulic flow on drive wheel speed response, the effect of energy transfer within the hydraulic system on the inertial response of mechanical components, and the dynamic coupling effects caused by various hydraulic hysteresis and pipeline elasticity. Subsequently, the hydraulic circuit-dynamics component correlation feature data is used as a constraint condition to couple the loader hydraulic circuit simulation model and the multibody dynamics simulation model. During the coupling process, a real-time mapping is established between the hydraulic cylinder's mechanical input and the force output of the multi-body joints. Changes in hydraulic pressure, flow rate, and temperature are used as driving signals to act on the multi-body dynamics model. Simultaneously, joint displacement, angular velocity, acceleration, and inertial force are fed back to the hydraulic circuit simulation model to correct hydraulic load and flow distribution. The coupling process also considers the nonlinear response between the hydraulic and mechanical systems, including real-time feedback of hydraulic cylinder filling hysteresis, pipeline flexible vibration, joint friction, and damping effects. By iteratively solving the hydraulic dynamics equations and the multi-body dynamics equations, the hydraulic circuit pressure fluctuations and the force changes of the mechanical components are synchronized, achieving multi-domain collaborative simulation of the entire machine. The final generated loader hydraulic circuit-dynamics simulation coupling model contains a complete relationship between hydraulic drive and mechanical response, accurately simulating the load history and dynamic behavior of the loader under different working conditions, providing fundamental data support for subsequent load history prediction and working condition optimization.

[0096] Furthermore, as an embodiment of the present invention, reference is made to... Figure 3 As shown, Figure 1 A detailed flowchart of step S3 is shown below. In this embodiment, step S3 includes the following steps:

[0097] Step S31: Obtain multi-source load data of the loader;

[0098] In this embodiment of the invention, multi-source load data is acquired for the loader under different working environments and tasks, such as unloaded straight-line driving, unloaded single-sided turning, unloaded full-circumference turning in place, rated load straight-line driving, rated load turning driving, rated load full-circumference turning in place, loading and unloading stones, and leveling material piles. The collected load data includes boom hydraulic cylinder thrust, bucket hinge point force, bucket torque, wheel ground reaction force, overall machine inertial load, and hydraulic system pressure, flow rate, and temperature data. High-precision strain gauge sensors and force sensors deployed at the hydraulic cylinder, boom hinge point, and bucket connection points record the force and load changes at each joint in real time; rotary encoders and accelerometers acquire boom and bucket angular velocities, accelerations, and displacement information; and hydraulic pressure sensors and flow meters acquire hydraulic cylinder pressure fluctuations and flow rate changes. During the data collection process, different operating conditions are divided into loading, lifting, unloading and walking conditions to ensure that the data covers the entire operating cycle. A standardized set of operating condition load data is generated according to the time series to provide accurate load input conditions for multi-domain simulation and ensure that the subsequent simulation model can reflect the coordinated response characteristics of the hydraulic system and the dynamic system.

[0099] Step S32: Transmit the multi-source load data of the loader's working condition to the multi-domain collaborative simulation model of the loader's working condition to perform load history characteristic analysis of the multi-domain collaborative simulation of the loader, so as to obtain the load history characteristic data of the multi-domain collaborative simulation of the loader.

[0100] In this embodiment of the invention, the acquired multi-source load data is used in a multi-domain collaborative simulation model. The model consists of a hydraulic circuit dynamic simulation module, a multibody dynamics module, and a boom and bucket mechanical response module. By analyzing the dynamic coupling relationship between hydraulic cylinder thrust and hydraulic circuit pressure fluctuations and flow distribution, and combining the multibody dynamic response of boom angle, velocity, and acceleration with bucket torque and joint forces, a time-series simulation of the entire machine's load history is achieved. During the simulation, the load data is periodically segmented, and key features such as load peak value, average value, amplitude, load rise rate, fall rate, duration, and vibration frequency are extracted. This forms multi-domain collaborative simulation load history characteristic data for the loader, which can simultaneously reflect the nonlinear coupling characteristics between the hydraulic system input signal and the mechanical system output mechanical response, providing a multi-domain collaborative time-series data foundation for load history prediction and energy consumption analysis.

[0101] Step S33: Perform load history change gradient feature analysis on the load history feature data of the multi-domain collaborative simulation of the loader, generate the load history change gradient feature data of the collaborative simulation, and establish the load history prediction mapping relationship of the multi-domain collaboration of the loader through the load history change gradient feature data of the collaborative simulation to obtain the load history prediction model of the multi-domain collaboration of the loader.

[0102] In this embodiment of the invention, load change gradient analysis is performed on the multi-domain collaborative simulation load history characteristic data. This includes calculating the rate of change, upward gradient, downward gradient, and abrupt change rate of hydraulic cylinder thrust, boom force, and bucket torque over different time periods, forming collaborative simulation load history change gradient characteristic data. Based on the load gradient characteristics, the coupling relationship between working condition changes, load changes, hydraulic system response, and mechanical structure forces is analyzed. A load history prediction mapping relationship is established, nonlinearly mapping the working condition input conditions, hydraulic pressure and flow response, boom and bucket forces, and load change gradient to achieve a predictive mapping from input working conditions to load change trends. This mapping relationship is used to generate a multi-domain collaborative load history prediction model for the loader, enabling the prediction of the time-series load changes of the entire machine under different working conditions, including peak, average, and transient load change characteristics, providing a foundation for subsequent working condition optimization and energy consumption analysis.

[0103] Step S34: Based on the multi-domain collaborative simulation load history characteristic data of the loader, perform energy consumption loss impact characteristic analysis on the simulation load history, and generate energy consumption loss impact characteristic data on the simulation load history.

[0104] In this embodiment of the invention, multi-domain collaborative simulation load history characteristic data is used to analyze the energy consumption and losses of various hydraulic and mechanical systems of a loader. By establishing a nonlinear mapping relationship between hydraulic cylinder power consumption, hydraulic pump energy output, pipeline friction loss, boom joint friction, and bucket torque variation, instantaneous power consumption and energy distribution under different working conditions are calculated, generating characteristic data on the impact of simulated load history energy consumption and losses. The peak energy consumption and energy distribution patterns of the hydraulic system under high load, long-cycle operation, and rapid lifting conditions are analyzed to form a load history-energy consumption nonlinear matrix. This characteristic data not only reflects the coupled energy consumption characteristics of the hydraulic and mechanical systems but also provides a key basis for subsequent temperature anomaly detection, energy consumption optimization, and load history prediction mapping relationship accuracy optimization, realizing multi-domain collaborative energy management analysis.

[0105] Step S35: Extract and process the oil circuit temperature anomaly of the simulated load history based on the load history-energy consumption characteristic data of the loader collaborative simulation to obtain the simulated load history oil circuit temperature anomaly data. Analyze the simulation drive adjustment coefficient of the load history temperature protection feedback based on the preset temperature self-feedback protection engine to generate the temperature protection feedback simulation drive adjustment coefficient.

[0106] In this embodiment of the invention, load history-energy consumption characteristic data from a loader's collaborative simulation are used to identify and extract abnormal hydraulic circuit temperatures. Based on the hydraulic pump output power, hydraulic cylinder flow rate, hydraulic line pressure, and force data of the boom and bucket joints, the accumulated heat energy and instantaneous temperature changes of each hydraulic circuit are calculated, constructing a time-series curve of the hydraulic circuit temperature. Subsequently, a temperature self-feedback protection engine is applied. By performing gradient change analysis, peak over-limit detection, and accumulated energy threshold judgment on the temperature curve, abnormal hydraulic circuit temperatures are identified. For example, under rated load full-circuit rotation, excessively high hydraulic circuit temperatures caused by excessively fast rotation speed generate simulated load history hydraulic circuit temperature anomaly data. Based on this, the impact of temperature anomalies on hydraulic oil viscosity changes, flow resistance, and power consumption is considered to perform simulation drive adjustment coefficient analysis for load history temperature protection feedback, such as the temperature self-feedback protection function. By collecting the hydraulic oil temperature of the working hydraulic system and transmitting this oil temperature signal to the engine control unit (ECU), when the oil temperature exceeds a set value, the fuel supply to the fuel system is reduced, causing a decrease in engine output torque while maintaining the engine speed, achieving rapid heat dissipation and protecting critical hydraulic components, thus improving equipment reliability. The adjustment coefficient calculation is constrained by the temperature control requirements of the hydraulic system. It quantifies the adjustment range of the hydraulic pump output pressure, flow distribution and dynamic response caused by temperature anomalies, thereby generating the temperature protection feedback simulation-driven adjustment coefficient, providing accurate control parameters for subsequent energy consumption characteristic adjustment.

[0107] Step S36: Based on the temperature protection feedback simulation drive adjustment coefficient, adjust the energy consumption and loss influence characteristics of the simulated load history energy consumption and loss feature data to generate adjusted simulated load history energy consumption and loss influence feature data.

[0108] In this embodiment of the invention, the energy consumption loss characteristic data of the simulated load history is adjusted based on the temperature protection feedback simulation-driven adjustment coefficient. A coupling mapping is established between the temperature protection feedback simulation-driven adjustment coefficient and energy consumption characteristics such as hydraulic pump power, hydraulic cylinder energy output, pipeline friction loss, and boom joint friction energy consumption. By adjusting the energy distribution curve of the hydraulic system under different working conditions, the energy consumption ratio of each joint and oil circuit is recalculated, forming an energy consumption loss characteristic matrix after temperature protection adjustment. Combining the impact of bucket operation, boom lifting, and travel conditions on the thermal load of the hydraulic system, the adjusted energy consumption curve is time-series optimized to ensure that the hydraulic system does not experience abnormal power loss due to temperature anomalies during high-load operations, while maintaining overall system energy balance. Finally, the adjusted simulated load history energy consumption loss characteristic data is generated, providing quantifiable input conditions for subsequent accuracy optimization of the load history prediction mapping relationship.

[0109] Step S37: Optimize the accuracy of the load history prediction mapping relationship of energy consumption loss in the multi-domain collaborative load history prediction model of the loader by using the characteristic data of energy consumption loss influence of the adjusted simulated load history, so as to obtain an optimized multi-domain collaborative load history prediction model of the loader.

[0110] In this embodiment of the invention, the energy consumption loss influence characteristic data of the simulated load history are used to optimize the energy consumption loss accuracy of the multi-domain collaborative load history prediction model of the loader. The adjusted energy consumption loss characteristic data is coupled with the hydraulic response characteristics and mechanical structure force response characteristics in the original multi-domain collaborative load history prediction model for analysis. The nonlinear mapping relationship of the prediction model is adjusted, and the predicted load history values ​​of boom mechanical response, bucket torque, and hydraulic flow are recalculated. By analyzing the load peak value, gradient change, and energy consumption distribution deviation under different working conditions, the prediction model parameters are gradually corrected to achieve minimum error matching between the load history prediction output and the actual multi-domain collaborative response. An optimized multi-domain collaborative load history prediction model for the loader is generated. This model can accurately predict the load history of the loader's hydraulic and dynamic systems under high load, rapid lifting, and complex working conditions, providing a reliable basis for real-time working condition control, energy consumption management, and temperature protection, and realizing closed-loop management of multi-domain collaborative load prediction and energy consumption optimization.

[0111] Furthermore, step S34 includes the following steps:

[0112] Step S341: Based on the load history characteristic data of the multi-domain co-simulation of the loader, perform nonlinear mapping relationship analysis between the load history and energy consumption characteristics of the loader co-simulation, and generate load history-energy consumption characteristic data of the loader co-simulation.

[0113] In this embodiment of the invention, load history characteristic data from multi-domain collaborative simulation of a loader is used to perform a nonlinear mapping analysis of the load history and energy consumption characteristics of the loader. Based on multi-domain collaborative characteristics such as boom lifting, bucket digging, travel speed, and load changes, the load history characteristics are decomposed into high, medium, and low load stages according to the time series. The hydraulic pump power consumption, hydraulic cylinder output energy, and oil circuit heat loss at each stage are calculated using energy consumption curves. Subsequently, the load history characteristics and various energy consumption indicators are nonlinearly coupled and analyzed. Using multivariate mapping logic, a correlation model is established between hydraulic energy consumption and dynamic forces, movement speed, and operating angles, forming load history-energy consumption characteristic data from the loader's collaborative simulation. This analysis ensures that the dynamic response relationship between load fluctuations and energy consumption is fully reflected under different working conditions, providing an accurate foundation for subsequent energy loss characteristic extraction.

[0114] Step S342: Extract the first simulation load history-energy consumption characteristic data and the second simulation load history-energy consumption characteristic data based on the loader collaborative simulation load history-energy consumption characteristic data;

[0115] In this embodiment of the invention, phased feature data extraction is performed based on the load history-energy consumption characteristic data of the loader's collaborative simulation. The entire load history is segmented according to time series and working condition stages, and indicators such as hydraulic energy consumption, joint torque, boom force, and bucket output force under high load conditions are extracted to generate the first simulated load history-energy consumption characteristic data. The hydraulic system energy consumption, travel system dynamic load, and bucket inertial force influence under medium and low load and no-load operation stages are extracted to generate the second simulated load history-energy consumption characteristic data. Through phased extraction, the energy consumption distribution of the load history under different operating intensities is described in detail, providing a phased and structured data foundation for subsequent energy loss impact analysis. This ensures that the differences in energy consumption characteristics under different working conditions can be clearly quantified, and the total work generated by the energy consumption or time series history of the loader's collaborative simulation load history-energy consumption characteristic data is greater than that generated by the energy consumption or time series history of the second simulated load history-energy consumption characteristic data, meaning that the two are loader load history sample data with significant evaluation differences.

[0116] Step S343: Analyze the energy consumption loss impact characteristics of the simulated load history using the first simulated load history-energy consumption characteristic data and the second simulated load history-energy consumption characteristic data, and generate simulated load history energy consumption loss impact characteristic data.

[0117] In this embodiment of the invention, the first and second simulated load history-energy consumption characteristic data are aligned according to working condition categories to ensure a one-to-one correspondence between load values, boom forces, bucket loads, and hydraulic energy consumption information at each time point. Based on the aligned data, the impact of load changes on hydraulic system pressure, pump output power, and travel system drive energy consumption under different working conditions is calculated, forming a load-energy consumption correlation matrix. Subsequently, by statistically analyzing the energy consumption peak, energy consumption accumulation, and energy consumption fluctuation rate in the matrix, the sensitivity characteristics of the load history to energy consumption are extracted, and a load energy consumption change gradient map is generated to characterize the instantaneous and overall impact of load changes on energy loss. Furthermore, a nonlinear mapping analysis method is used to correlate the load history with energy loss characteristics, establishing a quantitative mapping relationship between load changes and energy loss, thereby achieving an accurate assessment of the energy consumption impact under different working conditions. Finally, the above analysis results were compiled into simulated load history energy consumption loss impact characteristic data, including key indicators such as load peak, energy consumption peak, energy consumption gradient, and cumulative energy consumption. This provides a complete data foundation for subsequent temperature protection adjustment and load history prediction accuracy optimization. The generated simulated load history energy consumption loss impact characteristic data is used to conduct refined analysis of load changes in the load history while considering energy loss. During load history prediction, the energy consumption loss characteristic data can reflect the impact of hydraulic system pressure fluctuations, drive system power changes, and load peak on overall energy loss, thereby correcting the load response under high load conditions or extreme conditions and improving the reliability and accuracy of prediction results. At the same time, this characteristic data can provide a basis for dynamic adjustment of the prediction model. For example, under sudden load increases or long-term high load conditions, by introducing the energy consumption loss impact characteristics, the smoothness and safety of the load history curve can be optimized, avoiding the risk of over-predicting or underestimating the load peak.

[0118] Furthermore, in step S342, the energy consumption value of the first simulated load history-energy consumption characteristic data is greater than the energy consumption value of the second simulated load history-energy consumption characteristic data, and / or, the time series interval of the first simulated load history-energy consumption characteristic data is greater than the time series interval of the second simulated load history-energy consumption characteristic data.

[0119] Furthermore, step S4 includes the following steps:

[0120] Step S41: Obtain the orientation signal parameters of the electric control handle;

[0121] In this embodiment of the invention, the sensor system of the loader's electronic control handle acquires azimuth signal parameters, including the spatial position, deflection angle, and movement rate information of the boom lifting handle, bucket control handle, and travel control handle. A multi-dimensional sensor module measures the handle's angular offset, speed change, and acceleration response, converting these into digital signals to form a continuous time-series azimuth parameter. This signal data accurately records the instantaneous state of each handle's operation in time, ensuring that the loader's control behavior under different working conditions can be completely captured, while also providing working condition triggering conditions for load prediction. During the measurement process, the electronic control handle azimuth signal parameters are filtered to eliminate signal jitter and external interference, and a time synchronization mechanism unifies the azimuth signals of different handles onto the same time axis, thereby achieving accurate mapping of multi-source handle actions and providing complete input for subsequent working condition load monitoring.

[0122] Step S42: Perform load monitoring and processing on the loader operating condition based on the orientation signal parameters of the electric control handle to obtain loader operating condition load data. The loader operating condition load data includes load data under normal conditions and load data under extreme conditions. The load data under extreme conditions includes rated load data under straight driving conditions, rated load data under stationary full-circuit turning conditions, and loading and unloading conditions.

[0123] In this embodiment of the invention, the loader's operating load is monitored and classified in real time using the handle orientation signal parameters. The handle orientation signal is associated with the mechanical models of the loader's lifting mechanism, bucket load, and travel system to calculate the instantaneous load values ​​of the boom force, bucket output force, and travel system driving force under different operating conditions. The monitoring process is divided into normal operating conditions and extreme operating conditions. The extreme operating conditions include the rated load under straight-line driving conditions, the rated load under stationary full-circuit turning conditions, and the instantaneous maximum load under bucket loading and unloading conditions. Through mechanical mapping analysis, the instantaneous force conditions are mapped to the handle operation signals to generate an accurate operating load data sequence. During the processing of the operating load data, a real-time feature extraction method is used to identify the load fluctuation amplitude, peak time, and load duration, providing accurate load input conditions for subsequent prediction models.

[0124] Step S43: Transmit the loader's working condition load data to the optimized loader multi-domain collaborative load history prediction model for loader load history prediction processing, including: when extreme condition load data of the loader is identified in the loader's working condition load data, the extreme condition load history accuracy of the loader's extreme condition load data is optimized by adjusting the energy consumption and loss influence feature data of the simulated load history built into the optimized loader multi-domain collaborative load history prediction model, and the extreme condition load history prediction data of the loader is output; when extreme condition load data of the loader is identified in the loader's working condition load data, the normal condition load data of the loader is output, and the normal condition load history prediction data of the loader is output; the extreme condition load history prediction data of the loader and the normal condition load history prediction data of the loader are integrated to obtain the loader load history prediction data.

[0125] In this embodiment of the invention, in optimizing the multi-domain collaborative load history prediction model for loaders, the load data of the loaders under various operating conditions is first classified and identified, dividing the data into extreme condition operating conditions and normal condition operating conditions. For extreme condition load data, such as high load during straight-line driving, full-circuit turning in place, and full-load bucket operation, the load fluctuation is refined by adjusting the energy consumption loss influence feature data of the simulated load history built into the load history prediction model. This includes dynamically correcting the peak pressure of the hydraulic system, the power fluctuation of the drive system, and the nonlinear response of the bucket's motion inertial force to eliminate the impact of abnormal energy consumption or sudden load increases on prediction accuracy, as well as adjusting the engine parameters of the load history of the loader by adjusting the hydraulic circuit temperature overload, thereby generating extreme condition load history prediction data with optimized accuracy. For normal condition load data, such as light-load driving and low-frequency shoveling operation, the load change trend under the operating conditions is smoothly fitted by the prediction model, and the dynamic response deviation under low load is corrected by combining the energy consumption loss feature data to generate normal condition load history prediction data. Finally, through the system's internal fusion mechanism, the extreme condition load history prediction data and the conventional condition load history prediction data are aligned in time and working condition to form complete loader load history prediction data. This provides a reliable prediction basis for subsequent intelligent auxiliary control, enabling high-precision and dynamic response prediction of loader load changes under all working conditions.

[0126] Step S44: Execute intelligent auxiliary control of loader load history using loader load history prediction data.

[0127] In this embodiment of the invention, load history prediction data of the loader is used to intelligently assist and control the operation of the loader. Specific operations include: generating load control commands based on the predicted load history, combined with the boom force, bucket output, and the driving state of the travel system, to keep the hydraulic system pressure, drive torque, and control signals within a safe operating range under extreme and normal conditions. During loading and unloading operations, the bucket angle and boom speed are adjusted in advance using the predicted data to achieve load balance and minimize energy consumption. During travel and steering, the hydraulic pump output and drive power are adjusted according to the predicted load to achieve preventative control of load peaks. During the control process, a continuous feedback mechanism dynamically compares the predicted load with the actual load, generating fine-tuning commands in real time to achieve refined control of the load history, ensuring operational safety and energy efficiency.

[0128] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0129] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A load history prediction method for loaders based on simulation data, characterized in that, Includes the following steps: Step S1: Obtain the structural parameters of the skid steer loader; establish a mechanical mapping relationship for optimizing the loading and lifting capacity based on the structural parameters of the skid steer loader to obtain a mechanical relationship model for optimizing the loading and lifting capacity; Step S1 includes the following steps: Step S11: Obtain the structural parameters of the skid steer loader; Step S12: Perform multi-source structural type abstraction processing on the skid steer loader structural parameters to obtain multi-source skid steer loader structural abstract data; Step S13: Based on the abstract data of the multi-source skid steer loader structure, perform data analysis and processing on the loader lifting characteristics and bucket trajectory to obtain the loader lifting characteristic data and bucket trajectory data, and extract the boom parameters and corresponding hinge point parameters of the skid steer loader structure from the abstract data of the multi-source skid steer loader structure. Step S14: Perform topological analysis of the boom force characteristics based on the boom rod parameters and corresponding hinge point parameters of the skid steer loader structure to obtain the boom force characteristics topological data. Based on the boom force characteristics topological data, perform boom mechanical relationship modeling for loader lifting to obtain the loader lifting boom mechanical relationship model. Step S15: Based on the loader lifting characteristic data and the loader lifting boom mechanical relationship model, perform a minimum lifting force characteristic analysis of the boom with maximum lifting capacity to obtain lifting capacity-boom lifting force characteristic data, and design a loading lifting optimization objective function based on the lifting capacity-boom lifting force characteristic data. Step S16: Map the bucket trajectory data as loading and lifting constraints and the loading and lifting optimization objective function as loading and lifting optimization objective conditions to the loader lifting boom mechanical relationship model to perform mechanical mapping relationship analysis for loading and lifting capacity optimization, so as to obtain the loading and lifting capacity optimization mechanical relationship model. Step S2: Establish a multi-domain loader hydraulic circuit-dynamics simulation coupled model; perform loader lifting mechanics co-optimization processing on the multi-domain loader hydraulic circuit-dynamics simulation coupled model through the loading lifting capacity optimization mechanical relationship model to obtain a multi-domain co-simulation model of loader working conditions; Step S2 includes the following steps: Step S21: Obtain the loader's walking hydraulic parameters and corresponding loader walking dynamics data, as well as the loader's working hydraulic parameters and corresponding loader working dynamics data; Step S22: Perform dynamic simulation analysis of the hydraulic circuit based on the loader's walking hydraulic parameters and working hydraulic parameters to obtain the loader's hydraulic circuit simulation model; Step S23: Perform multibody dynamics simulation analysis of the loader based on the loader's walking dynamics data and working dynamics data to obtain the loader's multibody dynamics simulation model; Step S24: Based on the loader hydraulic circuit simulation model and the loader multibody dynamics simulation model, perform multi-domain simulation model coupling processing to obtain a multi-domain loader hydraulic circuit-dynamics simulation coupled model; Step S25: The lifting mechanics of the multi-domain loader hydraulic circuit-dynamic simulation coupling model is optimized by using the loading lifting capacity optimization mechanical relationship model to obtain the multi-domain collaborative simulation model of the loader working condition; Step S3: Obtain multi-source load data of the loader; transmit the multi-source load data of the loader to the multi-domain collaborative simulation model of the loader to perform load history feature analysis of the multi-domain collaborative simulation of the loader, so as to obtain load history feature data of the multi-domain collaborative simulation of the loader; establish the accuracy optimization load history prediction mapping relationship of the multi-domain collaborative simulation of the loader based on the load history feature data of the multi-domain collaborative simulation of the loader, and generate an optimized load history prediction model of the multi-domain collaborative simulation of the loader. Step S3 includes the following steps: Step S31: Obtain multi-source load data of the loader; Step S32: Transmit the load data of the multi-source working condition of the loader to the multi-domain collaborative simulation model of the loader working condition to perform load history characteristic analysis of the multi-domain collaborative simulation of the loader, so as to obtain the load history characteristic data of the multi-domain collaborative simulation of the loader. Step S33: Perform load history change gradient feature analysis on the load history feature data of the multi-domain collaborative simulation of the loader, generate the load history change gradient feature data of the collaborative simulation, and establish the load history prediction mapping relationship of the multi-domain collaboration of the loader through the load history change gradient feature data of the collaborative simulation to obtain the load history prediction model of the multi-domain collaboration of the loader. Step S34: Based on the multi-domain collaborative simulation load history characteristic data of the loader, perform energy consumption loss impact characteristic analysis on the simulation load history, and generate energy consumption loss impact characteristic data on the simulation load history. Step S35: Extract and process the oil circuit temperature anomaly of the simulated load history based on the energy consumption loss influence characteristic data of the simulated load history to obtain the simulated load history oil circuit temperature anomaly data. Analyze the simulation drive adjustment coefficient of the load history temperature protection feedback based on the preset temperature self-feedback protection engine to generate the temperature protection feedback simulation drive adjustment coefficient. Step S36: Based on the temperature protection feedback simulation drive adjustment coefficient, adjust the energy consumption and loss influence characteristics of the simulated load history energy consumption and loss feature data to generate adjusted simulated load history energy consumption and loss influence feature data. Step S37: Optimize the accuracy of the load history prediction mapping relationship of energy consumption loss in the multi-domain collaborative load history prediction model of the loader by using the characteristic data of energy consumption loss influence of the adjusted simulated load history, so as to obtain the optimized multi-domain collaborative load history prediction model of the loader. Step S4: Obtain the orientation signal parameters of the electric control handle; perform load monitoring and processing on the loader operating condition based on the orientation signal parameters of the electric control handle to obtain the loader operating condition load data; transmit the loader operating condition load data to the optimized loader multi-domain collaborative load history prediction model for load history prediction processing to obtain the loader load history prediction data; execute the intelligent auxiliary control operation of the loader load history based on the loader load history prediction data. Step S4 includes the following steps: Step S41: Obtain the orientation signal parameters of the electric control handle; Step S42: Perform load monitoring and processing on the loader based on the orientation signal parameters of the electric control handle to obtain loader load data. The loader load data includes load data under normal conditions and load data under extreme conditions. The load data under extreme conditions includes rated load data for straight-line driving, rated load data for stationary full-circuit turning, and rated load data for loading and unloading. Step S43: Transmit the loader's working condition load data to the optimized loader multi-domain collaborative load history prediction model for loader load history prediction processing, including: when extreme condition load data of the loader is identified in the loader's working condition load data, the extreme condition load history accuracy of the loader's extreme condition load data is optimized by adjusting the energy consumption loss influence feature data of the simulated load history built into the optimized loader multi-domain collaborative load history prediction model, and the extreme condition load history prediction data of the loader is output; when normal condition load data of the loader is identified in the loader's working condition load data, the normal condition load history prediction data of the loader is output; and the extreme condition load history prediction data and the normal condition load history prediction data of the loader are integrated to obtain the loader load history prediction data. Step S44: Execute intelligent auxiliary control of loader load history using loader load history prediction data.

2. The loader load history prediction method based on simulation data as described in claim 1, characterized in that, Step S22 includes the following steps: The hydraulic circuit components of the loader are analyzed based on the hydraulic parameters of the loader's travel and working to obtain the data of the hydraulic circuit components. The hydraulic circuit structure model of the loader is then established based on the data of the hydraulic circuit components. Based on the hydraulic parameters of the loader's travel and working, the hydraulic circuit structure model of the loader is subjected to dynamic simulation analysis to obtain the simulation model of the loader's hydraulic circuit.

3. The loader load history prediction method based on simulation data as described in claim 2, characterized in that, Step S23 includes the following steps: Based on the loader's walking dynamics data and working dynamics data, a multi-body dynamics component analysis of the loader is conducted to obtain the multi-body dynamics component data of the loader. A multi-body dynamics structural model of the loader is then established based on the multi-body dynamics component data of the loader. Based on the loader's walking dynamics data and working dynamics data, a multi-body dynamics simulation analysis of the loader's multi-body dynamics structural model is conducted to obtain a multi-body dynamics simulation model of the loader.

4. The loader load history prediction method based on simulation data as described in claim 3, characterized in that, Step S24 includes the following steps: Based on the hydraulic circuit component data and multibody dynamics component data of the loader, the correlation characteristics of the hydraulic circuit and multibody dynamics component of the loader are analyzed to generate the correlation characteristic data of the hydraulic circuit and dynamics component of the loader. Then, the simulation model coupling processing of the hydraulic circuit simulation model and the multibody dynamics simulation model of the loader is performed through the correlation characteristic data of the hydraulic circuit and dynamics component of the loader to obtain the multi-domain hydraulic circuit-dynamics simulation coupling model of the loader.

5. The loader load history prediction method based on simulation data as described in claim 1, characterized in that, Step S34 includes the following steps: Step S341: Based on the load history characteristic data of the multi-domain co-simulation of the loader, perform nonlinear mapping relationship analysis between the load history and energy consumption characteristics of the loader co-simulation, and generate load history-energy consumption characteristic data of the loader co-simulation. Step S342: Extract the first simulation load history-energy consumption characteristic data and the second simulation load history-energy consumption characteristic data based on the loader collaborative simulation load history-energy consumption characteristic data; Step S343: Analyze the energy consumption loss impact characteristics of the simulated load history using the first simulated load history-energy consumption characteristic data and the second simulated load history-energy consumption characteristic data, and generate simulated load history energy consumption loss impact characteristic data.

6. The loader load history prediction method based on simulation data as described in claim 5, characterized in that, Step S342: The energy consumption value of the first simulated load history-energy consumption characteristic data is greater than the energy consumption value of the second simulated load history-energy consumption characteristic data, and / or, the time series interval of the first simulated load history-energy consumption characteristic data is greater than the time series interval of the second simulated load history-energy consumption characteristic data.

Images