Forklift structure health state assessment method and system

Abstract

The invention belongs to the technical field of multi-variable measurement, and particularly relates to a forklift structure health state assessment method and system. The method comprises the steps that a plurality of strain type force sensors are symmetrically installed on the root portion of a left pallet fork and the root portion of a right pallet fork of the forklift, and the strain type force sensors are used for collecting stress signals of the left pallet fork and the right pallet fork in the three-dimensional space in real time; the six-degree-of-freedom motion state data of the forklift are collected in real time through an inertia measurement unit installed on a forklift main body, and the six-degree-of-freedom motion state data comprise linear accelerations in three orthogonal axial directions and angular velocities around the three orthogonal axial directions. According to the system, a high-bandwidth strain sensing array is deployed at the root of a pallet fork, the six-degree-of-freedom motion state of the whole vehicle is synchronously obtained, a dynamic disturbing force online decoupling mechanism based on a rigid-flexible coupling dynamic model is constructed, and the dependence of a traditional weighing technology on a static working condition is fundamentally broken through.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of multiple variable measurement, and particularly relates to a forklift structure health state evaluation method and system. BACKGROUND

[0002] With the continuous improvement of modern logistics and warehousing automation, the running safety and structural reliability of forklifts, as core material handling equipment, are increasingly valued. The forklift frequently bears dynamic load during operation, especially under high-speed driving, sudden stopping or turning conditions, and the force distribution on the forks is prone to instantaneous deviation. Traditional structure health monitoring methods mainly rely on static weighing sensors to measure the load when the vehicle is completely stationary, which cannot capture real-time force changes during the handling process. Such methods ignore the multi-dimensional mechanical response of the forklift under complex motion states, making it difficult to accurately assess whether the forks have risks such as unbalanced load, twisting or local stress concentration, thus failing to effectively warn potential structural instability or fatigue damage.

[0003] Dynamic evaluation of the health state of the fork structure has become a key technology direction to ensure the safe operation of the forklift. This direction aims to obtain multi-point force information of the forks during the motion process through a perception system in real time, and to make online judgments on the balance and rationality of the force distribution combined with an intelligent analysis model. The core goal is to realize continuous monitoring and early abnormal identification of the fork structure state without interrupting normal operation, providing data support for preventing structural failure.

[0004] Although some high-end forklifts have integrated basic weighing functions, their sensing systems generally lack adaptability to dynamic conditions, and no mapping relationship between force distribution and structural health has been established. On the one hand, a single static sensor cannot analyze complex loads under multiple degrees of freedom; on the other hand, even if multiple sensors are deployed, if there is a lack of lightweight, low-latency edge intelligent processing mechanism, massive dynamic data is difficult to convert into effective structure state criteria.

[0005] In typical scenarios such as high-frequency handling, turning in a narrow space, or driving on uneven ground, existing solutions cannot timely identify the risk of structural imbalance caused by unbalanced load due to response lag, insufficient accuracy or excessive computational overhead, leading to long-term potential safety hazards. SUMMARY

[0006] The application provides a forklift structure health state evaluation method and system, aiming to solve the technical problem that the structural imbalance risk caused by unbalanced load of the forklift forks during dynamic handling cannot be identified and warned in real time. In the prior art, the weighing sensor equipped on the forklift can only measure the weight in the completely static state of the vehicle, and its working principle relies on the static mechanical balance condition. When the forklift is in dynamic working conditions such as driving, steering, acceleration and deceleration, or road bumping, the sensor output signal is seriously polluted by inertial force, centrifugal force and vibration interference, and cannot accurately reflect the actual load distribution of the forks, thereby losing the ability to distinguish the uneven force on the left and right forks. This technical limitation makes the operator unable to know whether the forks are in a dangerous unbalanced load state, and long-term operation may easily cause safety accidents such as mast deformation, fork fracture or vehicle overturning.

[0007] To overcome the above defects, the application provides a forklift structure health state evaluation method and system based on multi-source dynamic perception and adaptive mechanical decoupling. This method discards the traditional weighing logic which relies on static balance assumption, and instead builds a closed-loop evaluation system integrating high-frequency mechanical sensing, motion state recognition and real-time dynamics modeling. The system symmetrically arranges a high-bandwidth strain force sensor array at the roots of the left and right forks, synchronously collects the six-degree-of-freedom motion state data of the forklift, uses the established forklift rigid-flexible coupling dynamics model to decouple the dynamic disturbance force online, thereby accurately separating the pure static component caused by cargo unbalanced load, realizing millisecond-level quantitative evaluation of the force difference on the left and right forks, and generating a structure health index and a safety warning signal on this basis.

[0008] The application provides a forklift structure health state evaluation method, which comprises: symmetrically installing a plurality of strain force sensors at the roots of the left and right forks of the forklift, the strain force sensors being used to collect force signals of the left and right forks in three-dimensional space in real time; collecting six-degree-of-freedom motion state data of the forklift in real time through an inertial measurement unit installed on the main body of the forklift, the six-degree-of-freedom motion state data including linear acceleration along three orthogonal axes and angular velocity around three orthogonal axes; based on a preset forklift rigid-flexible coupling dynamics model, taking the six-degree-of-freedom motion state data as input, calculating the inertial disturbance force component acting on the fork root under the current dynamic working condition; subtracting the corresponding inertial disturbance force component from the force signals of the left and right forks respectively to obtain the left fork net load component and the right fork net load component; calculating the fork unbalanced load rate according to the left fork net load component and the right fork net load component; comparing the fork unbalanced load rate with a preset safety threshold, and triggering a structure health warning signal when the fork unbalanced load rate exceeds the safety threshold.

[0009] As one embodiment of the present invention, the symmetrical installation of multiple strain gauge force sensors at the root of the left and right forks of the forklift specifically includes: installing a strain gauge in each of the four key stress concentration areas at the root of each fork: the upper flange, the lower flange, the inner side of the web, and the outer side of the web. Each strain gauge constitutes one arm of a Wheatstone bridge, forming a full-bridge measurement circuit to simultaneously sense the composite strain caused by bending moment, shear force, and torque.

[0010] In one embodiment of the present invention, the inertial measurement unit is fixedly installed at the center of the lower crossbeam of the forklift mast, and its sampling frequency is not less than 500 Hz. Its linear acceleration measurement range is ±10 times the gravitational acceleration, and its angular velocity measurement range is ±200 radians per second.

[0011] As one embodiment of the present invention, the process of establishing the rigid-flexible coupling dynamic model of the forklift includes: simplifying the forklift mast system into a two-degree-of-freedom transmission system composed of a rigid main beam and flexible forks; defining the generalized coordinates as the vertical displacement of the fork's center of mass and the pitch angle around the mast hinge point; and establishing a set of dynamic differential equations including the fork mass, moment of inertia, suspension stiffness, damping coefficient, and external excitation terms according to the Lagrange equations; wherein, the external excitation terms are determined by coordinate transformation of the linear acceleration and angular velocity output by the inertial measurement unit.

[0012] As one embodiment of the present invention, the calculation of the inertial disturbance force component acting on the root of the fork under the current dynamic working condition specifically includes: converting the linear acceleration data output by the inertial measurement unit to the local coordinate system of the root of the fork; calculating the translational inertial force by combining the mass distribution parameters of the fork; obtaining the angular acceleration by differentiating the angular velocity data with respect to time, and calculating the rotational inertial torque by combining the moment of inertia of the fork about the root; projecting the translational inertial force and the rotational inertial torque onto the measurement direction of the strain gauge force sensor to obtain the inertial disturbance force component corresponding to each sensor.

[0013] As one embodiment of the present invention, the formula for calculating the fork off-center load rate is: the fork off-center load rate is equal to the absolute value of the difference between the net load component of the left fork and the net load component of the right fork divided by the sum of the two, and then multiplied by 100%.

[0014] As one embodiment of the present invention, the preset safety threshold is dynamically adjusted according to the rated lifting capacity and fork length of the forklift, and its value ranges from 15% to 25%.

[0015] This invention provides a forklift structural health status assessment system, which includes: The fork force sensing module is used to symmetrically install multiple strain gauge force sensors at the root of the left and right forks of the forklift to collect the force signals of the left and right forks in three-dimensional space in real time. The motion state sensing module is used for collecting six-degree-of-freedom motion state data of the forklift in real time through an inertial measurement unit installed on the main body of the forklift. The dynamic interference decoupling module is used for taking the six-degree-of-freedom motion state data as input, calculating the inertial interference force component acting on the root of the fork under the current dynamic working condition based on a preset rigid-flexible coupling dynamics model of the forklift, and subtracting the corresponding inertial interference force component from the stress signals of the left fork and the right fork respectively to obtain the left fork net load component and the right fork net load component. The unbalance load evaluation module is used for calculating the fork unbalance load rate according to the left fork net load component and the right fork net load component. The safety warning module is used for comparing the fork unbalance load rate with a preset safety threshold, and triggering a structure health warning signal when the fork unbalance load rate exceeds the safety threshold.

[0016] As an embodiment of the present application, the strain type force sensor in the fork force sensing module adopts a temperature self-compensation type foil strain gauge, the nominal resistance of which is 120 ohms, the sensitivity coefficient is 2.0, and the working temperature range is -40℃ to +80℃.

[0017] As an embodiment of the present application, the dynamic interference decoupling module is arranged in a forklift vehicle-mounted controller, the controller adopts a dual-core processor architecture, wherein the core is dedicated to running a dynamics decoupling algorithm, and the algorithm execution period is 2 milliseconds.

[0018] As an embodiment of the present application, the safety warning module outputs the warning signal through an audible and visual alarm device in the cab of the forklift, the audible and visual alarm device includes a red flashing warning light and a buzzer, a yellow warning is started when the fork unbalance load rate exceeds 80% of the safety threshold, and a red emergency warning is started when the fork unbalance load rate exceeds the safety threshold.

[0019] As an embodiment of the present application, the system further includes a data storage unit, which is used for recording the fork net load component, the fork unbalance load rate and the time stamp of the warning event in each operation process, forming a structure health history file, and used for subsequent fatigue life analysis and preventive maintenance decision.

[0020] Compared with the prior art, the present application has the following beneficial effects: The present application fundamentally breaks through the dependence of the traditional weighing technology on the static working condition by deploying a high-bandwidth strain sensing array at the root of the fork and synchronously acquiring the six-degree-of-freedom motion state of the whole vehicle, and constructing an online decoupling mechanism of the dynamic interference force based on the rigid-flexible coupling dynamics model.

[0021] The present application can separate the pure static load component caused by cargo unbalanced load in real time and accurately under the complex dynamic scene of forklift driving, steering, acceleration and deceleration and passing through uneven road, so as to realize millisecond level quantitative evaluation of uneven force on left and right forks. The structure health index and grading warning signal generated based on the evaluation result can effectively prevent major safety accidents such as gantry plastic deformation, fork root crack propagation and even vehicle instability and overturning caused by long-term unbalanced load operation.

[0022] The structure health history file recorded by the system provides a high-value data basis for the preventive maintenance and residual life prediction of the forklift, and significantly improves the safety, reliability and whole life cycle management efficiency of the forklift operation. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is the overall technical scheme architecture schematic diagram of the forklift structure health state evaluation method and system proposed by the present application; Figure 2 is the dynamic disturbance force self-adaptive decoupling core principle framework schematic diagram based on the rigid-flex coupled dynamics model in the present application; Figure 3 is the multi-source dynamic perception logic flow framework diagram of the fork force perception and motion state synchronous acquisition in the present application; Figure 4 is the data processing logic flow framework diagram of the inertial disturbance force component online calculation and net load separation in the present application; Figure 5 is the safety evaluation logic flow framework diagram of the fork unbalanced load rate calculation, threshold comparison and grading warning generation in the present application; Figure 6 is the multi-level interaction relationship and data flow schematic diagram between the fork force perception module and the vehicle-mounted controller in the present application. DETAILED DESCRIPTION

[0024] Please refer to Figures 1 to 6 The present application provides a forklift structure health state evaluation method and system, aiming to solve the technical problem that the structural imbalance risk caused by cargo unbalanced load during the dynamic handling operation of the forklift cannot be identified and warned in real time. In the prior art, the weighing sensor configured on the forklift can only measure the weight in the completely static state of the vehicle, and its working principle depends on the static mechanics balance condition. When the forklift is in the dynamic working condition of driving, steering, acceleration and deceleration or passing through uneven road, the sensor output signal is seriously polluted by the inertial force, centrifugal force and vibration disturbance, and cannot accurately reflect the actual load distribution of the fork, thereby losing the ability to distinguish the uneven force on the left and right forks. This technical limitation makes the operator unable to know whether the fork is in a dangerous unbalanced load state, and long-term operation is easy to cause safety accidents such as gantry plastic deformation, fork root crack propagation and even vehicle overturning.

[0025] To overcome the above defects, the application proposes a forklift structure health state evaluation method based on multi-source dynamic perception and adaptive mechanical decoupling. The method discards the traditional weighing logic which relies on static balance assumption, and constructs a closed-loop evaluation system which integrates high-frequency mechanical sensing, motion state identification and real-time dynamics modeling. The system symmetrically arranges a high-bandwidth strain force sensor array at the root of the left and right forks, synchronously collects the six-degree-of-freedom motion state data of the forklift, uses the established rigid-flexible coupled dynamics model of the forklift to decouple the dynamic disturbance force online, thereby accurately separates the pure static component caused by the load imbalance of the goods, realizes the millisecond-level quantitative evaluation of the force difference between the left and right forks, and generates a structure health index and a safety warning signal on this basis.

[0026] The forklift structure health state evaluation method comprises the following steps: S1, symmetrically installing a plurality of strain force sensors at the root of the left fork and the root of the right fork of the forklift, the strain force sensors being used to collect force signals of the left fork and the right fork in three-dimensional space in real time; S2, collecting six-degree-of-freedom motion state data of the forklift in real time through an inertial measurement unit installed on the main body of the forklift, the six-degree-of-freedom motion state data including linear acceleration along three orthogonal axes and angular velocity around three orthogonal axes; S3, based on a preset rigid-flexible coupled dynamics model of the forklift, taking the six-degree-of-freedom motion state data as input, calculating the inertial disturbance force component acting on the fork root under the current dynamic working condition; S4, subtracting the corresponding inertial disturbance force component from the force signals of the left fork and the right fork respectively to obtain the left fork net load component and the right fork net load component; S5, calculating the fork load imbalance rate according to the left fork net load component and the right fork net load component; S6, comparing the fork load imbalance rate with a preset safety threshold, and triggering a structure health warning signal when the fork load imbalance rate exceeds the safety threshold.

[0027] In step S1, a plurality of strain force sensors are symmetrically installed at the left and right fork root of the forklift. Each fork root has four key stress concentration areas, i.e., the upper flange, the lower flange, the inner side of the web, and the outer side of the web, each of which is equipped with a strain gauge. Each strain gauge constitutes an arm of a Wheatstone bridge, forming a full-bridge measurement circuit to simultaneously sense the combined strain caused by bending moment, shear force and torque. The strain force sensor uses a temperature self-compensating foil strain gauge with a nominal resistance of 120 ohms, a sensitivity coefficient of 2.0, and a working temperature range of -40°C to +80°C. All strain gauges are connected to a signal conditioning module through shielded twisted pair wires. The signal conditioning module integrates a constant current excitation source, a low-pass anti-aliasing filter, and a programmable gain amplifier. The sampling frequency is set to 2000 Hz to ensure complete capture of the high-frequency dynamic response of the fork under bumpy road conditions or sharp turning conditions.

[0028] In step S2, an inertial measurement unit is fixedly installed at the center of the lower beam of the forklift mast. The sampling frequency is not less than 500 Hz, the linear acceleration measurement range is ±10 times the gravitational acceleration, and the angular velocity measurement range is ±200 rad / s. The inertial measurement unit includes a three-axis micro-electro-mechanical system (MEMS) accelerometer and a three-axis MEMS gyroscope, which share the same reference coordinate system and have undergone six-face calibration and temperature drift compensation before leaving the factory. The data output by the inertial measurement unit passes through a time stamp synchronization module, which aligns the data stream with that of the strain force sensor in the time domain, with a time synchronization error controlled within 50 microseconds. The time stamp synchronization module uses a hardware trigger mechanism, with a unified sampling enable signal sent by the main controller to ensure that the multi-source sensing data are strictly consistent in physical time.

[0029] In step S3, based on a preset rigid-flexible coupled dynamics model of the forklift, the six-degree-of-freedom motion state data are taken as inputs to calculate the inertial disturbance force component acting on the fork root under the current dynamic working condition. The establishment process of the rigid-flexible coupled dynamics model of the forklift includes: simplifying the forklift mast system into a two-degree-of-freedom transmission system composed of a rigid main beam and a flexible fork; defining the generalized coordinates as the vertical displacement of the fork mass center and the pitch angle around the mast hinge point; establishing a set of differential equations of motion containing the fork mass, the moment of inertia, the suspension stiffness, the damping coefficient, and the external excitation term according to the Lagrange equation; wherein the external excitation term is determined by the linear acceleration and angular velocity output by the inertial measurement unit after coordinate transformation.

[0030] Specifically, let the total mass of the fork be , the horizontal distance from the mass center of the fork to the hinge point of the mast be , the moment of inertia of the fork around the hinge point be , the equivalent suspension stiffness of the mast be , and the equivalent damping coefficient be . Define the generalized coordinate vector , where is the vertical displacement of the forklift's center of mass, is the pitch angle of the forklift around the hinge point. The Lagrangian function of the system is:

[0031] Considering the small angle approximation and introducing the damping dissipation function , the motion differential equations are obtained:

[0032] where, and are the external excitation force and torque, which are calculated from the vehicle's linear acceleration and angular acceleration measured by the inertial measurement unit (IMU) after coordinate transformation. Let the linear acceleration of the vehicle body measured by the IMU be and the angular velocity be , then the linear acceleration of the forklift's root in the local coordinate system is:

[0033] where, is the rotation matrix from the vehicle body coordinate system to the forklift's root local coordinate system, is the vector from the IMU mounting point to the forklift's root, is the angular acceleration. The angular acceleration is calculated by numerical differentiation and substituted into the above equation, which gives the instantaneous acceleration field of the forklift's root. Combined with the mass distribution parameters of the forklift, the linear inertia force is calculated; the angular acceleration is obtained by taking the derivative of the angular velocity data with respect to time, and combined with the moment of inertia of the forklift around the root, the rotational inertia torque is calculated. Finally, the linear inertia force and rotational inertia torque are projected onto the sensitive direction of each strain force sensor to obtain the inertia interference force component corresponding to each sensor.

[0034] In step S4, the corresponding inertia interference force components are subtracted from the force signals of the left and right forks respectively to obtain the left and right fork net load components. The force signal of each fork is output by a full-bridge circuit composed of 4 strain gauges, converted into a voltage signal after signal conditioning, and then inversely solved into a three-dimensional force vector and by a calibration coefficient matrix. Since the forklift mainly bears vertical load, the original load signals are taken as and . Correspondingly, the vertical components of the inertia interference force calculated from the dynamic model are and . Then the left fork net load component Net load component of the right fork To improve robustness, a sliding window mid-range filter with a window length of 20 milliseconds is applied to the net load component to suppress residual high-frequency noise.

[0035] In step S5, the fork off-center load ratio is calculated based on the net load components of the left and right forks. The formula for calculating the fork off-center load ratio is:

[0036] This formula ensures that the eccentricity ratio varies continuously between 0 and 100%. The eccentricity ratio is 0 when the loads on the left and right sides are completely equal, and reaches 100% when one side is completely unloaded while the other side bears the full weight. During the calculation, if... If the load is less than 5% of the rated lifting capacity, it is determined to be in a state of no effective load, and the off-center load rate is set to 0 to avoid false alarms caused by no-load shaking.

[0037] In step S6, the fork off-center load rate is compared with a preset safety threshold. When the fork off-center load rate exceeds the safety threshold, a structural health warning signal is triggered. The preset safety threshold is dynamically adjusted based on the forklift's rated lifting capacity and fork length, and its value ranges from 15% to 25%. Specifically, the safety threshold... Determined by the following formula:

[0038] in, The base threshold is set at 15%. This is the effective length of the currently installed forks; The standard fork length is typically 1.2 meters. The length correction factor is set to 10%. Therefore, for a two-meter-long extended fork, the safety threshold is 25%. The system has a two-level warning mechanism: when the fork offset rate exceeds 80% of the safety threshold, a yellow warning is activated, and the yellow warning light in the cab remains on, prompting the operator to adjust the cargo position; when the fork offset rate exceeds the safety threshold, a red emergency warning is activated, and the red flashing warning light and buzzer are activated simultaneously, forcing the operator to immediately stop work and reload.

[0039] The forklift structure health state evaluation system comprises a fork force sensing module, a motion state sensing module, a dynamic interference decoupling module, a load bias evaluation module and a safety warning module. The fork force sensing module symmetrically installs a plurality of strain force sensors at the left fork root and the right fork root to collect three-dimensional force signals in real time. The motion state sensing module collects six-degree-of-freedom motion state data through an inertial measurement unit. The dynamic interference decoupling module is arranged in a forklift vehicle-mounted controller. The controller adopts a dual-core processor architecture, wherein a core is dedicated to running a dynamic decoupling algorithm, and the algorithm execution period is 2 milliseconds. The load bias evaluation module calculates the load bias rate according to the decoupled net load component. The safety warning module outputs a graded warning signal through an audible and visual alarm device. The system further comprises a data storage unit for recording the fork net load component, the fork load bias rate and the time stamp of the warning event in each operation process, forming a structure health history file for subsequent fatigue life analysis and preventive maintenance decision. The data storage unit adopts a non-volatile flash memory with a storage capacity of 8 gigabytes, supports cyclic overwrite writing, and retains complete operation records for the last 360 days.

[0040] The working process of the whole system is as follows: after power-on initialization, each sensor completes self-checking and zero-point calibration; after the operation starts, the fork force sensing module and the motion state sensing module synchronously collect data; the dynamic interference decoupling module performs decoupling operation once every 2 milliseconds, and outputs the net load component; the load bias evaluation module updates the load bias rate in real time; the safety warning module continuously compares the threshold and drives the alarm device; and the data storage unit records the key parameters at a second granularity. All modules are interconnected through a controller area network bus, the communication baud rate is 500 kilobits per second, and the real-time and reliability of data transmission are ensured.

[0041] Through the above method and system, the present application realizes real-time monitoring of dynamic load bias of the forklift under all working conditions, solves the fundamental defect that the traditional static weighing technology cannot be applied to mobile scenes, and significantly improves the intelligent level and accident prevention ability of forklift structure safety monitoring.

Claims

1. A method for evaluating the health state of a forklift structure, characterized by, The method comprises the following steps: A plurality of strain force sensors are symmetrically installed at the left fork root and the right fork root of the forklift, and the strain force sensors are used to collect force signals of the left fork and the right fork in three-dimensional space in real time; A six-degree-of-freedom motion state data of the forklift is collected in real time by an inertial measurement unit installed on the main body of the forklift, and the six-degree-of-freedom motion state data includes linear acceleration along three orthogonal axes and angular velocity around three orthogonal axes; Based on a preset rigid-flexible coupling dynamics model of the forklift, the six-degree-of-freedom motion state data is taken as input to calculate an inertial interference force component acting on the fork root under a current dynamic working condition; The inertial interference force component is subtracted from the force signals of the left fork and the right fork respectively to obtain a left fork net load component and a right fork net load component; The left fork net load component and the right fork net load component are used to calculate a fork load deviation rate; The fork load deviation rate is compared with a preset safety threshold, and a structure health warning signal is triggered when the fork load deviation rate exceeds the safety threshold.

2. The method of assessing the structural health of a forklift truck according to claim 1, wherein A plurality of strain force sensors are symmetrically installed at the left fork root and the right fork root of the forklift, and the method comprises the following steps: One strain gauge is installed at each of four key stress concentration areas of the upper flange, the lower flange, the inner side of the web and the outer side of the web of each fork root, each strain gauge constitutes one arm of a Wheatstone bridge, and a full-bridge measurement circuit is formed to simultaneously sense the composite strain caused by the bending moment, the shear force and the torque.

3. The method of assessing the structural health of a forklift truck according to claim 2, wherein, The inertial measurement unit is fixedly installed at the center position of the lower crossbeam of the forklift mast, the sampling frequency is not less than 500 Hz, the linear acceleration measurement range is ±10 times of the gravity acceleration, and the angular velocity measurement range is ±200 rad / s.

4. The method of assessing the structural health of a forklift truck according to claim 3, wherein The establishment process of the rigid-flexible coupling dynamics model of the forklift comprises the following steps: The forklift mast system is simplified into a two-degree-of-freedom transmission system composed of a rigid main beam and a flexible fork; A generalized coordinate is defined as the vertical displacement of the mass center of the fork and the pitch angle around the hinge point of the mast; According to the Lagrange equation, a set of differential equations of dynamics is established, which includes the mass of the fork, the moment of inertia, the suspension stiffness, the damping coefficient and an external excitation term; The external excitation term is determined by the linear acceleration and the angular velocity output by the inertial measurement unit after coordinate transformation.

5. The method of assessing the structural health of a forklift truck according to claim 4, wherein, The inertial interference force component acting on the fork root under the current dynamic working condition is calculated, and the method comprises the following steps: The linear acceleration data output by the inertial measurement unit is converted to the local coordinate system of the fork root; The translational inertia force is calculated in combination with the mass distribution parameters of the fork; The angular acceleration is obtained by differentiating the angular velocity data with respect to time, and the rotational inertia moment is calculated in combination with the moment of inertia of the fork around the root; The translational inertia force and the rotational inertia moment are projected to the measurement direction of the strain force sensor to obtain the corresponding inertial interference force component of each sensor.

6. The method of assessing the structural health of a forklift truck according to claim 5, wherein, The left fork net load component and the right fork net load component are obtained, and the method comprises the following steps: The force signals of the left fork and the right fork are respectively converted into three-dimensional force vectors; The vertical component of the three-dimensional force vector is extracted as an original load signal; The left fork net load component and the right fork net load component are obtained by subtracting the vertical component of the corresponding inertial interference force from the original load signal respectively. The left and right fork net load components are subjected to sliding window median filtering processing.

7. The method of assessing the structural health of a forklift truck according to claim 6, wherein, The formula for calculating the fork load imbalance rate is: The fork load imbalance rate is equal to the absolute value of the difference between the left and right fork net load components divided by the sum of the two, multiplied by 100%. When the sum of the left and right fork net load components is less than 5% of the forklift's rated load, the fork load imbalance rate is set to 0.

8. The method of assessing the structural health of a forklift truck according to claim 7, wherein, The preset safety threshold is dynamically adjusted according to the forklift's rated load and fork length, with a value range of 15% to 25%. When the fork load imbalance rate exceeds 80% of the safety threshold, a yellow pre-warning signal is triggered. When the fork load imbalance rate exceeds the safety threshold, a red emergency pre-warning signal is triggered.

9. A forklift structure health status evaluation system characterized by, It includes: A fork force sensing module for symmetrically installing multiple strain force sensors at the left and right fork roots of the forklift to real-time collect force signals of the left and right forks in three-dimensional space; A motion state sensing module for real-time collecting six-degree-of-freedom motion state data of the forklift through an inertial measurement unit installed on the forklift's main vehicle body; A dynamic interference decoupling module for calculating the inertial interference force components acting on the fork roots under the current dynamic working condition based on a preset forklift rigid-flex coupling dynamics model, taking the six-degree-of-freedom motion state data as input, and subtracting the corresponding inertial interference force components from the force signals of the left and right forks respectively to obtain the left and right fork net load components; An imbalance evaluation module for calculating the fork load imbalance rate according to the left and right fork net load components; A safety pre-warning module for comparing the fork load imbalance rate with a preset safety threshold and triggering a structural health pre-warning signal when the fork load imbalance rate exceeds the safety threshold.

10. The forklift structural health state assessment system according to claim 9, characterized by, The strain force sensors in the fork force sensing module use temperature self-compensating foil strain gauges with a nominal resistance of 120 ohms, a sensitivity coefficient of 2.0, and a working temperature range of -40°C to +80°C.

Images