Method and system for driving a body sculptor

Abstract

The application relates to a body shaping machine driving method and system, solves the problem that the existing body shaping machine adopts fixed gear open-loop control, cannot be personalized adapted according to the weight and standing posture of a user, ignores vibration cross coupling and load disturbance among multiple excitation sources, and causes actual vibration to deviate from expectation, the method comprises the following steps: obtaining weight and pressure distribution, combining preset parameters to construct a man-machine coupling model, and then solving a multi-source transfer operator containing a direct channel and a cross coupling channel, wherein the cross coupling channel is caused by a finite value of rotational constraint stiffness; then driving the vibration mechanism after sequentially performing feedforward compensation, amplitude and phase correction and vibration field consistency optimization on the expected vibration parameters input by the user; monitoring the motor current and rotating speed during operation, and performing load disturbance compensation correction on the driving signal until the operation is completed or a stop instruction is received. The application has the following effects: realizing precise control and load adaptive compensation of multi-source vibration of the body shaping machine, and improving vibration stability and uniformity.

Description

Technical Field

[0001] This invention relates to the field of body shaping machine control technology, and in particular to a body shaping machine driving method and system. Background Technology

[0002] A body shaping machine is a type of home fitness equipment. Users stand on the foot pedals, and the machine's internal drive mechanism causes these pedals to vibrate rhythmically, transmitting the vibrations to the body to achieve muscle relaxation and body shaping effects. The core capability of a body shaping machine lies in its ability to accurately and evenly vibrate the foot pedals according to the user's preset vibration pattern, thus providing an ideal vibration experience.

[0003] Most body shaping machines on the market today operate in a relatively simple and direct manner. Users set the vibration intensity or select a preset vibration level via the control panel. The machine's control circuit then directly drives the vibration motor to operate at a fixed speed or a preset voltage curve. The motor drives an eccentric wheel or linear motion mechanism to generate vibration, which is transmitted to the foot pedal area through the mechanical structure. Some products incorporate simple feedback adjustments, such as using sensors to detect the actual motor speed. When the speed deviates from the set value due to load changes, the drive voltage is automatically adjusted to maintain a relatively stable speed.

[0004] Traditional control methods can meet basic vibration output requirements, but because the determination of their drive commands mainly relies on preset fixed programs or simple closed-loop speed adjustment, they do not fully consider the changes in the vibration characteristics of the entire system when the user stands on the machine. When the user's weight or standing posture changes, the actual vibration effect of the machine's foot pedal area will vary significantly, making it difficult to guarantee a consistent and stable vibration experience under different usage conditions. Summary of the Invention

[0005] In order to achieve precise control of multi-source vibration and adaptive load compensation of the body shaping machine, and improve vibration stability and uniformity, this application provides a body shaping machine driving method and system.

[0006] In a first aspect, this application provides a method for driving a body shaping machine, which adopts the following technical solution:

[0007] A method for driving a body shaping machine, the body shaping machine including a base frame body, a top frame body, a movable support member, a vibration mechanism, and a sensing unit, the top frame body having a foot pedal area, the movable support member having two ends rotatably connected to the top frame body and the base frame body respectively, the vibration mechanism being located on the base frame body and connected to the top frame body, and the sensing unit being arranged in the foot pedal area; including the following steps:

[0008] The system acquires body weight and pressure distribution through sensing units, retrieves preset human body parameters, top frame structural parameters and rotational constraint stiffness, determines the mass matrix of the human-machine coupling model based on the body weight and preset human body parameters according to preset mass distribution rules, determines the load distribution matrix of the user-top frame contact interface based on the pressure distribution according to preset load mapping rules, and integrates the mass matrix, load distribution matrix, top frame structural parameters and rotational constraint stiffness to construct the human-machine coupling model.

[0009] Based on the human-machine coupling model, the installation position of the vibration mechanism is taken as the excitation point and the foot pedal area is taken as the response point. The multi-source transmission operator containing direct channels and cross-coupled channels is solved according to the preset transmission operator construction rules. The cross-coupled channels are triggered by the rotational constraint stiffness being a finite value.

[0010] The system receives the desired vibration parameters input by the user, performs feedforward compensation on the desired vibration parameters based on the multi-source transfer operator according to the preset feedforward compensation rules, and obtains the theoretical excitation displacement sequence.

[0011] The theoretical excitation displacement sequence is corrected according to the preset amplitude and phase correction rule, and the vibration field is optimized according to the preset consistency optimization rule. The vibration mechanism is then converted into a drive control signal according to the preset drive conversion rule.

[0012] During vibration operation, the operating current and speed of the motor of the vibration mechanism are monitored. Based on the motor operating current, speed and multi-source transmission operator, the drive control signal is corrected according to the preset load compensation rules to compensate for load disturbances until the user-set running time is reached or a stop command is received.

[0013] By adopting the above technical solution, a differentiated human-machine coupling model can be established according to the weight and standing posture of different users. By compensating for the vibration cross-coupling and load disturbance between multiple excitation sources, the actual vibration response of the body shaping machine can be made close to the expected trajectory, and the vibration uniformity and operation stability of the foot pedal area can be improved.

[0014] Secondly, this application provides a body shaping machine drive system, which adopts the following technical solution:

[0015] A body shaping machine drive system, the body shaping machine includes a base frame, a top frame, a movable support component, a vibration mechanism, and a sensing unit. The top frame has a foot pedal area. The two ends of the movable support component are rotatably connected to the top frame and the base frame, respectively. The vibration mechanism is located on the base frame and connected to the top frame. The sensing unit is arranged in the foot pedal area. The system includes:

[0016] The sensing unit is used to acquire the user's weight and pressure distribution;

[0017] The model building unit is used to retrieve preset human body parameters, top frame structural parameters and rotational constraint stiffness. The mass matrix of the human-machine coupling model is determined by the weight and preset human body parameters according to the preset mass distribution rules. The load distribution matrix of the user-top frame contact interface is determined by the pressure distribution according to the preset load mapping rules. The human-machine coupling model is constructed by fusing the mass matrix, load distribution matrix, top frame structural parameters and rotational constraint stiffness.

[0018] The transfer operator solving unit is used to solve the multi-source transfer operator, which includes direct channels and cross-coupled channels, based on the human-machine coupling model, with the installation position of the vibration mechanism as the excitation point and the foot pedal area as the response point, according to the preset transfer operator construction rules. The cross-coupled channels are triggered by the rotational constraint stiffness being a finite value.

[0019] The feedforward compensation unit is used to receive the desired vibration parameters input by the user, and perform feedforward compensation on the desired vibration parameters based on the multi-source transfer operator according to the preset feedforward compensation rules to obtain the theoretical excitation displacement sequence.

[0020] The correction, optimization, and drive unit is used to correct the amplitude and phase of the theoretical excitation displacement sequence according to a preset amplitude and phase correction rule, optimize the vibration field consistency according to a preset consistency optimization rule, and convert it into a drive control signal to drive the vibration mechanism according to a preset drive conversion rule.

[0021] The load compensation unit is used to monitor the motor operating current and speed of the vibration mechanism during vibration operation, and to correct the drive control signal based on the motor operating current, speed and the multi-source transmission operator according to the preset load compensation rules to compensate for load disturbances.

[0022] The operation control unit is used to control the vibration mechanism to stop operating when the user-set operating time is reached or when a stop command is received. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall process of a body shaping machine driving method according to an embodiment of this application.

[0024] Figure 2 This is a comparison chart of the relative deviation of the amplitude of the foot pedal area between the present application and the prior art in the full frequency band. The horizontal axis is the vibration frequency, the vertical axis is the relative deviation of the amplitude of the foot pedal area, and the dotted line in the figure is the limit of the preset tolerance of 10%.

[0025] Figure 3 This is a schematic diagram of the body shaping machine according to an embodiment of this application.

[0026] In the diagram, 1 is the base frame body; 2 is the top frame body; 3 is the movable support component; and 4 is the vibration mechanism. Detailed Implementation

[0027] The present application will be further described in detail below with reference to the accompanying drawings.

[0028] Reference Figure 3 The body shaping machine of this application embodiment includes a base frame body 1, a top frame body 2, a movable support member 3, a vibration mechanism 4, and a sensing unit. The base frame body 1 includes two opposing side beams, which serve as the load-bearing base of the entire machine. During use, it is placed on the ground or a support surface to bear the weight of the entire machine and isolate the transmission of vibration to the ground. The top frame body 2 is a grid-like frame structure, positioned above the base frame body 1. The upper surface of the top frame body 2 has a footrest area for the user to stand on. The movable support member 3 is located between the top frame body 2 and the base frame body 1. Its two ends are rotatably connected to the top frame body 2 and the base frame body 1 via a pivot or hinge structure, allowing the top frame body 2 to perform vertical translation relative to the base frame body 1 and a small rotation around a horizontal axis. An elastic element (such as a rubber bushing or torsion spring) is provided at the rotatable connection point to provide elastic resistance to the rotation of the top frame body 2 relative to the base frame body 1, i.e., the rotational constraint stiffness described in this application. The vibration mechanism 4 is located on the base frame body 1 and connected to... In the top frame body 2, the vibration mechanism 4 can be a dual-head vibration motor (with eccentric blocks installed at both ends of its rotor, generating excitation force through the rotation of the eccentric blocks), a horizontal vibration component, or a vertical vibration component, used to generate controlled mechanical vibration under the action of the drive control signal and transmit it to the top frame body 2; the sensing unit is arranged in the foot pedal area, including a pressure sensing array and a weight detection module, used to acquire the user's weight and foot pressure distribution; the body shaping machine also includes a control unit (not shown in the figure), which is electrically connected to the sensing unit and the vibration mechanism 4, used to execute the driving method described in this application and output the drive control signal; the control unit is equipped with a memory, a human-machine interface, and a power drive module, the memory is used to pre-store various preset parameters, preset rules, databases, and mapping tables described in this application, the human-machine interface is used to receive the user's set desired vibration parameters and start / stop commands, and the power drive module is used to amplify the drive control signal into a drive current and output it to the vibration mechanism 4. The above structure is roughly the same as the conventional structure of existing body shaping machines, and the improvement of this application lies in the driving method rather than the mechanical structure itself.

[0029] Reference Figure 1 The present application discloses a method for driving a body shaping machine, comprising:

[0030] Step S100: Obtain weight and pressure distribution through the sensing unit, retrieve preset human body parameters, top frame structural parameters and rotational constraint stiffness, determine the mass matrix of the human-machine coupling model based on the weight and preset human body parameters according to the preset mass distribution rules, determine the load distribution matrix of the user-top frame contact interface based on the pressure distribution according to the preset load mapping rules, and integrate the mass matrix, load distribution matrix, top frame structural parameters and rotational constraint stiffness to construct the human-machine coupling model.

[0031] The sensing unit refers to the combination of a pressure sensing array and a weight detection module deployed in the foot pedal area. In this application, it specifically refers to a measurement unit that can simultaneously output the user's overall weight and two-dimensional pressure distribution data of the sole of the foot. The preset human body parameters refer to the reference parameters stored in the system memory in advance to describe the user's human body geometry and inertial characteristics. In this application, these include height, gender, and lower limb length-to-diameter ratio. The lower limb length-to-diameter ratio refers to the ratio of thigh length to calf length, used to characterize the geometric proportions of the user's lower limbs.

[0032] The structural parameters of the top frame refer to the physical quantities that characterize the material and geometric properties of the top frame, including the elastic modulus, moment of inertia of the section, material density and geometric dimensions in this application; the rotational constraint stiffness refers to the elastic resistance of the rotating connection at both ends of the movable support 3 to the rotation of the top frame relative to the base frame, and in this application specifically refers to the boundary elastic coefficient in units of Newton-meter per radian (N·m / rad).

[0033] The stiffness of the human-top frame contact interface refers to the stiffness parameter describing the elastic contact characteristics between the user's foot and the footrest area of ​​the top frame. In this application, it includes vertical contact stiffness and rotational coupling stiffness, which are used to realize the elastic coupling between the human body's degrees of freedom and the top frame's degrees of freedom. The human-machine coupling model refers to the overall dynamic system obtained by integrating the user's human body mass-inertia characteristics, the top frame's structural dynamic characteristics, and the human-machine contact interface coupling characteristics into a unified model. In this application, it specifically refers to the multi-degree-of-freedom coupled dynamic equation described by the mass-stiffness-damping matrix. The transverse horizontal axis refers to the horizontal axis along the width direction of the top frame and passing through the geometric center of the top frame. The rotational degrees of freedom of the top frame are all defined around this axis.

[0034] The preset mass distribution rule refers to the rule of allocating the user's weight to each limb segment according to anthropometry proportions, calculating the inertia of each segment, and then assembling the system mass matrix. One implementation method is the lookup table method, which obtains the mass proportion of each segment by querying the preset human body segment mass distribution database stored in memory based on the user's gender and height. The preset load mapping rule refers to the rule of mapping the two-dimensional pressure distribution of the foot to a load distribution matrix through filtering, mesh discretization, pressure integration, node allocation, boundary correction, and degree of freedom assembly.

[0035] The necessary process is as follows: The user stands on the footrest area of ​​the top frame, and the weight detection module in the sensing unit outputs the user's weight. The pressure sensor array outputs a two-dimensional pressure distribution field on the sole of the foot. The system retrieves preset human body parameters (height) from memory. Gender identification and lower limb length-to-diameter ratio ), Top frame structural parameters (elastic modulus) Moment of inertia of cross section Material density ,length ,width and thickness Rotational constraint stiffness at the rotating connection of movable support member 3 and the vertical stiffness of the human-top frame contact interface Coupling stiffness with rotation Elastic modulus With the moment of inertia of the cross section Used for offline verification of the first-order bending natural frequency of the top frame. ,in This represents the cross-sectional area of ​​the top frame.

[0036] For the process of determining the mass matrix, please refer to steps S110 to S160 in Example 2; for the process of determining the load distribution matrix, please refer to steps S1A0 to S1F0 in Example 3; for the process of constructing the human-machine coupling model, please refer to steps S1a0 to S1d0 in Example 4.

[0037] Step S200: Based on the human-machine coupling model, with the installation position of the vibration mechanism 4 as the excitation point and the foot pedal area as the response point, solve the multi-source transmission operator containing direct channels and cross-coupled channels according to the preset transmission operator construction rules. The cross-coupled channels are triggered by the rotational constraint stiffness being a finite value.

[0038] Among them, the multi-source transfer operator refers to the frequency domain function matrix that describes the vibration transfer relationship from each excitation degree of freedom to each response degree of freedom. In this application, it specifically refers to the complete frequency response matrix assembled according to the frequency discrete axis, the excitation degree of freedom number and the response degree of freedom number; the frequency discrete axis refers to the set of discrete frequency points determined according to the preset distribution law in the target vibration frequency band, and the multi-source transfer operator takes a value at each frequency point in the set.

[0039] A direct path refers to the vibration transmission path from a certain excitation degree of freedom to its geometrically corresponding response degree of freedom; a cross-coupling path refers to the non-diagonal vibration transmission path between the excitation degree of freedom and the geometrically non-corresponding response degree of freedom because the rotational constraint stiffness is a finite value and the boundary rotational degree of freedom is not completely fixed; the pre-defined transmission operator construction rule refers to the rule of constructing the excitation-response frequency response matrix based on the mass-stiffness-damping coefficient matrix, through modal solution and modal superposition on the frequency discrete axis.

[0040] The necessary process is as follows: Solving the multi-source transfer operator with the mass matrix of the human-machine coupling model. Damping matrix and overall stiffness matrix The coefficient matrix is ​​used to construct the frequency response matrix from the excitation degree of freedom to the response degree of freedom at each frequency point on the preset frequency discrete axis. Specifically, the corresponding degree of freedom number is first determined according to the physical installation position of the excitation point and the response point, and then the dominant natural frequency and mode shape are solved by the coefficient matrix. Then, the complete frequency response matrix is ​​constructed according to the modal superposition principle. Since the boundary rotation constraint stiffness is a finite value, the projection of the mode shape on the non-corresponding degree of freedom causes the frequency response matrix to naturally have off-diagonal terms, i.e., cross-coupling channels.

[0041] For the specific solution process of the multi-source transit operator, refer to steps S210 to S240 and steps S241 to S245.

[0042] Step S300: Receive the desired vibration parameters input by the user, and perform feedforward compensation on the desired vibration parameters based on the multi-source transfer operator according to the preset feedforward compensation rule to obtain the theoretical excitation displacement sequence.

[0043] Among them, the desired vibration parameters refer to the target vibration conditions input by the user through the control panel or mobile terminal, which include vibration frequency, vibration amplitude and vibration mode identifier in this application; feedforward compensation refers to the technique of performing inverse calculation on the desired response based on the known transmission characteristics of the system before vibration output, and pre-calculating the required excitation displacement; theoretical excitation displacement sequence refers to the time-domain displacement command sequence obtained after feedforward compensation for driving the vibration mechanism 4.

[0044] The necessary process is as follows: The overall idea of ​​feedforward compensation is to take the vibration response of the user's desired foot pedal area as the target, and use the known system transmission characteristics of the multi-source transfer operator to perform inverse operation to deduce the excitation displacement required to generate the response; specifically, firstly, the received frequency, amplitude and mode identifier are checked and analyzed to obtain the desired vibration response, then the frequency response matrix of the corresponding frequency is taken out from the multi-source transfer operator and inverse operation is performed to obtain the frequency domain excitation displacement, and finally the theoretical excitation displacement sequence in the time domain is reconstructed according to the sampling period.

[0045] For details on the feedforward compensation process, refer to steps S310 to S360.

[0046] Step S400: The theoretical excitation displacement sequence is corrected according to the preset amplitude and phase correction rule, and the vibration field is optimized according to the preset consistency optimization rule. The vibration mechanism 4 is then converted into a drive control signal according to the preset drive conversion rule.

[0047] Among them, amplitude-phase correction refers to the process of adjusting the amplitude gain and compensating for phase lead / lag of the direct channel at the target vibration frequency for the theoretical excitation displacement sequence, based on the amplitude-frequency attenuation and phase-frequency lag characteristics of the direct channel in the multi-source transfer operator. This compensation is for the dominant transfer channel, and residual coupling is suppressed by subsequent consistency optimization. Consistency optimization refers to the process of optimizing the allocation of the excitation sequence after amplitude-phase correction with the goal of minimizing the amplitude deviation between the vibration responses of each degree of freedom in the foot pedal area. The drive control signal refers to the electrical control command obtained by converting the vibration mechanism into type 4 after consistency optimization.

[0048] The necessary process is as follows: The theoretical excitation displacement sequence undergoes two stages of processing before drive execution. The first stage is amplitude and phase correction, which is used to offset the amplitude attenuation and phase lag of the direct channel at the target frequency. The second stage is consistency optimization, which is used to suppress the uneven vibration amplitude of each response point caused by the cross-coupling channel. Specifically, the amplitude and phase of the theoretical excitation displacement sequence are first extracted and inverse channel compensation is performed. Then, the excitation is optimized and allocated with the goal of minimizing the predicted amplitude deviation of each response point. Finally, it is converted into a voltage signal or PWM duty cycle signal output according to the vibration mechanism type 4.

[0049] For details on amplitude and phase correction, consistency optimization, and drive conversion, please refer to steps S410 to S450.

[0050] In step S500, during vibration operation, the operating current and speed of the motor of the vibration mechanism 4 are monitored. The drive control signal is corrected according to the preset load compensation rules based on the motor operating current, speed and multi-source transmission operator to compensate for load disturbances until the user-set running time is reached or a stop command is received.

[0051] Among them, the load compensation rule refers to estimating the actual load disturbance based on the real-time operating condition feedback of the motor during vibration operation, and mapping the disturbance force into excitation degree of freedom compensation displacement through the transfer function of the multi-source transfer operator at the vibration frequency, thereby correcting the closed-loop compensation strategy of the drive output; the motor operating current refers to the real-time current of the motor winding of the drive vibration mechanism 4; the speed refers to the real-time angular velocity of the motor rotor.

[0052] The necessary process is as follows: Load compensation is a closed-loop process. The motor operating current and speed are used as measurable feedback quantities to estimate the load disturbance in real time. The disturbance is converted into a compensation displacement of the excitation degree of freedom and superimposed on the drive control signal to cancel it out. Specifically, the actual load torque is estimated from the motor current and speed through the rotor dynamics equation. The load disturbance force is obtained by subtracting the load torque from the nominal excitation force. The disturbance force is inversely mapped into the disturbance equivalent displacement by the transfer function of the multi-source transfer operator at the vibration frequency. The compensation signal is generated from the disturbance equivalent displacement and superimposed on the drive output. The process is repeated until the set time is reached or a stop command is received.

[0053] For details on the load compensation process, please refer to steps S510 to S550.

[0054] Furthermore, considering that the mass matrix of the human-machine coupling model needs to accurately reflect the inertial characteristics and distribution of the user's body and the top frame, the mass matrix of the human-machine coupling model is determined based on body weight and preset human body parameters according to preset mass distribution rules, including the following steps:

[0055] Step S110: According to the preset mass allocation rules, the total mass of the human body is determined by weight and gender correction coefficient, the total mass of the lower limbs is determined by the total mass of the human body and the preset lower limb mass ratio, and the mass of the thigh segment and the mass of the lower limb segment are determined by the total mass of the lower limbs and the preset thigh-to-calf mass ratio.

[0056] The preset human body segment mass distribution database refers to a database that is pre-established based on anthropometry statistics (such as statistical data on human body segment parameters of adults) and stored in memory for lookup during operation. It records the mass distribution ratio of each segment for different genders and heights. In this application, the mass ratio of the lower limbs and the mass ratio of the thigh and calf are determined by looking up a table in this database according to the user's gender and height. The preset mass distribution rule refers to the rule for distributing the total human body mass layer by layer to the lower limbs, thighs and calves according to the proportion obtained from the table. The gender correction coefficient refers to the correction coefficient used to correct the deviation of the weighing system caused by the difference in body water content and foot contact area between different genders. It is determined by the gender identifier in the preset human body parameters.

[0057] The necessary procedures are as follows:

[0058] The total body mass is determined by combining the weight measurement value with the gender correction factor. ,in The gender correction factor is used, with one implementation setting being 1.00 for males and 0.99 for females. When the gender correction factor is 1.00, it degenerates into determining the total body mass directly from weight measurements. Based on the user's gender and height, a preset body segment mass distribution database is consulted to determine the lower limb mass percentage. (Typical value 0.30~0.40) and thigh-to-calf mass ratio (Typical values ​​1.1~1.4); the quality of each segment is:

[0059] .

[0060] .

[0061] in, The total mass of the lower limbs, For thigh segmentation quality, The mass of the lower leg is segmented.

[0062] Step S120: According to the preset geometric ratio rules, determine the relative coordinates of the local center of mass of the thigh and the local center of mass of the lower leg based on the length-to-diameter ratio of the lower limb, and synthesize the overall center of mass coordinates of the lower limb from the segmented mass of the thigh, the segmented mass of the lower leg, and the local relative coordinates of the center of mass.

[0063] Among them, the relative coordinates of the local centroids refer to the position coordinates of the centroids of each limb segment in the local coordinate system with the proximal joint of the segment as the origin, which is used to characterize the distribution of segment mass along the longitudinal direction of the limb; the preset geometric proportion rule refers to the rule for determining the position of the centroids of each limb segment along the longitudinal direction of the limb according to the proportional constant of anthropometry statistics.

[0064] The necessary procedures are as follows:

[0065] Let the thigh length be... The length of the lower leg is Lower limb length-to-diameter ratio With the hip joint as the origin Along the longitudinal direction of the thigh towards the knee joint The positive axis direction; the relative coordinate of the local centroid of the thigh is taken at 43.3% of the thigh length, i.e. The relative coordinates of the lower leg's local centroid, with the knee joint as the local origin, are taken at 43.3% of the lower leg length, and then converted to a global coordinate system with the hip joint as the origin. The aforementioned 43.3% is derived from the proportion of the center of mass of each segment from the proximal joint in the statistics of human biomechanical parameters; the mass of the thigh segments... Lower leg segmental quality And synthesize the overall centroid coordinates of the lower limb from the relative coordinates of the local centroids. :

[0066] ;

[0067] Step S130: According to the preset axis-shifting synthesis rules, calculate the overall rotational inertia of the lower limb from the segmental mass of the thigh, the segmental mass of the lower leg, the relative coordinates of the local center of mass, the coordinates of the overall center of mass of the lower limb, and the length-to-diameter ratio of the lower limb.

[0068] Among them, the axis shifting synthesis rule refers to the calculation rule based on the parallel axis theorem, which translates the rotational inertia of each segment about its own center of mass to the center of mass of the lower limb as a whole and then superimposes them.

[0069] The necessary process is as follows: The moment of inertia of the thigh about its own center of mass is calculated using the slender rod model, and the same applies to the lower leg:

[0070] , .

[0071] By the parallel axis theorem, the overall rotational inertia of the lower limbs for:

[0072] .

[0073] Step S140: According to the preset trunk and upper limb inertia correction rules, determine the equivalent rotational inertia of the human body from the overall rotational inertia of the lower limbs and the preset trunk and upper limb inertia correction coefficient.

[0074] Among them, the preset trunk and upper limb inertia correction rule refers to the rule that, based on the rotational inertia of the lower limbs, the rotational inertia contribution of the trunk and upper limbs is converted and superimposed into the equivalent rotational inertia of the whole body according to the preset trunk and upper limb inertia correction coefficient.

[0075] The necessary procedures are as follows:

[0076] ;

[0077] in, The equivalent rotational inertia of the human body. The regression coefficients for the rotational inertia of the trunk and upper limbs are based on anthropometric statistics (range 0.12~0.18).

[0078] The derivation and engineering basis of this formula are as follows:

[0079] The moment of inertia of the entire human body about its center of mass cannot be obtained through simple geometric analytical formulas because the mass distribution of the trunk and upper limbs is complex and varies greatly from person to person. This application adopts an engineering regression approach, decomposing the moment of inertia of the entire body into the sum of known quantities of the lower limbs and unknown quantities of the trunk and upper limbs. Based on the statistical analysis of segmental parameters in human biomechanics, the moment of inertia of the entire body... With human body mass ,height Satisfying statistical regression relationship ,in Dimensionless regression coefficients; lower limb rotational inertia The rotational inertia contributions of the torso and upper limbs have been accurately calculated using the parallel axis theorem in step S130, therefore, are expressed as follows: ,in These are the regression coefficients obtained by fitting an anthropometric database; regarding dimensional verification, Dimensionless Dimensions are It has the same dimensions as the moment of inertia.

[0080] Step S150: According to the preset equivalent mass-inertia conversion rules, calculate the equivalent mass and equivalent rotational inertia of the top frame from the material density and geometric dimensions in the top frame structural parameters.

[0081] Among them, the preset equivalent mass-inertia conversion rule refers to the rule that simplifies the top frame into a homogeneous rectangular plate and calculates its mass and moment of inertia about the geometric central axis based on the material density and geometric dimensions.

[0082] The necessary process is as follows: Top frame volume From the length of the top frame ,width and thickness calculate:

[0083] ;

[0084] Equivalent mass of top frame From material density and volume calculate:

[0085] ;

[0086] The equivalent moment of inertia of the top frame is calculated using the formula for the moment of inertia of a homogeneous rectangular plate about a transverse horizontal axis passing through its geometric center and parallel to its width:

[0087] ;

[0088] This formula originates from the integral definition of the moment of inertia of a rigid body. For a homogeneous rectangular plate, establish a coordinate system with the axis of rotation as... The axis (parallel to the width direction) is located on the plate surface. - Plane, mass element The mass element to The square of the distance along the axis is ,right , , integral:

[0089] .

[0090] Step S160: According to the preset diagonal assembly rules, fill the diagonal elements with the total mass of the human body, the equivalent moment of inertia of the human body, the equivalent mass of the top frame, and the equivalent moment of inertia of the top frame in a preset order of degrees of freedom to obtain the mass matrix.

[0091] Among them, the preset degree of freedom sorting refers to the arrangement order of the generalized coordinate vectors of the system. In this application, it is sorted as [vertical translation of the top frame, rotation of the top frame around the horizontal axis, vertical translation of the human body, and rotation of the human body around the horizontal axis]. The preset diagonal assembly rule refers to the rule of filling the diagonal elements of the mass matrix with the mass and moment of inertia of each rigid body according to the preset degree of freedom and setting the off-diagonal elements to zero. In this application, the human-machine inertial coupling effect is reflected in the load distribution matrix and dynamic equations through the contact interface force vector. The mass matrix adopts a block diagonal form to reduce the real-time calculation complexity.

[0092] The necessary process is as follows: Let the system's generalized coordinate vector be... ,in This refers to the vertical translational displacement of the top frame. The rotation angle of the top frame around the horizontal axis. This refers to the vertical translational displacement of the human body. The rotation angle of the human body around the horizontal axis; mass matrix for:

[0093] .

[0094] Furthermore, considering that the plantar pressure distribution directly determines the load input accuracy of the human-machine interface, and that the original pressure signal is subject to sensor noise and boundary support effect interference, the load distribution matrix of the user-top frame contact interface is determined by the pressure distribution according to a preset load mapping rule, including the following steps:

[0095] Step S1A0: Denoise the pressure distribution according to the preset filtering rules to obtain the preprocessed pressure distribution field.

[0096] Among them, the preset filtering rule refers to the rule for spatially smoothing the pressure distribution field to suppress sensing noise.

[0097] The necessary process is as follows: For the original pressure distribution field... Perform two-dimensional Gaussian filtering:

[0098] .

[0099] in, This is the filtered pressure distribution field. It is a two-dimensional Gaussian kernel function. The filter window half-width is set to 2, and the Gaussian kernel standard deviation is set to... Use a grid cell spacing of 1.5; 2D Gaussian kernel function. Separate definition is used: And the kernel function is in the window The sum is normalized to 1 to ensure that the total pressure remains unchanged before and after filtering.

[0100] Step S1B0: Divide the foot pedal area into several preset shape sensing unit grids according to the preset grid discretization rules, determine the coordinates of each grid node and the area of ​​each grid unit, and spatially register the preprocessed pressure distribution field with each grid unit.

[0101] Among them, the preset grid discretization rule refers to the rule of dividing the continuous foot pedal area into rectangular grid cells and establishing the correspondence between pressure sampling points and grid cells.

[0102] The necessary procedures are as follows:

[0103] Divide the foot pedal area evenly along its length into Divide into equal parts, evenly distributed along the width direction. Divide into equal parts, forming A rectangular sensing unit grid; wherein, the number of grid divisions is... , Determine the area of ​​each grid cell according to the preset grid density rules. No larger than the physical area of ​​a single sensing unit in the pressure sensing array, i.e. , ,in , These represent the resolutions in the length and width directions of the pressure sensing array, respectively (i.e., the center-to-center distance between adjacent sensing units). In this embodiment... , , To round up, ensure the grid resolution is not lower than the physical resolution of the sensing unit; Area of ​​each grid cell ,in The grid spacing is in the length direction. The grid spacing is set in the width direction; the preprocessed pressure distribution field is... Samples were taken at the center point of each grid cell to obtain the registered cell center pressure value. In step S1C0, according to the preset pressure integration rules, the registered pressure data and cell area within each grid cell are numerically integrated to obtain the equivalent cell resultant force of each grid cell.

[0104] Among them, the preset pressure integration rule refers to the rule of multiplying the registered unit pressure value by the unit area and integrating the distributed pressure into the unit concentrated resultant force.

[0105] The necessary procedures are as follows:

[0106] .

[0107] in, For the first The equivalent element resultant force of each grid cell.

[0108] In step S1D0, according to the preset node allocation rules, the element resultant force is allocated to each grid node by the equivalent element resultant force and the preset shape function value, so as to obtain the equivalent nodal force of each grid node.

[0109] Among them, the shape function refers to the basis function in the finite element method that describes the interpolation relationship between the distribution of physical quantities inside the element and the values ​​of the element nodes. One implementation of the preset shape function value is the rectangular bilinear shape function value, which takes a value of 0.25 at the origin of the natural coordinate system for all four nodes, indicating that the load at the center of the element is evenly distributed to the four nodes. The preset node allocation rule refers to the rule for distributing the resultant force of the element to each node of the element according to the shape function weight.

[0110] The necessary procedures are as follows:

[0111] For the first A rectangular grid cell, the shape function values ​​of its four nodes. ( (The four nodes of the element are numbered) according to the natural coordinates of the isoparametric element. Calculation, where , These are the dimensionless coordinates along the length and width directions of the element in the natural coordinate system, respectively; the pressure sensor is located at the center of the element (the origin of the natural coordinate system). Equivalent nodal force The total equivalent nodal force of a grid node is obtained by superimposing the forces distributed among all adjacent elements sharing the same node. .

[0112] In step S1E0, according to the preset boundary correction rules, the node distance is calculated from the grid node coordinates and the preset installation position of the movable support 3, and the boundary correction coefficient is determined from the node distance and rotational constraint stiffness. The equivalent node force of the grid nodes that are less than the preset boundary threshold from the geometric boundary of the foot pedal area is corrected for the support boundary effect.

[0113] Among them, the boundary correction coefficient refers to the weighted coefficient that corrects the nodal forces near the boundary by considering the pressure redistribution caused by the boundary support effect of the movable support member. In this application, it is determined by the nodal distance and the rotational constraint stiffness. The support boundary effect refers to the mechanical phenomenon that the pressure distribution near the boundary shifts to the central region due to the elastic rotational constraint provided by the movable support member at the boundary.

[0114] The necessary procedures are as follows:

[0115] Let the length direction of the top frame be... Axis, width direction is Axis, vertical direction is Axis; by the first Calculate the node distance by using the coordinates of each grid node and the preset installation position coordinates of the active support 3. This refers to the minimum distance from the node to the installation position of each movable support 3; the distance from the node to the geometric boundary of the foot pedal area is calculated, only for distances less than a preset boundary threshold. The boundary pressure redistribution coefficient of the selected nodes is calculated and corrected, while the boundary pressure redistribution coefficient of the remaining nodes is set to 1, i.e., no correction is applied. The distance of a node from the geometric boundary of the foot-feeding area is denoted as... The boundary pressure redistribution coefficient is calculated from the grid node coordinates and the geometric boundary coordinates of the foot-feet area; the boundary pressure redistribution coefficient is calculated using the following formula:

[0116] ;

[0117] in, For the first The boundary pressure redistribution coefficient of each node This is the boundary effect intensity coefficient (taken as 0.3). The node distance is the distance from the node to the installation position of the nearest active support 3. For the characteristic length (take) ), Preset reference rotational constraint stiffness (take) ), A preset boundary threshold for determining whether a node is affected by boundary effects; The distance from the node to the geometric boundary of the foothold area has different physical meanings. Used to determine whether a node is affected by boundary effects. Used to quantify the degree of decay of boundary effects.

[0118] The construction logic and derivation basis of this formula are as follows:

[0119] The boundary correction coefficient in this application is not directly derived from the analytical solution of elasticity mechanics, but is a semi-empirical formula based on dimensional analysis and physical intuition. Its design follows the following three principles: First, the distance attenuation principle. According to Saint-Venant's principle, the local stress disturbance caused by the boundary concentration force at the installation position of the movable support 3 attenuates rapidly with the increase of distance from the installation position. In engineering, exponential functions are often used to describe this kind of boundary layer attenuation behavior, so a spatial attenuation term is introduced. ,when When the time decay term is 1, the correction effect is the largest. The time decay term approaches zero, the correction effect disappears, which aligns with the intuitive understanding of boundary layer mechanics; secondly, according to the stiffness modulation principle, the intensity of the boundary effect is directly related to the stiffness of the boundary rotation constraint. The larger the value, the stronger the resistance of the boundary to the rotation of the top frame, the more the plantar pressure shifts towards the central region, and the more significant the force attenuation at the boundary nodes should be. To make this physical relationship dimensionless, a stiffness ratio is introduced. The ratio varies with Monotonically increasing, when When the time approaches 0, The time tends to 1, and the physical behavior is correct; third, according to the principle of bounded linear superposition, the above two items are combined by multiplication and subtracted from 1, and the coefficients... Control the overall correction range to ensure There will be no non-physical results with negative values ​​or values ​​greater than 1.

[0120] The corrected nodal forces are:

[0121] ;

[0122] in, For the first The equivalent nodal force after correction of each grid node.

[0123] Step S1F0: According to the preset load assembly rules, the corrected equivalent nodal forces of each grid node are sorted and assembled into a load distribution matrix according to the preset degrees of freedom.

[0124] Among them, the preset load assembly rule refers to the rule of sorting and merging the forces at each node into an external force vector according to the system's degrees of freedom.

[0125] The necessary procedures are as follows:

[0126] Assume the system has a total of There are several grid nodes; the forces at each node are summed to obtain the resultant vertical force of the top frame; the forces at each node are multiplied by their length arm to the geometric center and summed to obtain the resultant moment about the horizontal axis; load distribution matrix (system external force vector). Assemble according to preset degrees of freedom:

[0127] .

[0128] in, For the first The length direction coordinates of each node Here are the coordinates of the geometric center of the top frame along its length; the first term is the resultant vertical force of the top frame, the second term is the resultant moment of the top frame about its horizontal axis, and the third and fourth terms are the external forces of the human body's degrees of freedom, transmitted by the top frame through the contact interface and set to zero during initial assembly; it should be noted that this application performs linearization analysis on the dynamic components of vibration, and the human body's self-weight, as a static preload, does not participate in the dynamic external force assembly, which conforms to the standard treatment of linearization analysis of small vibrations.

[0129] Furthermore, considering that after the mass matrix and load distribution matrix are determined, the elastic and damping characteristics of the top frame, boundary supports, and human-machine interface need to be uniformly assembled into a solvable dynamic system, a human-machine coupling model is constructed by integrating the mass matrix, load distribution matrix, top frame structural parameters, and rotational constraint stiffness, including the following steps:

[0130] Step S1a0: Construct the stiffness matrix of the top frame according to the preset top frame discretization rules, using the elastic modulus, moment of inertia of the cross section, and geometric dimensions.

[0131] Among them, the preset top frame discretization rule refers to the rule of simplifying the top frame into a finite element model and discretizing it; in this application, the top frame mainly exhibits rigid body motion characteristics in the vertical vibration frequency band, and the contribution of the top frame's own elastic deformation to the low-frequency dynamic response of the system is negligible, and the top frame stiffness matrix is... It degenerates into a zero matrix in the global stiffness matrix; the elastic recovery effect of the top frame is determined by the boundary support matrix. and stiffness matrix of human body-top frame contact interface This is provided to ensure that the overall stiffness matrix and mass matrix have the same dimension and that the system equations are closed.

[0132] The necessary procedures are as follows:

[0133] In this application, the elastic deformation of the top frame within the target vibration frequency band (5Hz to 50Hz) is much smaller than that of the boundary supports and the human-machine interface. Therefore, the top frame is treated as a rigid body, and its stiffness matrix is... The vertical translation and rotation about the horizontal axis of the top frame are elastically restoring due to boundary supports and contact stiffness; the rigid body assumption is verified by the following offline check: if the check yields... (That is, more than 5 times the upper limit of the target frequency band, in this embodiment) If this is true, then the rigid body assumption holds within the target frequency band, allowing the use of... It degenerates into a zero matrix.

[0134] Step S1b0: According to the preset boundary support rules, construct the boundary support matrix based on the rotational constraint stiffness and the preset installation position of the movable support 3. The boundary support matrix includes the elastic constraint terms for rotational degrees of freedom and the rigid constraint terms for translational degrees of freedom. According to the preset human-top frame contact rules, construct the stiffness matrix of the human-top frame contact interface based on the vertical stiffness and rotational coupling stiffness of the human-top frame contact interface.

[0135] The necessary process is as follows: Boundary support matrix This stems from the local elastic constraint at the rotating connection points at both ends of the movable support member 3. There are two movable support members 3, symmetrically installed at both ends of the bottom surface of the top frame. The distance from each installation position to the geometric center (centroid) of the top frame in the longitudinal direction is... In the local coordinate system of the nodes, each support point ( It has vertical translational degrees of freedom. and rotational degrees of freedom about the horizontal axis The corresponding local diagonal stiffness matrix is ,in The vertical approximate rigid constraint stiffness of the movable support component. This refers to the rotational constraint stiffness of the elastic element at the rotational connection.

[0136] Node coordinates and generalized coordinates of the center of mass of the rigid body of the top frame The transformation relationship is as follows:

[0137] .

[0138] .

[0139] According to the preset coordinate transformation rules, the equivalent matrix of the boundary support matrix in the generalized coordinates of the top frame is: Expanding on this, we get:

[0140] .

[0141] Among them, the non-diagonal elements cancel each other out because the two movable supports are symmetrically arranged and their installation positions are symmetrical about the centroid. The natural frequency of the boundary mode corresponding to the vertical translational degree of freedom of the top frame is much higher than the target vibration frequency band (about tens of kHz). In the subsequent mode truncation, it is discarded as a high-frequency boundary mode, which does not affect the dynamic response in the 5Hz to 50Hz frequency band. This is the equivalent boundary constraint stiffness for the top frame rotating about the transverse horizontal axis. The boundary support matrix is ​​obtained by expanding the matrix to a fourth-order matrix with the same dimension as the overall stiffness matrix according to the preset degree-of-freedom alignment rules (with zeros padded at the corresponding positions of the human body degrees of freedom). Stiffness matrix of human body-top frame contact interface In the system's generalized coordinates The non-zero elements below are:

[0142] ;

[0143] ;

[0144] All other elements are zero; this matrix realizes the elastic coupling between the top frame and the human body's degrees of freedom. Its construction is based on the stiffness assembly rules of the nodes at both ends of the contact spring. The contact spring force is proportional to the relative displacement at both ends, so the diagonal elements are positive and the cross elements are negative.

[0145] In step S1c0, according to the preset degree of freedom alignment rules, the overall stiffness matrix is ​​constructed from the top frame stiffness matrix and the boundary support matrix; according to the preset Rayleigh damping rules, the damping matrix is ​​determined from the mass matrix, the overall stiffness matrix and the preset damping ratio; and the mass-stiffness-damping system matrix of the human-machine coupling model is constructed from the mass matrix, the overall stiffness matrix and the damping matrix.

[0146] Among them, the preset degree-of-freedom alignment rule refers to the rule of superimposing and assembling the elements corresponding to the generalized coordinates in the boundary support matrix and the contact interface stiffness matrix according to the same degree-of-freedom numbering system. One implementation method is to first expand each sub-matrix to a fourth-order matrix with the same dimension as the overall stiffness matrix, and then add the corresponding degree-of-freedom numbers element by element after placing them in the same row and column positions. The preset Rayleigh damping rule refers to the engineering damping model that represents the viscous damping matrix as a linear combination of the mass matrix and the stiffness matrix. The full-band damping characteristics can be determined by calibrating the damping ratio of a few modes.

[0147] The necessary procedures are as follows:

[0148] The overall stiffness matrix is ​​assembled by aligning and superimposing the top frame stiffness matrix and the boundary support matrix according to their degrees of freedom. In one embodiment, the stiffness matrix of the human-top frame contact interface is incorporated as an equivalent parallel component of the rotational constraint stiffness under the user's standing condition, and the overall stiffness matrix is ​​also included. ,because Therefore, in this implementation method All three are in the same generalized coordinate system. The elements below are superimposed, and the dimensions of each element are consistent with the corresponding generalized coordinates (the dimensions of the translational term are...). The dimensions of the rotation-rotation term are: The system satisfies the stiffness matrix assembly rules for multi-degree-of-freedom systems. According to the Rayleigh damping rule, the damping matrix is... Rayleigh damping coefficient is based on the premise that the damping ratios of the first two modes are equal to the preset modal damping ratio. Assuming a value of 0.05, the first two dominant natural angular frequencies of the system... Sure:

[0149] ;

[0150] in, and The Rayleigh damping coefficient is... The preset modal damping ratio is used to characterize the rate of decay of vibration of a given modality. The modal damping ratio is the ratio of the actual damping of each mode to the critical damping.

[0151] The derivation of this formula is as follows: For the first... 1st-order mode, left multiplication Right multiplication Utilizing modal orthogonality and The modal damping term is obtained as follows:

[0152] ;

[0153] Modal damping ratio is defined as Therefore:

[0154] ;

[0155] If the damping ratio of the first two modes is required to be equal to the preset value... Then we establish a system of two linear equations in two variables:

[0156] ;

[0157] Multiply the first equation by Multiply by the second expression ,have to and Subtracting the two equations cancels out the expression. :

[0158] .

[0159] like Both sides made an appointment. ,have to ;

[0160] Substitution First Form: .

[0161] Step S1d0: According to the preset external force mapping rules, the load distribution matrix is ​​mapped to the system external force vector. The mass matrix, damping matrix and overall stiffness matrix are used as the coefficient matrix of the dynamic equation, and the system external force vector is used as the excitation term to construct the human-machine coupled dynamic equation and obtain the human-machine coupled model.

[0162] Among them, the preset external force mapping rule refers to the rule of directly assigning the concentrated forces and moments in the load distribution matrix to the excitation terms on the right side of the dynamic equation according to the degree of freedom. One implementation method is that the load distribution matrix itself is assembled according to the generalized coordinates, so it is directly substituted into the right side of the dynamic equation as the external force vector of the system.

[0163] The necessary procedures are as follows:

[0164] The human-machine coupling dynamics equation is: ;

[0165] in, , , These are the aforementioned mass matrix, damping matrix, and overall stiffness matrix, respectively. , They are generalized coordinate vectors The first derivative with respect to time (generalized velocity vector) and the second derivative with respect to time (generalized acceleration vector). The system external force vector is mapped from the load distribution matrix. It is a time variable.

[0166] Furthermore, considering that after the human-machine coupling model is established, it is necessary to obtain the quantitative transfer characteristics from the excitation point to the response point in the foot pedal area in order to provide the basis for inverse calculation for subsequent feedforward compensation, the multi-source transfer operator containing direct channels and cross-coupled channels is solved based on the human-machine coupling model according to the preset transfer operator construction rules, including the following steps:

[0167] Step S210: The mass matrix, damping matrix and overall stiffness matrix are read from the human-machine coupling model as coefficient matrices, and the excitation degree of freedom number and response degree of freedom number corresponding to the installation position of the vibration mechanism 4 and the foot pedal area are determined according to the preset excitation response mapping rules.

[0168] Among them, the preset excitation response mapping rule refers to the rule that establishes the correspondence between the physical installation location and the finite element degree of freedom number.

[0169] The necessary process is as follows: A mapping table of installation position coordinates and degree of freedom numbers is pre-stored in the memory. The mapping table is generated according to the following preset mapping construction rules: The geometric dimensions in the top frame structure parameters are spatially registered with the node coordinates of the discrete grid of the foot pedal area. A correspondence between the spatial coordinates and the nearest grid node index is established for each physical component (installation position of vibration mechanism 4, response point of foot pedal area, installation position of movable support 3). The nearest grid node index is determined by the Euclidean distance minimization criterion, that is, the spatial coordinates of the component are... Traverse the set of coordinates of all grid nodes , take Minimum node index As the mapping node for this component; according to the preset excitation response mapping rules, the vertical translational degree of freedom number is retrieved from the installation position mapping node index of vibration mechanism 4. and the numbering of the degrees of freedom of rotation about the horizontal axis As the excitation degree of freedom number, its vertical translational degree of freedom number is retrieved from the index of the response point mapping node in the foot pedal area and used as the response degree of freedom number; in this embodiment, the vibration mechanism 4 is installed at the center of the bottom surface of the top frame, and the corresponding excitation degree of freedom number is the vertical translational degree of freedom number of the top frame. and the numbering of the degrees of freedom of rotation of the top frame about the transverse horizontal axis The response points in the foot pedal area are taken at three locations: both ends and the midpoint of the upper surface of the top frame. The corresponding degrees of freedom are numbered as follows: (Left end vertical) (Midpoint perpendicular) (Vertical direction at the right end).

[0170] Step S220: According to the preset modal truncation rule, the dominant natural frequencies and corresponding mode shapes within the preset order range are solved by the coefficient matrix.

[0171] Among them, the preset mode truncation rule refers to the rule of retaining only the low-order modes that dominate the dynamic response of the target frequency band and discarding the high-order modes.

[0172] The necessary process is as follows: Solving the generalized eigenvalue problem The natural angular frequency is obtained. and normalized mode shapes (satisfy The modal effective mass ratio (EMR) refers to the proportion of the effective mass of each mode in a specified motion direction to the total mass of the system in that direction. It is used to measure the contribution of each mode to the dynamic response in that direction. The EMR is accumulated sequentially in ascending order of natural frequency. The process stops when the accumulated EMR in all excitation degrees of freedom within the cutoff frequency range is greater than a preset threshold (0.95), and the retained order is recorded as . This ensures that the mode set has dynamic completeness within the target frequency band; wherein, the "direction of each excitation degree of freedom" includes the vertical translational direction of the top frame. The direction of rotation of the top frame around the horizontal axis Two directions; the first First mode in The modal effective mass ratio in the direction is ,exist The modal effective mass ratio in the direction is ,in For the first Concentrated mass with one degree of freedom For this degree of freedom coordinate, The coordinates of the top frame's geometric center; modal truncation preserves the order. Take and The smallest value.

[0173] Step S230: According to the preset direct channel construction rules, construct the direct transmission channel from each excitation degree of freedom to the corresponding response degree of freedom by constructing the dominant natural frequency, the corresponding mode shape and the preset frequency discrete axis.

[0174] The necessary process is as follows: On the discrete frequency axis (covering 5Hz to 50Hz, taking 200 logarithmically uniformly distributed frequency points), directly calculate the channel frequency response function. for:

[0175] ;

[0176] in, To incentivize the degree of freedom, The modal order is... The imaginary unit, To excite the angular frequency, For the first The dominant natural angular frequency, For the first The first mode shape is in the 1st order. The component at each degree of freedom of the incentive For the first The first modal damping ratio; this formula is the standard frequency response expression of the mass-normalized modal superposition method, with the denominator having the dimension of the square of the angular frequency, and the numerator having the dimension of the reciprocal of the mass after mass normalization, and its standard compliance dimension is... In this application, the excitation is applied in the form of a specified displacement of the excitation degree of freedom, and the specified displacement is applied through the equivalent stiffness of the excitation channel. Converted to equivalent excitation force, this equivalent stiffness is incorporated into the standard frequency response during offline calibration of the multi-source transfer operator. Therefore, the dimensions of each element of the multi-source transfer operator are the ratio of displacement to displacement (dimensionless), and the physical meaning is the ratio of steady-state displacement response of the response degrees of freedom under a unit excitation displacement input, directly establishing the mapping relationship from excitation displacement to response displacement.

[0177] Step S240: According to the preset cross-coupling construction rules, the cross-coupling gain is determined by the boundary rotational degrees of freedom corresponding to the finite value of the rotational constraint stiffness and the modal participation factor of the mode shape on the boundary rotational degrees of freedom. The cross-coupling gain, the dominant natural frequency, the corresponding mode shape and the frequency discrete axis are used to construct the cross-coupling channels from each excitation degree of freedom to the non-corresponding response degree of freedom. According to the preset transfer operator assembly rules, the direct channel and the cross-coupling channel are assembled into a multi-source transfer operator according to the frequency discrete axis, the excitation degree of freedom number and the response degree of freedom number.

[0178] Among them, the cross-coupling gain refers to the dimensionless equivalent coefficient of the excitation-response off-diagonal coupling strength caused by the limited stiffness of the boundary rotation constraint. In this application, it is used to correct the contribution of the truncated mode to the cross-coupling term in the modal superposition method. The preset transfer operator assembly rule refers to the rule of adding the direct channel matrix and the cross-coupling channel matrix element by element and storing them as a complex matrix point by point according to the frequency discrete axis. One implementation is that at each discrete frequency point, the diagonal block position is filled with the direct channel frequency response value and the off-diagonal block position is filled with the cross-coupling channel frequency response value. The specific determination process of the cross-coupling gain is referred to steps S241 to S245.

[0179] The necessary process is as follows: Assume the system has Each degree of freedom of incentive and One response degree of freedom; multi-source transit operator for 3D complex matrix:

[0180] ;

[0181] in, It is a direct channel matrix (diagonal block dominant). This is a cross-coupled channel matrix (off-diagonal block); matrix elements Indicates from the first The incentive degrees of freedom to the first The frequency response function of each degree of freedom is determined by the direct path formula in step S230 and the cross-coupling gain in steps S241 to S245.

[0182] Furthermore, considering that modal truncation will lose the contribution of higher-order modes to boundary coupling, and that rotational constraint stiffness has manufacturing tolerances and material discreteness, failure to correct this will lead to inaccurate estimation of cross-coupling strength. Therefore, the cross-coupling gain is calibrated and perturbation corrected, including the following steps:

[0183] Step S241: The movable support 3 is installed at a preset position, and the boundary rotational degree of freedom number is determined according to the preset boundary degree of freedom identification rules.

[0184] Among them, the boundary degree of freedom identification rule refers to the rule of locating the corresponding rotational degree of freedom index in the finite element node degree of freedom according to the physical installation position of the active support 3 on the top frame.

[0185] The necessary process is as follows: Search for the node closest to the installation coordinates in the set of grid node coordinates determined in step S1B0, and include the rotational degree of freedom number of this node about the horizontal axis into the boundary rotational degree of freedom set; In this embodiment, the movable support is installed at two connection positions on the bottom surface of the top frame, and the rotational degree of freedom number of the first connection node is... The rotational degree of freedom of the second connecting node is numbered as follows: The set of boundary rotation degrees of freedom is as follows .

[0186] Step S242: According to the preset modal participation factor extraction rules, extract the modal participation factors of each dominant mode on the boundary rotation degree of freedom based on the corresponding mode shape and boundary rotation degree of freedom number.

[0187] The modal participation factor refers to the component value of the mode shape at a specific degree of freedom. In this application, it specifically refers to the modal shape component at the boundary rotational degree of freedom, which is used to characterize the degree of motion participation of the mode in the boundary rotational direction. The preset modal participation factor extraction rule refers to the rule of directly reading the component value of the normalized mode shape vector at the specified degree of freedom number.

[0188] The necessary process is as follows: For the first The dominant mode, with its rotational degrees of freedom at the boundary. The modal participation factor is:

[0189] ;

[0190] in, For the first The first mode in the 1st order Modal participation factor on each boundary rotational degree of freedom For the first The mode shape vectors of the first order are numbered in the degree of freedom. The component at that location.

[0191] Step S243: According to the preset residual flexibility correction rule, calculate the truncated mode residual flexibility matrix from the overall stiffness matrix and the corresponding mode shape, and calculate the equivalent boundary stiffness correction coefficient on the boundary rotational degrees of freedom from the residual flexibility matrix and the rotational constraint stiffness.

[0192] Among them, residual compliance refers to the contribution of higher-order modes not included due to modal truncation to the low-frequency compliance of the system, which is calculated in this application by the difference between the inverse of the overall stiffness matrix and the retained modal compliance; the equivalent boundary stiffness correction coefficient refers to the equivalent elastic support stiffness at the boundary rotational degrees of freedom after considering residual compliance, which is used to correct the coupling effect caused by finite rotational constraints.

[0193] The necessary steps are as follows: truncate the modal residual compliance matrix. for:

[0194] ;

[0195] The derivation of this formula is based on the following: the complete compliance matrix of the system (static compliance) is... According to the modal superposition method, the compliance matrix can also be expressed as the sum of the compliance of all modes. ,in For the total modal order of the system's generalized eigenvalue problem; when only the first... When the higher-order modes are dominant, the compliance contribution of truncating the higher-order modes is the residual compliance. This reflects the residual influence of neglected high-frequency modes on the low-frequency response; the equivalent boundary stiffness correction coefficient on the boundary rotational degrees of freedom. The residual compliance matrix is ​​the reciprocal of the diagonal elements of the boundary rotation degrees of freedom.

[0196] .

[0197] in, is the diagonal element of the residual compliance matrix at the boundary rotation degrees of freedom.

[0198] Step S244: Calculate the nominal cross-coupling gain according to the preset nominal gain calculation rules, using the modal participation factor, equivalent boundary stiffness correction coefficient, and rotational constraint stiffness.

[0199] The formula is constructed and derived as follows: when the boundary rotation constraint stiffness When the value is finite, the boundary rotational degrees of freedom are not completely fixed, resulting in additional coupling between the excitation and response degrees of freedom through boundary rotation. The strength of this coupling is related to the rotational components of the mode shapes at the boundary and the ratio of the boundary support stiffness to the system's equivalent stiffness. In this application, the nominal cross-coupling gain is defined as a weighted combination of boundary modal participation factors, with the weights determined by the boundary stiffness ratio.

[0200] ;

[0201] in, The nominal cross-coupling gain is dimensionless. Normalized modal participation factor (dimensionless):

[0202] ;

[0203] in, To iterate through the summation index of all dominant modes, For the first The modal participation factor of the first mode on the boundary rotational degrees of freedom (when there are multiple boundary rotational degrees of freedom, the sum of the absolute values ​​of the modal participation factors of each boundary rotational degree of freedom).

[0204] The physical analysis is as follows: The dimensionless property is eliminated by normalization. It is a ratio of dimensionless stiffness, and it varies with... Increase monotonically, when When the time approaches 0, When the value approaches 1, it physically indicates that the weaker the boundary constraint, the weaker the cross-coupling effect, and the stronger the boundary constraint, the more significant the cross-coupling. This gain is used to correct the frequency response amplitude of the cross-coupling channel, making the cross-coupling estimation under the truncated mode closer to the behavior of the complete system.

[0205] Step S245: According to the preset perturbation correction rule, calculate the sensitivity coefficient of the cross-coupling gain to the rotational constraint stiffness based on the nominal cross-coupling gain, rotational constraint stiffness, and preset stiffness perturbation range. Determine the perturbation-corrected cross-coupling gain based on the sensitivity coefficient and the preset stiffness perturbation reference value.

[0206] Among them, the sensitivity coefficient refers to the partial derivative of the cross-coupling gain with respect to the rotational constraint stiffness parameter, reflecting the sensitivity of the gain to changes in this parameter; the stiffness perturbation range refers to the possible fluctuation range of the rotational constraint stiffness after considering manufacturing tolerances and material dispersion, which is taken as ±15% in this application; the stiffness perturbation reference value refers to the rotational constraint stiffness perturbation amount used for sensitivity analysis, which is taken as 10% of the nominal value.

[0207] The necessary process is as follows: the sensitivity coefficient of cross-coupling gain to rotational constraint stiffness. for:

[0208] ;

[0209] This equation can be derived using the chain rule. Differentiation yields, i.e. Preset stiffness perturbation reference value The cross-coupling gain after perturbation correction is corrected using a first-order Taylor expansion.

[0210] ;

[0211] in, This is the cross-coupling gain after perturbation correction; this correction covers the most unfavorable coupling conditions within the manufacturing tolerance band with a +10% unidirectional perturbation, and is a conservative engineering correction; the final corrected elements of the cross-coupling channel frequency response matrix are:

[0212] ;

[0213] in, To incentivize freedom To the response degrees of freedom The frequency response function of the cross-coupled channel. For the first Mode shape in response degrees of freedom The component at that location.

[0214] Furthermore, considering that the multi-source transfer operator already contains the complete frequency response characteristics of the system, the desired response can be inversely calculated before vibration output to pre-compensate the system transfer error. Therefore, based on the multi-source transfer operator, feedforward compensation is performed on the desired vibration parameters to obtain the theoretical excitation displacement sequence, including the following steps:

[0215] Step S310: According to the preset parameter receiving rules, receive the vibration frequency, vibration amplitude and vibration mode identifier input by the user, and perform protocol parsing, validity verification and anti-misclick filtering to obtain the desired vibration parameter set to be processed.

[0216] Among them, the preset parameter receiving rules refer to the receiving processing rules for parsing, verifying and confirming the original input instructions.

[0217] The necessary process is as follows: The system receives the original command frame through the human-machine interface and parses it according to the preset communication protocol to obtain the vibration frequency. Vibration amplitude Vibration mode identifier Perform a validity check; if or If the input fails, the input is rejected and a prompt for re-entry is displayed. The above range is consistent with the typical operating conditions of a whole-body vibration training device (frequency 15-50Hz, amplitude 2-4mm) and includes a margin. Anti-mistouch filtering is implemented, and the input signal is confirmed with a 500ms delay. If the change in the input value during the confirmation period is less than a preset threshold, it is considered a valid input, and the desired vibration parameter set is obtained. The "preset threshold" is the allowable change amount for determining input stability. In this embodiment, it is taken as the change amount of vibration frequency among the desired vibration parameters being less than... The change in vibration amplitude is less than Furthermore, the vibration mode identifier does not change; the input is considered valid only if all three conditions are met simultaneously within the entire 500ms delay confirmation window.

[0218] Step S320: According to the preset parameter analysis rules, determine the vibration mode type and the preset weight of each vibration direction component from the vibration mode identifier in the expected vibration parameter set.

[0219] The necessary process is as follows: Query the vibration mode identifier mapping table pre-stored in memory, and select the whole-body vertical vibration mode ( ), oscillation mode ( This is achieved through differential phase excitation of the vibration mechanism, utilizing the rotation of the top frame around the transverse horizontal axis to generate an equivalent horizontal displacement component at the edge of the foot pedal area, thus achieving a swinging effect and a torsional shaping mode. (Achieved through single-end synchronous stimulus), custom combination mode ( );in Preset weights for the vertical direction. To reverse the direction to the preset weight.

[0220] Step S330: According to the preset expected response mapping rules, the expected vibration response is determined by the preset weights of the vibration frequency, vibration amplitude, vibration mode type and each vibration direction component in the expected vibration parameter set.

[0221] The necessary procedures are as follows:

[0222] The active degrees of freedom of the top frame in the human-machine coupled model only include vertical translation. Rotation about the horizontal axis It has no independent horizontal translational degree of freedom; the horizontal vibration component rotates about the transverse horizontal axis through the top frame. The equivalent horizontal displacement component generated at the edge of the foot pedal area is realized, therefore both the oscillating mode and the torsional shaping mode are mapped to... Degrees of freedom are distinguished by preset weights and excitation phase configurations to differentiate vibration modes; the vertical component is mapped to the vertical translational degrees of freedom of the top frame. The torsional component is mapped to the rotational degree of freedom of the top frame about the transverse horizontal axis. The expected response of the human body's degrees of freedom is determined by a weighted average of the corresponding degrees of freedom responses of the overhead frame transmitted through the human-overhead frame contact interface; the expected vibration response is:

[0223] ;

[0224] in, The angular frequency of vibration The expected vibration response vector at that location, The angular frequency of vibration, For vibration amplitude, Let be the unit basis vector for the vertical translational degree of freedom of the top frame. Let be the unit basis vector for the rotational degree of freedom of the top frame about the transverse horizontal axis; To predetermine the equivalent swing arm, the distance from the geometric center of the top frame to the edge of the footrest area is taken in the longitudinal direction. This distance is used to convert the desired linear displacement amplitude in the torsional direction into the desired angular amplitude (in radians) about the transverse horizontal axis, ensuring that the dimensions of the vertical and directional components in the desired response vector are consistent with their corresponding degrees of freedom; the swing mode about the axis corresponds to... Shared with torsional shaping mode The degree of freedom is achieved through differential phase control of the two vibration mechanisms 4 in the embodiment with two vibration mechanisms 4, and through single-end amplitude modulation in the embodiment with a single vibration mechanism 4, thereby realizing differentiated vibration modes.

[0225] Step S340: According to the preset frequency response value rules, obtain the corresponding frequency response matrix from the vibration frequency and the multi-source transmission operator.

[0226] The necessary steps are as follows: Calculate the angular frequency from the vibration frequency. The closest location is found on the discrete frequency axis of the multi-source transfer operator. Extract the corresponding frequency response matrix from the frequency points. Furthermore, linear interpolation is employed to improve the accuracy of the values.

[0227] Step S350: According to the preset feedforward compensation rule, the frequency response inverse operation is performed from the frequency response matrix and the desired vibration response to obtain the theoretical frequency domain excitation displacement.

[0228] The necessary process is as follows: theoretical frequency domain excitation displacement Calculated from the pseudo-inverse of the frequency response matrix:

[0229] ;

[0230] in, This is the theoretical frequency domain excitation displacement vector. This is the Moore-Penrose pseudoinverse; in this application, the response degrees of freedom (3 response points) are greater than the excitation degrees of freedom, and the frequency response matrix is ​​a full-rank non-square matrix. This pseudoinverse gives the optimal solution in the least squares sense, i.e. This ensures that the vibration response of the foot pedal area is as close as possible to the desired response with the existing actuator configuration.

[0231] Step S360: According to the preset time-domain conversion rules, the amplitude and phase parameters, vibration frequency and preset sampling period of the theoretical frequency-domain excitation displacement are converted into the theoretical excitation displacement sequence.

[0232] The preset sampling period refers to the time interval for generating the time-domain excitation sequence. One implementation method is to take one-tenth of the vibration period, that is, each vibration period corresponds to 10 sampling points.

[0233] The necessary process is as follows: From Extract the amplitude of each excitation degree of freedom. and phase angle According to the preset sampling period Generate the theoretical excitation displacement sequence:

[0234] ;

[0235] in, For the first The incentive degrees of freedom in the first... The theoretical excitation displacement at each sampling time, For the first Frequency domain displacement amplitude of each excited degree of freedom For the first Frequency domain displacement phase angle of each excited degree of freedom , This represents the number of sampling points within a single vibration cycle; 10 sampling points per cycle is far higher than the lower limit of twice the bandwidth required by the Nyquist sampling theorem, which can completely preserve the amplitude and phase information of the sinusoidal waveform.

[0236] Furthermore, considering that feedforward compensation is based solely on frequency domain inverse operations, and that the actual drive still suffers from amplitude-frequency attenuation and phase-frequency lag in the direct channel, as well as amplitude inconsistencies at each response point caused by cross-coupling, the theoretical excitation displacement sequence is corrected for amplitude and phase, and the vibration field consistency is optimized before being converted into a drive control signal to drive the vibration mechanism 4. This process includes the following steps:

[0237] Step S410: According to the preset time-domain amplitude and phase extraction rules, extract the time-domain amplitude and phase parameters of each excitation degree of freedom from the theoretical excitation displacement sequence.

[0238] The necessary process is as follows: For the theoretical excitation displacement sequence, the first... Discrete-time series of excitation degrees of freedom The amplitude of the fundamental frequency component is extracted using discrete Fourier transform. and initial phase :

[0239] ;

[0240] ;

[0241] in, For the first The time-domain amplitude of each excitation degree of freedom, For the first The initial phase angle of each excited degree of freedom, This represents the number of sampling points per cycle.

[0242] The derivation of this formula under the condition of integer period sampling is as follows: Assume that the discrete time series is a single-frequency sinusoidal signal. ,in Let be the amplitude of the sinusoidal signal. This represents the initial phase of the sinusoidal signal. Let be the sampling point number; express it in complex exponential form using Euler's formula:

[0243] ;

[0244] Perform on the sequence Point-based Discrete Fourier Transform, fundamental frequency component (the first frequency component, denoted as...) )for:

[0245] ;

[0246] The first summation is... The second term is a geometric series, and its common ratio is denoted as . ,when And when it is an integer The sum of the series is ;therefore:

[0247] ;

[0248] Take the modulus Therefore, the fundamental frequency amplitude Regarding the phase angle, by Complex factors Introduction Due to the fixed phase shift, the initial phase of the sinusoidal excitation must be compensated based on the DFT argument. The fixed offset, i.e. To extract the phase and The definition of the sine function is strictly corresponding; if this offset is not compensated, all excitation degrees of freedom will have a systematic phase error of 90°, causing the phase reference for subsequent amplitude and phase compensation to fail; the above DFT amplitude estimation holds under the condition of integer period sampling, i.e., sampling period With vibration frequency satisfy ( (where the integer is an integer); if there are non-integer period samples, the time-domain sequence is weighted using a Hanning window before performing the DFT to suppress spectral leakage.

[0249] Step S420: According to the preset frequency-time compensation rules, the time-domain amplitude-phase compensation coefficient is determined by the amplitude-frequency characteristics and phase-frequency characteristics of the direct channel in the multi-source transfer operator at the vibration frequency; according to the preset amplitude-phase compensation rules, the amplitude gain and phase lead / lag compensation of the time-domain amplitude-phase parameters are adjusted by the time-domain amplitude-phase compensation coefficient to offset the amplitude-frequency attenuation and phase-frequency lag of the direct channel in the multi-source transfer operator, so as to obtain the amplitude-phase corrected excitation sequence.

[0250] The compensation is applied to the dominant transmission channel, while residual coupling is suppressed by subsequent consistency optimization steps.

[0251] The necessary process is as follows: The direct channel is extracted by the multi-source transfer operator. Frequency response function at point Amplitude compensation gain Phase compensation angle ; Corrected version:

[0252] ;

[0253] ;

[0254] in, For the first The amplitude after correction for each degree of freedom of the excitation For the first Phase angle after correction for each degree of freedom of excitation For the first The incentive degrees of freedom in the first... The amplitude and phase correction excitation displacement at each sampling time.

[0255] Step S430: According to the preset vibration field consistency optimization rules, with the goal of minimizing the amplitude deviation between the vibration responses of each degree of freedom in the foot-step region predicted by the multi-source transfer operator, the amplitude and phase correction excitation sequence is uniformly optimized and allocated to obtain the optimized excitation sequence.

[0256] Wherein, regularization coefficient The regularization parameter refers to the trade-off between the consistency objective and the excitation energy constraint. In this application, it is set to 0.01 to suppress excitation amplitude overshoot caused by ill-conditioned frequency response matrix. The selection is determined according to a preset regularization rule. One implementation of this preset regularization rule is the L-curve method, which calculates different... Consistency optimization objective function residual norm under different values with solution norm Plot the L-curve in a log-log coordinate system, and take the point corresponding to the maximum curvature at the inflection point of the curve. As the optimal value; the preferred implementation method is based on the empirical formula for the reciprocal of the condition number, taking... ,in For matrix condition number, ensure The introduction of The condition number drops below a preset threshold; in this embodiment, it is determined by the above rules. .

[0257] The necessary process is as follows: Correct the complex amplitude value for each excitation degree of freedom. (No. The amplitude of each component is Phase is Let the complex representation of be the variable to be determined, and establish the regularized least squares objective function:

[0258] ;

[0259] in, To optimize the objective function for consistency, For multi-source transfer operators at the oscillation angular frequency Frequency response submatrix (dimension) corresponding to the response degrees of freedom ), Let be the complex amplitude vector of the excitation to be optimized.

[0260] The construction basis and solution process of the objective function are as follows: First term The deviation between the predicted and expected responses is measured, and its minimization directly corresponds to the goal of consistent vibration amplitude across all degrees of freedom in the foot-feeding region; the second term For Tikhonov regularization, when When the gain of the cross-coupling channel is small and close to ill-conditioned, the simple least squares solution exhibits a drastic amplification of the excitation amplitude. The regularization term ensures the numerical stability of the solution by penalizing the energy of the solution; about Find the gradient and set it to zero:

[0261] ;

[0262] The analytical solution for optimizing the complex amplitude of the excitation is obtained as follows:

[0263] ;

[0264] Among them, superscript This represents the conjugate transpose of a complex matrix. for identity matrix; due to For positive semi-definite matrices, superimposed The latter is strictly positive definite, the inverse matrix always exists, the solution is unique and numerically stable; by Extract the optimized amplitude of each excitation degree of freedom and phase Generate an optimized stimulus sequence:

[0265] ;

[0266] in, For the first The incentive degrees of freedom in the first... Optimized excitation displacement at each sampling time. for The Each component.

[0267] Step S440: According to the preset drive conversion rules, the optimized excitation sequence is converted into the corresponding voltage drive signal or PWM duty cycle signal according to the vibration mechanism type 4 to obtain the drive control signal.

[0268] The necessary process is as follows: When vibration mechanism 4 is a horizontal vibration component, a vertical vibration component, or a dual-head vibration motor, the drive conversion rule is to calculate the target velocity of each excitation degree of freedom from the optimized excitation displacement sequence:

[0269] ;

[0270] in, For the first The target vibration velocity of each excitation degree of freedom; according to the preset electromechanical conversion coefficient of vibration mechanism 4. Map the target velocity to a voltage drive signal:

[0271] ;

[0272] in, For the first Each excitation degree of freedom corresponds to the voltage drive signal of the vibration mechanism 4; The preset electromechanical conversion coefficient (unit: mm / s / V) refers to the vibration velocity generated by the vibration mechanism 4 under a unit driving voltage, which is determined by the motor's factory calibration or offline identification; preset electromechanical conversion coefficient The calibration rule is as follows: When the vibration mechanism 4 is under no-load condition, an amplitude of [value] is applied. , frequency is The sweep frequency sinusoidal voltage signal was used to record the output velocity response amplitude of the vibration mechanism 4. Take each frequency point Then at the target vibration frequency Linear interpolation is performed at the point to obtain Alternatively, it can be simplified to directly using the slope of the rated speed-voltage characteristic curve provided by the motor manufacturer. ,in This refers to the rated angular velocity of the motor (in rad / s). The eccentric radius (in mm) of the eccentric block 4 of the vibration mechanism. The rated drive voltage is used; this embodiment adopts the latter method; when the vibration mechanism 4 adopts PWM speed regulation, the voltage drive signal is further adjusted to the rated voltage. Normalize and limit the amplitude to obtain the PWM duty cycle signal:

[0273] ;

[0274] in, This is the PWM duty cycle signal. As a limiting function, the duty cycle is constrained to Range to prevent overdrive.

[0275] Step S450: The vibration mechanism 4 is driven by the drive control signal according to the preset output drive rules.

[0276] The necessary process is as follows: the drive control signal of each vibration mechanism 4 is amplified by the power drive module and output to the motor winding of the corresponding vibration mechanism 4, so that each vibration mechanism 4 generates controlled mechanical vibration according to the optimized excitation sequence, which is transmitted to the foot pedal area through the top frame to form a vibration field consistent with the desired vibration parameters.

[0277] Furthermore, considering that changes in user posture and weight distribution in actual use will introduce unpredictable load disturbances to the feedforward system, closed-loop load compensation is performed on the drive control signal based on the motor operating current and speed during vibration operation until the user-set running time is reached or a stop command is received. This includes the following steps:

[0278] Step S510: According to the preset load observation rules, the actual load torque applied to the vibration mechanism 4 is estimated in real time by combining the motor torque constant and back electromotive force constant in the preset electromagnetic parameters of the motor with the operating current and speed of the motor of the vibration mechanism 4.

[0279] Among them, the preset electromagnetic parameters of the motor refer to the electromagnetic characteristic parameters determined by the motor factory calibration or offline identification, which in this application include the motor torque constant. (Unit: N·m / A) and back electromotive force constant (Unit: V·s / rad).

[0280] The necessary process is as follows: During vibration operation, the system uses a preset sampling period. Real-time acquisition of motor operating current and speed The electromagnetic torque of the motor is estimated from the torque constant as follows: ,in The electromagnetic torque of the motor; the back electromotive force term in the motor terminal voltage balance equation. Used to verify the motor operating point, among which The back electromotive force; based on the motor rotor dynamics equations, the actual load torque is estimated using the following formula:

[0281] ;

[0282] in, The actual load torque applied to the vibration mechanism 4, This represents the equivalent moment of inertia of the motor rotor and eccentric blocks. This is the equivalent viscous damping coefficient of the motor shaft system. The rotational speed change rate is obtained from the rotational speed difference between adjacent sampling times. This formula originates from the rotational form of Newton's second law. The electromagnetic torque is equal to the sum of the load torque, the inertial torque, and the damping torque. Rearranging the terms yields the load torque estimation formula, which belongs to the standard construction of the load disturbance observer.

[0283] Step S520: According to the preset excitation force estimation rules, the nominal excitation force is calculated from the preset electromechanical conversion coefficient of the drive control signal and the vibration mechanism 4; the load disturbance force is determined from the actual load torque and the nominal excitation force.

[0284] The necessary process is as follows: from the current drive control signal and the preset electromechanical conversion coefficient of vibration mechanism 4 Calculate the nominal incentive force ,in The nominal excitation force under ideal, undisturbed operating conditions; The preset force-to-electric conversion coefficient (unit: N / V) refers to the electromechanical conversion coefficient between the excitation force output by the vibration mechanism 4 under a unit driving voltage and the velocity dimension in step S440. They belong to two calibration channels, one for speed and one for force, with different signs and dimensions, and are not interchangeable in use; the actual load torque is transmitted through the eccentric radius of the vibration mechanism 4. Converted to equivalent load force ,in For equivalent load force, The eccentric radius of the eccentric block of the dual-head vibratory motor is: The load disturbance force is:

[0285] ;

[0286] in, The load disturbance force characterizes the deviation of actual working conditions from nominal working conditions, such as user weight distribution and attitude deviation.

[0287] Step S530: According to the preset disturbance feedforward compensation rule, the load disturbance force is mapped to the disturbance equivalent displacement of each excitation degree of freedom by the transfer function from each excitation degree of freedom to each response degree of freedom in the multi-source transfer operator.

[0288] Among them, the preset disturbance feedforward compensation rule refers to the rule that converts the load disturbance force into the equivalent displacement of the excitation degree of freedom through the inverse mapping of the system transmission characteristics.

[0289] The necessary process is as follows: Take the multi-source transfer operator at the oscillation angular frequency. The transfer function submatrix from each excitation degree of freedom to each response degree of freedom in the footing region. They are aggregated into excitation channel frequency response vectors according to their excitation degrees of freedom. The excitation channel frequency response vector refers to the scalar frequency response obtained by aggregating the amplitudes of the transfer functions from the same excitation degree of freedom to each response degree of freedom using a preset aggregation rule. It is used to comprehensively characterize the overall transfer strength of the excitation channel. One implementation of the preset aggregation rule is root mean square aggregation, i.e., for the first... One degree of freedom for incentives:

[0290] .

[0291] in, In response to the total number of degrees of freedom, For the multi-source transit operator from the first The incentive degrees of freedom to the first The frequency response function with one degree of freedom; another implementation is maximum amplitude aggregation, i.e. In this embodiment, root mean square aggregation is preferably used to comprehensively reflect the average influence intensity of the excitation channel on each response point; the response generated by the load disturbance force acting on the system is equivalent to the additional displacement of the excitation degree of freedom, and the equivalent displacement of the disturbance is:

[0292] .

[0293] in, For the first The equivalent displacement of the disturbance of each degree of freedom. For the first The aggregate transfer function corresponding to each excitation degree of freedom The reference equivalent mass is a preset value; the reference equivalent mass The mass matrix of the human-machine coupling model is determined according to the following rules: The modal truncation results in step S220 are used to calculate the target vibration frequency. Modal quality of the nearby dominant mode (The mass is always 1 after normalization), take the equivalent single-degree-of-freedom mass corresponding to the dominant mode. , Take the root mean square value of the equivalent mass of each dominant mode; as a simplified engineering implementation, when the dominant mode is absolutely dominant in the vertical translational degree of freedom of the top frame (modal participation factor is greater than 0.95). The total mass of the human body is directly taken as the sum of the equivalent mass of the top frame; in this embodiment, a simplified method is used. The physical basis of this mapping is the frequency domain form of Newton's second law. The perturbation force causes the reference inertia to produce vibrational acceleration. In steady-state single-frequency vibration, the displacement amplitude and acceleration amplitude satisfy... Therefore, when inferring the equivalent excitation displacement from the disturbance force, the inverse mapping should be performed using the reference inertia, the square of the angular frequency of vibration, and the amplitude of the transfer function as divisors; regarding dimensional verification, the molecular weight dimension is... The denominator has dimensions of The dimensions of the two are: Its dimensions are consistent with those of the equivalent displacement of the disturbance.

[0294] Step S540: According to the preset drive correction rule, a drive compensation signal is generated from the disturbance equivalent displacement and superimposed on the drive control signal to correct the drive output.

[0295] The necessary process is as follows: A time-domain driving compensation signal is generated from the amplitude and phase of the equivalent displacement of the disturbance.

[0296] .

[0297] in, For the first The drive compensation signal for each degree of freedom of the excitation; the drive compensation signal is superimposed on the current drive control signal:

[0298] .

[0299] in, The corrected drive control signal is then subjected to amplitude limiting processing in step S440 to ensure that it does not exceed the drive capability boundary of the vibration mechanism 4. Through the above closed-loop compensation, the load disturbance caused by changes in user posture or weight distribution is canceled in real time, and the deviation between the actual vibration response of the foot pedal area and the user's expected vibration parameters is maintained within the preset tolerance range, with the response amplitude deviation not exceeding 10% of the expected amplitude.

[0300] In step S550, according to the preset operation termination judgment rule, when the cumulative running time of the vibration mechanism 4 reaches the user-set running time or a stop command is received, the drive correction is stopped and the vibration output is terminated.

[0301] The necessary process is as follows: cumulative running time since the system's self-vibration output starts. ; A termination check is performed for each sampling period, if ( If the user sets the runtime (or the human-machine interface receives a stop command), the amplitude of the drive control signal will be linearly decayed to zero within 1 second according to the preset slow stop rule, the drive correction will be stopped and the vibration output will be terminated to avoid the shock caused by sudden stop; at the same time, the actual vibration parameters and load compensation records of this operation will be written into the system memory for later retrieval.

[0302] Furthermore, considering that each step of the above method needs to be executed collaboratively by corresponding functional units to be implemented, this embodiment discloses a body shaping machine driving system for executing the above body shaping machine driving method. The correspondence between each functional unit and the method steps is as follows.

[0303] The sensing unit is deployed in the foot pedal area to acquire the user's weight and pressure distribution, and transmits them to the model building unit. This unit corresponds to the data acquisition stage in step S100. Its specific composition (combination of pressure sensing array and weight detection module) and data output format are described in step S100.

[0304] The model building unit is used to retrieve preset human body parameters, top frame structural parameters, and rotational constraint stiffness. The mass matrix is ​​determined by body weight and preset human body parameters, and the load distribution matrix is ​​determined by pressure distribution. The mass matrix, load distribution matrix, top frame structural parameters, and rotational constraint stiffness are fused to build a human-machine coupled model. This unit corresponds to step S100. For the determination of the mass matrix, refer to steps S110 to S160. For the determination of the load distribution matrix, refer to steps S1A0 to S1F0. For the construction of the human-machine coupled model, refer to steps S1a0 to S1d0.

[0305] The transfer operator solving unit is used to solve the multi-source transfer operator, which includes direct channels and cross-coupled channels, based on the human-machine coupling model, with the installation position of the vibration mechanism 4 as the excitation point and the foot pedal area as the response point. The cross-coupled channels are triggered by the rotational constraint stiffness being a finite value. This unit corresponds to step S200. For the specific solution process, refer to steps S210 to S240 and steps S241 to S245.

[0306] The feedforward compensation unit is used to receive the desired vibration parameters input by the user, and to perform feedforward compensation on the desired vibration parameters based on the multi-source transfer operator to obtain the theoretical excitation displacement sequence; this unit corresponds to step S300, and the specific process is referred to steps S310 to S360.

[0307] The correction and optimization unit is used to correct the amplitude and phase of the theoretical excitation displacement sequence and optimize the vibration field consistency. It converts the vibration mechanism 4 into a drive control signal according to the preset drive conversion rule. This unit corresponds to step S400. For the specific process, refer to steps S410 to S450.

[0308] The load compensation unit is used to monitor the motor operating current and speed during vibration operation, and correct the drive control signal based on the current, speed and multi-source transmission operator to compensate for load disturbances; this unit corresponds to step S500, and the specific process is referred to steps S510 to S540.

[0309] The operation control unit is used to control the vibration mechanism 4 to stop operating when the user-set operating time is reached or a stop command is received; this unit corresponds to the operation termination judgment link in step S500, and the specific process is referred to step S550.

[0310] The above-mentioned units can be physically integrated into the same controller, or they can be deployed in a distributed manner according to the sensing layer, computing layer, and driving layer. Data transmission between units is completed through the system bus or wireless communication link. The unit division is only a functional logic division and does not constitute a limitation on the hardware form.

[0311] To verify the technical effect of the proposed solution, two driving strategies were used to conduct comparative tests on the same body shaping machine prototype. The test conditions were as follows: the user's weight was 75kg and height was 175cm, the expected vibration parameter was an amplitude of 3mm, and three response points were taken at the left, middle and right ends of the foot pedal area. The maximum relative deviation between the actual amplitude and the expected amplitude at each response point was used as the evaluation index, and the preset tolerance was 10%, which is the allowable range of response amplitude deviation mentioned in step S540.

[0312] like Figure 2 As shown, in the full-band sweep test from 5Hz to 50Hz, the relative amplitude deviation of the traditional open-loop drive (existing technology, i.e., directly outputting the drive signal according to the expected parameters without any compensation) shows significant peaks near the two natural frequencies (approximately 18Hz and 38Hz), reaching 40% and 35% respectively, and even far from the resonance region, it reaches 13% to 25%, exceeding the preset tolerance across the entire frequency band. The fundamental reason is that the open-loop drive cannot sense the frequency characteristics of the human-machine coupling system. Near the natural frequency, the system response is amplified by resonance, and the actual vibration intensity far exceeds the user-set value, posing a safety hazard of excessive vibration. The solution proposed in this application has a deviation of no more than 5.5% across the entire frequency band, with a peak value of only about 5% near the resonance region, always within the preset tolerance range. The mechanism is that feedforward compensation eliminates the steady-state error caused by the system's transmission characteristics, amplitude and phase correction precisely cancels the amplitude-frequency attenuation and phase-frequency lag of the direct channel, and consistency optimization further suppresses the amplitude unevenness between response points caused by the cross-coupled channel. The superposition of these three factors systematically suppresses the amplification effect in the resonance region.

[0313] In the load disturbance test, the user's center of gravity actively shifted at the 20th second of operation, simulating the posture adjustment condition in actual use. The amplitude deviation of the traditional open-loop drive continuously climbed from about 9% before the disturbance to more than 25%, and could not recover on its own until the end of the operation. The fundamental reason is that the open-loop system has no load sensing channel. The load disturbance caused by the change in user posture is completely transmitted to the vibration output, which manifests as a change in force when the person shifts their stance. The solution of this application jumps to about 9.5% at the moment of disturbance. Then, the load compensation unit estimates the load disturbance force in real time through the motor operating current and speed, and inversely maps the disturbance into the compensation displacement of the excitation degree of freedom and superimposes it on the drive control signal. The deviation drops back to below 5% within about 5 seconds and is maintained for a long time in the remaining operation time. This shows that the load closed-loop compensation of this application can sense and offset the load disturbance caused by the change in user posture in real time, ensuring the consistency of vibration intensity and vibration mode throughout the entire use period.

[0314] In summary, this application solves the technical problems of large vibration deviation in the entire frequency band, especially near the resonance region, and the inability to recover vibration intensity under load disturbance by human-machine coupling modeling, multi-source transfer operator with cross-coupling channels, two-level correction of feedforward compensation and amplitude-phase correction, vibration field consistency optimization, and load closed-loop compensation. It achieves accurate vibration output with consistency across the entire frequency band, all time period, and multiple response points.

[0315] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for driving a body shaping machine, characterized in that, The body shaping machine includes a base frame, a top frame, a movable support, a vibration mechanism, and a sensing unit. The top frame has a foot pedal area. The two ends of the movable support are rotatably connected to the top frame and the base frame, respectively. The vibration mechanism is located on the base frame and connected to the top frame. The sensing unit is located in the foot pedal area. The machine includes the following steps: The system acquires body weight and pressure distribution through sensing units, retrieves preset human body parameters, top frame structural parameters and rotational constraint stiffness, determines the mass matrix of the human-machine coupling model based on body weight and preset human body parameters, determines the load distribution matrix of the user-top frame contact interface based on pressure distribution, and integrates the mass matrix, load distribution matrix, top frame structural parameters and rotational constraint stiffness to construct the human-machine coupling model. Based on the human-machine coupling model, the installation position of the vibration mechanism is taken as the excitation point and the foot pedal area is taken as the response point. The multi-source transfer operator containing direct channels and cross-coupled channels is solved. The system receives the desired vibration parameters input by the user, performs feedforward compensation on the desired vibration parameters based on the multi-source transfer operator, and obtains the theoretical excitation displacement sequence. The amplitude and phase of the theoretical excitation displacement sequence are corrected, and the vibration field consistency is optimized, which is then converted into a drive control signal to drive the vibration mechanism. During vibration operation, the operating current and speed of the motor of the vibration mechanism are monitored. Based on the motor operating current, speed and multi-source transmission operator, the drive control signal is corrected to compensate for load disturbances until the user-set running time is reached or a stop command is received.

2. The method according to claim 1, characterized in that, Preset human body parameters include height, sex, and lower limb length-to-diameter ratio. The mass matrix of the human-machine coupling model is determined as follows: According to the preset mass allocation rules, the total mass of the human body is determined by weight and gender correction coefficient, the total mass of the lower limbs is determined by the total mass of the human body and the preset lower limb mass ratio, and the mass of the thigh segment and the mass of the lower limb segment are determined by the total mass of the lower limbs and the preset thigh-to-calf mass ratio. According to the preset geometric proportion rules, the relative coordinates of the local center of mass of the thigh and the local center of mass of the lower leg are determined by the length-to-diameter ratio of the lower limb. The overall center of mass coordinates of the lower limb are synthesized by the segmented mass of the thigh, the segmented mass of the lower leg, and the relative coordinates of the local centers of mass. According to the preset axis-shifting synthesis rules, the overall rotational inertia of the lower limb is calculated from the segmental mass of the thigh, the segmental mass of the lower leg, the relative coordinates of the local center of mass, the coordinates of the overall center of mass of the lower limb, and the length-to-diameter ratio of the lower limb. According to the preset trunk and upper limb inertia correction rules, the equivalent rotational inertia of the human body is determined by the overall rotational inertia of the lower limbs and the preset trunk and upper limb inertia correction coefficient. According to the preset equivalent mass-inertia conversion rules, the equivalent mass and equivalent rotational inertia of the top frame are calculated from the material density and geometric dimensions in the top frame structural parameters. According to the preset diagonal assembly rules, the total mass of the human body, the equivalent moment of inertia of the human body, the equivalent mass of the top frame, and the equivalent moment of inertia of the top frame are sorted according to the preset degrees of freedom and filled into the diagonal elements to obtain the mass matrix.

3. The method according to claim 2, characterized in that, Determining the load distribution matrix at the user-top frame contact interface includes: The pressure distribution is denoised according to the preset filtering rules to obtain the pre-processed pressure distribution field. The foot pedal area is divided into several preset shape sensing unit grids according to the preset grid discretization rules. The coordinates of each grid node and the area of ​​each grid unit are determined. The pre-processed pressure distribution field is spatially registered with each grid unit. According to the preset pressure integration rules, the registered pressure data and cell area in each grid cell are numerically integrated to obtain the equivalent cell resultant force of each grid cell. According to the preset node allocation rules, the element resultant force is allocated to each grid node by the equivalent element resultant force and the preset shape function value, so as to obtain the equivalent nodal force of each grid node; According to the preset boundary correction rules, the node distance is calculated from the grid node coordinates and the preset installation position of the movable support, and the boundary correction coefficient is determined from the node distance and rotational constraint stiffness. The equivalent node force of the grid node that is less than the preset boundary threshold from the geometric boundary of the foot pedal area is corrected for the support boundary effect. According to the preset load assembly rules, the corrected equivalent nodal forces of each grid node are sorted and assembled into a load distribution matrix according to the preset degrees of freedom.

4. The method according to claim 3, characterized in that, A human-machine coupled model is constructed by integrating the mass matrix, load distribution matrix, top frame structural parameters, and rotational constraint stiffness. The top frame structural parameters include elastic modulus, moment of inertia, and geometric dimensions. The steps include: According to the preset discretization rules of the top frame, the stiffness matrix of the top frame is constructed from the elastic modulus, the moment of inertia of the section and the geometric dimensions; According to the preset boundary support rules, a boundary support matrix is ​​constructed based on the rotational constraint stiffness and the preset installation position of the movable support component. The boundary support matrix includes elastic constraint terms for rotational degrees of freedom and rigid constraint terms for translational degrees of freedom. According to the preset degree of freedom alignment rules, the overall stiffness matrix is ​​constructed from the top frame stiffness matrix and the boundary support matrix; according to the preset Rayleigh damping rules, the damping matrix is ​​determined from the mass matrix, the overall stiffness matrix and the preset damping ratio, and the mass-stiffness-damping system matrix of the human-machine coupling model is constructed from the mass matrix, the overall stiffness matrix and the damping matrix. According to the preset external force mapping rules, the load distribution matrix is ​​mapped to the system external force vector. The mass matrix, damping matrix and overall stiffness matrix are used as the coefficient matrix of the dynamic equation, and the system external force vector is used as the excitation term to construct the human-machine coupled dynamic equation and obtain the human-machine coupled model.

5. The method according to claim 4, characterized in that, Based on the human-machine coupling model, taking the installation location of the vibration mechanism as the excitation point and the foot pedal area as the response point, the multi-source transfer operator, which includes direct channels and cross-coupled channels, is solved as follows: The mass matrix, damping matrix, and overall stiffness matrix are read from the human-machine coupling model as coefficient matrices. The excitation degree of freedom and response degree of freedom corresponding to the installation position of the vibration mechanism and the foot pedal area are determined according to the preset excitation-response mapping rules. According to the preset modal truncation rule, the dominant natural frequencies and corresponding mode shapes within the preset order range are solved by the coefficient matrix; According to the preset direct channel construction rules, the direct transmission channels from each excitation degree of freedom to the corresponding response degree of freedom are constructed by the dominant natural frequency, the corresponding mode shape, and the preset frequency discrete axis. According to the preset cross-coupling construction rules, the cross-coupling gain is determined by the boundary rotational degree of freedom corresponding to the finite value of the rotational constraint stiffness and the modal participation factor of the mode shape on the boundary rotational degree of freedom. The cross-coupling gain, the dominant natural frequency, the corresponding mode shape and the frequency discrete axis are used to construct the cross-coupling channel from each excitation degree of freedom to the non-corresponding response degree of freedom. According to the preset transfer operator assembly rules, the direct channel and the cross-coupled channel are assembled into a multi-source transfer operator according to the frequency discrete axis, excitation degree of freedom number and response degree of freedom number.

6. The method according to claim 5, characterized in that, According to the preset cross-coupling construction rules, the cross-coupling gain is determined by the boundary rotational degrees of freedom corresponding to the finite value of the rotational constraint stiffness and the modal participation factor of the mode shape on that boundary rotational degree of freedom, including: The boundary rotational degree of freedom number is determined by the preset installation position of the movable support component and according to the preset boundary degree of freedom identification rules. According to the preset modal participation factor extraction rules, the modal participation factors of each dominant mode on the boundary rotation degree of freedom are extracted by the corresponding mode shape and boundary rotation degree of freedom number; According to the preset residual flexibility correction rule, the truncated mode residual flexibility matrix is ​​calculated from the overall stiffness matrix and the corresponding mode shape, and the equivalent boundary stiffness correction coefficient on the boundary rotational degrees of freedom is calculated from the residual flexibility matrix and the rotational constraint stiffness. The nominal cross-coupling gain is calculated from the modal participation factor, the equivalent boundary stiffness correction coefficient, and the rotational constraint stiffness according to the preset nominal gain calculation rules. According to the preset perturbation correction rules, the sensitivity coefficient of the cross-coupling gain to the rotational constraint stiffness is calculated from the nominal cross-coupling gain, the rotational constraint stiffness, and the preset stiffness perturbation range. The perturbation-corrected cross-coupling gain is then determined from the sensitivity coefficient and the preset stiffness perturbation reference value.

7. The method according to claim 1, characterized in that, The system receives the desired vibration parameters input by the user, performs feedforward compensation on the desired vibration parameters based on a multi-source transfer operator, and obtains the theoretical excitation displacement sequence, including: According to the preset parameter receiving rules, the vibration frequency, vibration amplitude and vibration mode identifier input by the user are received, and the protocol is parsed, validity is checked and anti-misclick filtering is performed to obtain the desired vibration parameter set to be processed. According to the preset parameter analysis rules, the vibration mode type and the preset weight of each vibration direction component are determined by the vibration mode identifier in the expected vibration parameter set. According to the preset expected response mapping rules, the expected vibration response is determined by the preset weights of the vibration frequency, vibration amplitude, vibration mode type and each vibration direction component in the expected vibration parameter set. According to the preset frequency response value rules, the corresponding frequency response matrix is ​​obtained from the vibration frequency and the multi-source transmission operator; According to the preset feedforward compensation rules, the frequency response inverse operation is performed from the frequency response matrix and the desired vibration response to obtain the theoretical frequency domain excitation displacement; According to the preset time-domain conversion rules, the amplitude and phase parameters, vibration frequency, and preset sampling period of the theoretical frequency-domain excitation displacement are converted into the theoretical excitation displacement sequence.

8. The method according to claim 7, characterized in that, The theoretical excitation displacement sequence is corrected in amplitude and phase, and the vibration field consistency is optimized. This is then converted into a drive control signal to drive the vibration mechanism. According to the preset time-domain amplitude and phase extraction rules, the time-domain amplitude and phase parameters of each excitation degree of freedom are extracted from the theoretical excitation displacement sequence; According to the preset frequency-time compensation rules, the time-domain amplitude-phase compensation coefficients are determined by the amplitude-frequency characteristics and phase-frequency characteristics of the direct channel in the multi-source transfer operator at the vibration frequency. According to the preset amplitude-phase compensation rules, the amplitude gain and phase lead / lag compensation of the time-domain amplitude-phase parameters are adjusted by the time-domain amplitude-phase compensation coefficients to offset the amplitude-frequency attenuation and phase-frequency lag of the direct channel in the multi-source transfer operator, so as to obtain the amplitude-phase corrected excitation sequence. According to the preset vibration field consistency optimization rules, with the goal of minimizing the amplitude deviation between the vibration responses of each degree of freedom in the foot-stepped area predicted by the multi-source transfer operator, the amplitude-phase correction excitation sequence is uniformly optimized and allocated to obtain the optimized excitation sequence. According to the preset drive conversion rules, the optimized excitation sequence is converted into the corresponding voltage drive signal or PWM duty cycle signal according to the vibration mechanism type to obtain the drive control signal; The vibration mechanism is driven by the drive control signal according to the preset output drive rules.

9. The method according to claim 8, characterized in that, The drive control signal is corrected based on motor operating current, speed, and multi-source transmission operators to compensate for load disturbances until the user-set running time is reached or a stop command is received, including: According to the preset load observation rules, the actual load torque applied to the vibration mechanism is estimated in real time by combining the motor operating current and speed of the vibration mechanism with the motor torque constant and back electromotive force constant in the preset electromagnetic parameters of the motor. According to the preset excitation force estimation rules, the nominal excitation force is calculated from the preset electromechanical conversion coefficient of the drive control signal and the vibration mechanism; the load disturbance force is determined from the actual load torque and the nominal excitation force. According to the preset disturbance feedforward compensation rule, the load disturbance force is mapped to the disturbance equivalent displacement of each excitation degree of freedom by the transfer function from each excitation degree of freedom to each response degree of freedom in the multi-source transfer operator. According to the preset drive correction rules, a drive compensation signal is generated from the equivalent displacement of the disturbance and superimposed on the drive control signal to correct the drive output. According to the preset operation termination judgment rules, when the cumulative running time of the vibration mechanism reaches the user-set running time or a stop command is received, the drive correction is stopped and the vibration output is terminated.

10. A body shaping machine drive system, characterized in that, The body shaping machine driving method according to any one of claims 1 to 9 includes: The sensing unit, located in the foot pedal area, is used to acquire the user's weight and pressure distribution, and transmit them to the model building unit; The model building unit is used to retrieve preset human body parameters, top frame structure parameters and rotational constraint stiffness. The mass matrix is ​​determined by body weight and preset human body parameters, and the load distribution matrix is ​​determined by pressure distribution. The mass matrix, load distribution matrix, top frame structure parameters and rotational constraint stiffness are fused to build a human-machine coupled model. The transfer operator solving unit is used to solve the multi-source transfer operator, which includes direct channels and cross-coupled channels, based on the human-machine coupling model, with the installation position of the vibration mechanism as the excitation point and the foot pedal area as the response point. The cross-coupled channels are triggered by the rotational constraint stiffness being a finite value. The feedforward compensation unit is used to receive the desired vibration parameters input by the user, and to perform feedforward compensation on the desired vibration parameters based on the multi-source transfer operator to obtain the theoretical excitation displacement sequence. The correction and optimization unit is used to correct the amplitude and phase of the theoretical excitation displacement sequence and optimize the consistency of the vibration field. It converts the vibration mechanism into a drive control signal according to the preset drive conversion rule. The load compensation unit is used to monitor the motor's operating current and speed during vibration operation, and to correct the drive control signal based on the current, speed and multi-source transmission operator to compensate for load disturbances. The operation control unit is used to control the vibration mechanism to stop operating when the user-set operating time is reached or when a stop command is received.

Images