Multi-mode hierarchical control method, device and storage medium for electromechanical actuators

Abstract

This invention discloses a multi-mode hierarchical control method, device, and storage medium for an electromechanical brake, belonging to the field of braking control technology. The method includes: determining the current braking mode, which includes: no-travel braking mode, clamping force following braking mode, and retraction mode; when in no-travel braking mode, using a zero-force servo control method for braking, and determining whether the friction pad and brake disc have reached the contact proximity point; if so, entering the clamping force following braking mode; when in clamping force following braking mode, using a dual-fuzzy adaptive sliding mode control method based on the EMB dynamics model to track the target clamping force to enable braking; when in retraction mode, identifying the retraction intention and releasing the brake. This invention can seamlessly switch between multiple control modes, accurately identify the critical contact state of braking, improve tracking accuracy and stability, and avoid dragging or jerking during retraction.

Description

Technical Field

[0001] This invention belongs to the field of braking control technology, specifically relating to a multi-mode hierarchical control method, device, and storage medium for an electromechanical brake. Background Technology

[0002] In the field of automotive braking, electromechanical brakes (EMBs) use an electric motor as the core power source. Electronic sensors monitor braking demand in real time, and the controller calculates and drives the motor to transmit braking force to the brake disc or drum through a gear transmission mechanism. Because electromechanical brakes do not have hydraulic transmission and have high response potential, they have become an important direction to replace traditional hydraulic braking systems.

[0003] Existing technologies employ hierarchical control to control EMB, but these technologies lack multi-mode collaborative switching mechanisms, resulting in poor braking stability. Furthermore, each control mode has significant shortcomings: the idle travel mode relies on real-time torque mutations to identify the contact state, at which point actual contact is difficult to avoid impacts, and there is no self-calibration of the contact position; clamping force control modes mostly use fixed-parameter algorithms, making it difficult to adapt to load disturbances and resulting in low tracking accuracy; and the retraction mode lacks differentiated strategies, making it prone to dragging or jerking. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a multi-mode hierarchical control method, device, and storage medium for electromechanical brakes. It can seamlessly switch between multiple control modes, accurately identify the contact state between the friction pads and the brake disc, improve tracking accuracy and stability, and avoid dragging or jerking during retraction.

[0005] This invention provides the following technical solution:

[0006] Firstly, a multi-mode hierarchical control method for an electromechanical brake is provided, comprising:

[0007] The system continuously monitors the target clamping force applied by the driver and the actual clamping force output by the brake, and determines the current braking mode, which includes: free travel braking mode, clamping force following braking mode, and retraction mode.

[0008] When in the no-travel braking mode, the zero-force servo control method is used for braking. The friction pads and brake discs are judged to be close to the contact point by micro-pulse excitation and equivalent stiffness. If they are, the clamping force following braking mode is entered; otherwise, the no-travel braking mode continues.

[0009] When in clamping force following braking mode, based on the EMB dynamics model, a dual fuzzy adaptive sliding mode control method is used to track the target clamping force so that the brake outputs the actual clamping force.

[0010] When in retraction mode, the intention to retract is identified based on changes in the target clamping force to release the brake.

[0011] Optionally, the determination of the current braking mode specifically involves: when the target clamping force is not greater than 0, no braking operation is performed; when the target clamping force is greater than 0, it is determined whether the actual clamping force output by the brake is 0; if the actual clamping force output by the brake is 0, the system enters the no-stroke braking mode; if the actual clamping force is not 0, it is determined whether the target clamping force has decreased; if it has decreased, the system enters the retraction mode; otherwise, the system enters the clamping force following braking mode.

[0012] Optionally, the zero-force servo control method for braking specifically involves: under zero-force constraints, obtaining the target torque of the brake motor in the no-stroke braking mode using the following solution formula. ;

[0013] The target torque of the brake motor The solution formula is:

[0014] ;

[0015] The zero-force constraint is: ;

[0016] in, and The first The frictional torque and viscosity coefficient are estimated by the observer updated at the sampling time. and The first The angular velocity and angular displacement of the motor rotor measured at the sampling time. For the amplitude limiting function, For position servo gain, This is the prior value of the contact angle. This is a time-invariant constant, specifically the torque limiting during the positioning phase. The upper limit of the near-zero force is set.

[0017] Optionally, the step of determining whether the friction pad and the brake disc have reached the point of near contact through micro-pulse excitation and equivalent stiffness specifically involves:

[0018] In the At the sampling time, for the brake motor The brake motor is obtained by superimposing the shaft current with a micropulse excitation. The shaft current command is as follows:

[0019] ;

[0020] in, In the first The brake motor at the sampling time Shaft current command, This is the baseline current during the process of the friction pad approaching the brake disc. The set micro-pulse current amplitude, For symbolic functions, The frequency of the micropulse current. The sampling period of the brake motor. ,but Take 1, ,but Take 0, ,but Take -1;

[0021] Within the defined sliding window, the amplitude of the angular displacement fluctuation caused by the superimposed micropulse excitation is defined. and torque fluctuation amplitude And calculate the equivalent contact stiffness. and torque change rate ;

[0022] ; ; ; ;

[0023] in, and These represent the maximum and minimum angular displacements within the time period corresponding to the sliding window. Indicates the definition symbol, The length of the defined sliding window, Indicates the time period Inside, This is the torque coefficient of the brake motor. For the first The torque of the brake motor at the sampling moment;

[0024] If the following conditions are met during the duration: Then it enters the clamping force following braking mode;

[0025] in, , This represents the median rate of change of torque within the time period corresponding to the sliding window. The first time period corresponding to the sliding window The rate of change of torque, A function that takes the median value. To achieve an adaptive threshold, it follows the angular velocity of the motor rotor. Adjustment, For stiffness threshold, The logical AND operator.

[0026] Optionally, the method of tracking the target clamping force using a dual-fuzzy adaptive sliding mode control based on the EMB dynamics model, so that the brake outputs the actual clamping force, specifically involves:

[0027] Based on the EMB dynamics model, a linear sliding surface and a control law are defined. The control law includes an equivalent control term and a switching control term. The switching control term introduces a saturation function with respect to the boundary layer thickness.

[0028] Using clamping force deviation and clamping force deviation change rate as input variables, the first fuzzy controller performs fuzzy inference and defuzzification through a preset fuzzy rule table to obtain the change in sliding mode gain, and updates the sliding mode gain based on the set initial gain.

[0029] Using the actual clamping force change rate as the input variable, the second fuzzy controller performs fuzzy inference and defuzzification through a preset fuzzy rule table to obtain the change in boundary layer thickness, and updates the boundary layer thickness based on the set initial boundary layer thickness.

[0030] Based on the updated sliding mode gain and boundary layer thickness, a sliding mode control method is used to track the target clamping force so that the brake can output the actual clamping force.

[0031] Optionally, the fuzzy sets corresponding to the input and output variables of the two fuzzy controllers are both: ;

[0032] The triangular membership functions for the input and output variables of both fuzzy controllers are:

[0033] ; ;

[0034] ; ;

[0035] ; ;

[0036] ;

[0037] in, These are the input or output variables of the fuzzy controller. , , , , , , , , It is the universe of discourse of the input or output variables of the fuzzy controller. , , , , , and These are the input variables or output variables of the fuzzy controller. Belongs to negative large fuzzy set Membership degree, belonging to negative fuzzy set Membership degree, belonging to negative small fuzzy set Membership degree, belonging to zero fuzzy set Membership degree, belonging to positive small fuzzy set Membership degree, belonging to the central fuzzy set Membership degree and belonging to the large fuzzy set The degree of membership.

[0038] Optionally, when the first fuzzy controller obtains the change in sliding mode gain after fuzzy inference and defuzzification using a preset fuzzy rule table, it uses a product method to determine the activation intensity and outputs the change in sliding mode gain after defuzzification according to the following formula. ;

[0039] ;

[0040] in, For the first The activation strength of the rule, For effective fuzzy rules, This represents the change in sliding mode gain. Let the universe of discourse be the variable of sliding mode gain. In the first Under this rule, the change in sliding mode gain Belongs to fuzzy set The membership degree corresponding to time, fuzzy set It is one of the following: negative large fuzzy set, negative medium fuzzy set, negative small fuzzy set, zero fuzzy set, positive small fuzzy set, positive medium fuzzy set, and positive large fuzzy set. Indicates to Integrate the points.

[0041] Optionally, the retraction mode includes a rapid retraction mode and a smooth retraction mode. Specifically, identifying the retraction intention to release the brake based on changes in the target clamping force involves:

[0042] like In this mode, the vehicle enters rapid reversal mode and uses a maximum torque control algorithm to release the brakes; among which, The target clamping force for the previous control cycle. The target clamping force for the current control cycle. To represent the actual clamping force before the brake is released, The logical AND operator;

[0043] like If it is in smooth retraction mode, the smooth retraction cycle and the change gradient of clamping force are calculated using the following formula based on the target clamping force of the current control cycle, and the actual clamping force is gradually reduced based on the change gradient of clamping force.

[0044] ;

[0045] ;

[0046] ;

[0047] in, To smooth out the rollback cycle, and These are the maximum and minimum set smooth rollback periods, respectively. This is the clamping force offset. This is the maximum design clamping force of the brake. This represents the gradient of the clamping force during the current control cycle. This represents the actual clamping force output during the current control cycle.

[0048] In a second aspect, a computer device is provided, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the multi-mode hierarchical control method for the electromechanical actuator as described in any one of the first aspects.

[0049] Thirdly, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, it implements the steps of the multi-mode hierarchical control method for the electromechanical actuator described in any one of the first aspects.

[0050] Compared with the prior art, the beneficial effects of the present invention are:

[0051] In the no-travel braking mode, this invention uses zero-force servo and micro-pulse excitation to accurately identify the contact between the friction pad and the brake disc, eliminating interference and achieving impact-free approach. In the clamping force following braking mode, it adopts dual fuzzy adaptive sliding mode control to dynamically adapt parameters, suppressing chatter, and combining a hierarchical dynamic model to improve tracking accuracy and stability. In the retraction mode, it can recognize the driver's intention, distinguish between rapid retraction and smooth retraction, and avoid dragging and jerking. Overall, the multi-mode switching of this invention is seamless, with high stability, adaptable to intelligent driving, and without a hydraulic module, it is easy to integrate into the whole vehicle. In addition, it can reduce costs, energy consumption and pollution, and make up for many shortcomings of traditional electromechanical brakes. Attached Figure Description

[0052] Figure 1 This is a control flowchart of the multi-mode hierarchical control method for the electromechanical brake of the present invention.

[0053] Figure 2 This is a flowchart of a complete braking process using the multi-mode hierarchical control method of the present invention. Detailed Implementation

[0054] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the scope of protection of the present invention. It should be noted that the term "comprising" and any variations thereof in the specification, claims and the above-mentioned drawings of the present invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or devices.

[0055] Example 1:

[0056] like Figure 1 As shown, a multi-mode hierarchical control method for an electromechanical brake includes:

[0057] It continuously monitors the target clamping force applied by the driver and the actual clamping force output by the brake, and determines the current braking mode, which includes: free travel braking mode, clamping force following braking mode and retraction mode.

[0058] When in the no-travel braking mode, the zero-force servo control method is used for braking. The friction pads and brake discs are judged to be close to the contact point by micro-pulse excitation and equivalent stiffness. If they are, the clamping force following braking mode is entered; otherwise, the no-travel braking mode continues.

[0059] When in clamping force following braking mode, based on the EMB dynamics model, a dual fuzzy adaptive sliding mode control method is used to track the target clamping force so that the brake outputs the actual clamping force.

[0060] When in retraction mode, the intention to retract is identified based on changes in the target clamping force to release the brake.

[0061] In this embodiment, the current braking mode is determined as follows: when the target clamping force is not greater than 0, no braking operation is performed; when the target clamping force is greater than 0, it is determined whether the actual clamping force output by the brake is 0. If the actual clamping force output by the brake is 0, the no-stroke braking mode is entered; if the actual clamping force is not 0, it is determined whether the target clamping force has decreased. If it has decreased, the retraction mode is entered; otherwise, the clamping force following braking mode is entered. That is, if the target clamping force increases or remains unchanged, the clamping force following braking mode is entered.

[0062] 1. Free-travel braking mode.

[0063] In this embodiment, the first mode of the vehicle braking process is the no-travel braking mode, which is used to quickly, smoothly, and reproducibly move the actuator from the initial gap position to the brake pad contact critical point without generating effective clamping force. The core of the method includes four synergistic steps: zero-force servoing to eliminate friction and gap interference, micro-pulse excitation and equivalent impedance estimation to predict contact in advance, kinetic energy constraint soft landing to suppress contact impact, and online shortest-time solution to ensure the shortest approach time within electrical and energy constraints. Once contact is confirmed, it seamlessly switches to clamping force-following braking.

[0064] In the no-travel braking mode, to avoid generating actual clamping force, a zero-force servo drive is used to bring the actuator close to the contact point. Specifically, the target torque calculation formula for the brake motor is:

[0065] ;

[0066] In the formula, and The first The frictional torque and viscosity coefficient are estimated by the observer updated at the sampling time. and The first The angular velocity and angular displacement of the motor rotor measured at the sampling time. For the amplitude limiting function, For position servo gain, This is the prior value of the contact angle. It is a time-invariant constant, specifically the torque limit during the positioning stage.

[0067] Apply zero-force constraints: ;

[0068] In the formula, , To set the upper limit of near-zero force, To generate a minimum threshold for perceptible clamping force, a disturbance observer is designed for compensation, ensuring that the theoretical clamping force remains near zero during the approach process of the actuator, thereby eliminating the interference of mechanism clearance and friction on contact determination.

[0069] To accurately determine whether the friction pads and brake disc have reached the point of near contact, the motor is monitored during the approach process. shaft current Superimposed small-amplitude micropulse excitation, at the first Sampling time, The shaft current command is:

[0070] ;

[0071] In the formula, In the first The brake motor at the sampling time Shaft current command, The set micro-pulse current amplitude ( ), The maximum current amplitude is set. The frequency of the micropulse current is usually selected within the bandwidth of the current loop and to avoid mechanical resonance. The sampling period for the motor; The baseline current from the zero-force servo during the approach of the friction pads and brake disc. For symbolic functions, ,but Take 1, ,but Take 0, ,but Set to -1; the small-amplitude micro-pulse excitation method can adopt existing technology, and other specific parameters of the small-amplitude pulse excitation can also be determined based on expert experience or existing technology.

[0072] Of course, in some other embodiments, micro-vibration excitation and equivalent impedance estimation can also be used to predict contact in advance, and kinetic energy constraint soft landing can be used to suppress contact impact.

[0073] Define a sliding window with a length of Within the window, calculate the amplitude of angular displacement fluctuation caused by the excitation. and torque fluctuation amplitude :

[0074] ; ;

[0075] In the formula, and These represent the maximum and minimum angular displacements within the time period corresponding to the sliding window. Indicates the definition symbol, The length of the defined sliding window, Indicates the time period Inside, This is the torque coefficient of the brake motor.

[0076] Estimated equivalent contact stiffness :

[0077] ;

[0078] When the friction pads are about to release the brake disc Boundary reduction, equivalent contact stiffness A sudden increase is expected.

[0079] Simultaneously, calculate the torque change rate. :

[0080] ;

[0081] And use median to suppress noise: ;

[0082] In the formula, This represents the median rate of change of torque within the time period corresponding to the sliding window. The first time period corresponding to the sliding window The rate of change of torque, , It is a function that takes the median value.

[0083] After calculating the equivalent stiffness and torque change rate, a joint criterion of torque change rate and stiffness surge is used to identify the contact trend, requiring the duration to exceed [a certain value]. The specific formula is:

[0084] ;

[0085] In the formula, To achieve an adaptive threshold, it follows the angular velocity of the motor rotor. Adjustment, The stiffness threshold can be calibrated based on the motor rotor angular velocity and brake temperature. The shortest time to maintain the above conditions (usually 5-10 minutes) ), used to suppress transient noise.

[0086] When the friction pads and brake disc reach the point of near contact, define The online predicted contact angular displacement will serve as the target point for soft landing and mode switching. It provides an angular displacement threshold to prevent excessive impact at the moment of contact during subsequent braking processes. When the actual angular displacement... At this point, it is assumed that the friction pads are in complete contact with the brake disc, and a clamping force begins to be generated. This marks the end of the idle travel braking mode and the transition to the clamping force following braking mode.

[0087] II. Clamping force following braking mode.

[0088] First, a dynamic model of the electromechanical brake needs to be built. Taking a permanent magnet synchronous motor driven by a planetary ball screw EMB as an example, the dynamic model is built in three layers:

[0089] The dynamic equation of the motor is:

[0090] ;

[0091] In the formula, For the stator inductance of the motor, For the stator resistance of the motor, The back electromotive force constant is... This is the torque coefficient of the brake motor. The moment of inertia of the motor. The damping coefficient is... For load torque, for shaft current, Input voltage, This represents the angular velocity of the motor.

[0092] The dynamic equation of the speed reduction transmission mechanism is:

[0093] ;

[0094] In the formula, The reduction ratio, The linear velocity of the leadscrew. The angular velocity of the leadscrew. The lead radius of the leadscrew. , For the lead screw, The total transmission efficiency. For the lead screw torque, .

[0095] The dynamic characteristics of clamping force are:

[0096] ;

[0097] In the formula, , and These are equivalent mass, damping, and stiffness, respectively. Assuming the clamping force is the equivalent output force of the motor, its dynamic characteristics can be simplified to a first-order inertial system. Assuming the backlash is eliminated, the clamping force... , This is the equivalent stiffness coefficient for displacement and clamping force, reflecting the clamping force generated per unit displacement. This refers to the displacement caused by the deformation of the friction pads after they contact the brake disc. Replace We can obtain:

[0098] ;

[0099] Ignoring higher-order dynamics of clamping force, let and define the time constant. , Simplified to obtain clamping force response for:

[0100] ;

[0101] With the goals of rapidly tracking the clamping force target value and suppressing chattering, a sliding surface and control law are designed based on the above dynamic model.

[0102] When selecting a linear sliding surface, it is necessary to include clamping force deviation. Rate of change of clamping force deviation To ensure rapid system convergence: ,in, , This is the convergence coefficient; the larger the value, the faster the sliding mode curve converges. Represents the sliding surface. A saturation function is used. Design the sliding mode control rate by adjusting the boundary layer thickness. Suppressing high-frequency chattering in traditional sliding mode.

[0103] ;

[0104] Control rate Divided into two parts: . For equivalent control items, To switch control items.

[0105] based on The design of the sliding surface Let the derivative of the sliding surface Solve for the control input, ignoring And because Determined by the control input, its derivative dynamics can be ignored, and the equivalent control law can be obtained in the end: In the formula, This is the equivalent motor torque coefficient. , , The clamping force is the target force.

[0106] The switching control term is used to overcome load disturbances. It introduces a saturation function with respect to the boundary layer thickness. The formula for solving the switching control term is:

[0107] ;

[0108] in, For sliding mode gain, the initial value is taken as... That's it; subsequent adjustments will be made dynamically by the first fuzzy controller. The initial value is the boundary layer thickness. The value is set to 0.5~2, and then dynamically adjusted by the second fuzzy controller.

[0109] First fuzzy controller: clamping force deviation The rate of change of clamping force deviation is used express. and As input variable 1 and input variable 2, the fuzzy set is defined as follows: The two input variables have different universes of discourse, for example, The domain of discourse is The unit is N. The domain of discourse is The unit is N / s, and the universe of discourse can be evenly distributed according to the number of fuzzy sets. Sliding mode gain correction. Fuzzy set definition Sliding mode gain correction amount The domain of discourse is also different from the input variables.

[0110] The second fuzzy controller: Similar in design to fuzzy controller 1, the input variable is the rate of change of the actual clamping force. The domain of discourse is set as A fuzzy set is defined as: For the output variable boundary layer thickness correction amount The domain of discourse is A fuzzy set is defined as: Similarly, a triangular membership function is used, and a fuzzy rule base is established.

[0111] The triangular membership functions for the input and output variables of both fuzzy controllers are:

[0112] ; ;

[0113] ; ;

[0114] ; ;

[0115] ;

[0116] in, These are the input or output variables of the fuzzy controller, namely, clamping force deviation, clamping force deviation rate of change, actual clamping force rate of change, change in sliding mode gain, or change in boundary layer thickness. , , , , , , , , It is the universe of discourse of the input or output variables of the fuzzy controller. , , , , , and These are the input variables or output variables of the fuzzy controller. Belongs to negative large fuzzy set Membership degree, belonging to negative fuzzy set Membership degree, belonging to negative small fuzzy set Membership degree, belonging to zero fuzzy set Membership degree, belonging to positive small fuzzy set Membership degree, belonging to the central fuzzy set Membership degree and belonging to the large fuzzy set The degree of membership.

[0117] During defuzzification, the activation intensity is determined using a product method. For example, for a fuzzy rule: if yes and yes ,So yes (in , It is the input fuzzy set. It is the change in sliding mode gain. Activation strength of the fuzzy set to which it belongs The product method is used to determine this, that is: , Indicates clamping force deviation Belongs to fuzzy set The degree of membership corresponding to time. Indicates the rate of change of clamping force deviation Belongs to fuzzy set The degree of membership corresponding to the time.

[0118] Assuming there is a total The first valid fuzzy rule, the The activation strength of the rule is The accurate output after deblurring for:

[0119] ;

[0120] in, Let the universe of discourse be the variable of sliding mode gain. In the first Under this rule, the change in sliding mode gain Belongs to fuzzy set The membership degree corresponding to time, fuzzy set It is one of the following: negative large fuzzy set, negative medium fuzzy set, negative small fuzzy set, zero fuzzy set, positive small fuzzy set, positive medium fuzzy set, and positive large fuzzy set. Indicates to Integrate the points.

[0121] After obtaining the change in sliding mode gain, based on the set initial gain Update the sliding mode gain; In this embodiment, the solution is performed according to... Step-by-step solution Finally obtained Thus, the sliding mode gain is obtained. With boundary layer thickness This enables dual-fuzzy adaptive sliding mode control in clamping force following mode.

[0122] In this embodiment, the core control strategy of the first fuzzy controller is based on the differentiated requirements of two types of working conditions: when the clamping force deviates... The larger the value, the greater the deviation between the actual clamping force and the target, requiring an increase in sliding mode gain. To enhance the system's tracking response capability, the greater the response, the better; when the clamping force deviation change rate... When the deviation is larger, The rate of change is too fast, posing a risk of overshoot, requiring a reduction or even the application of a negative gain correction. To suppress overshoot and ensure the smoothness of the control process. Therefore, if yes and yes ,So yes The following fuzzy rules are established in this form:

[0123] R1: If It is NB and It's NB, then It's PM;

[0124] R2: If It is NB and If it's NM, then... It's PB;

[0125] R3: If It is NB and If it's NS, then... It's PB;

[0126] R4: If It is NB and It's ZO, then It's PB;

[0127] R5: If It is NB and If it's Photoshop, then... It's PB;

[0128] R6: If It is NB and If it's PM, then... It's PM;

[0129] R7: If It is NB and If it's PB, then... It's Photoshop;

[0130] R8: If It is NM and It's NB, then It's PM;

[0131] R9: If It is NM and If it's NM, then... It's PM;

[0132] R10: If It is NM and If it's NS, then... It's PB;

[0133] R11: If It is NM and It's ZO, then It's PB;

[0134] R12: If It is NM and If it's Photoshop, then... It's PB;

[0135] R13: If It is NM and If it's PM, then... It's PM;

[0136] R14: If It is NM and If it's PB, then... It's Photoshop;

[0137] R15: If It is NS and It's NB, then It's Photoshop.

[0138] It is worth noting that fuzzy rules reflect the influence relationship between input and output variables. The fuzzy rules R1-R15 given above do not represent all the rules. The fuzzy rules can be derived by referring to existing technologies.

[0139] In clamping force following mode, the brake continuously outputs the braking force requested by the driver until the target clamping force is released. Or it may be that a decrease in braking force demand is detected, i.e., a decrease occurs. This is a characteristic that the brake needs to switch to reversing mode at this time.

[0140] III. Rollback Mode.

[0141] Rollback modes include fast rollback mode and smooth rollback mode. It is necessary to determine whether the current rollback mode is fast rollback mode or smooth rollback mode based on the rollback intention.

[0142] like If the intention of retraction is rapid retraction, then the maximum torque control algorithm (i.e., the control algorithm for the no-travel mode) is used to release the brake, i.e., rapid retraction mode; where, The clamping force target value of the previous control cycle, The target clamping force for the current control cycle. To represent the actual clamping force before the brake is released, This is a logical AND operator. The fast retraction mode, designed for scenarios where the brake is completely released, employs a maximum permissible retraction torque control algorithm. Constrained by the mechanical response limit of the brake, it maximizes the retraction speed, achieving a reduction in clamping force from the brake's initial value to the final value while ensuring mechanical structural safety. The rapid decay to 0 ensures the timely output of vehicle power.

[0143] like If the intention is to perform a smooth retraction, an adaptive variable cycle number smooth retraction strategy is adopted to gradually reduce the actual clamping force, i.e., smooth retraction mode. Smooth retraction mode is used when the driver needs to fine-tune the brake clamping force under braking conditions (such as following distance adjustment, deceleration on a gentle slope, etc.). At this time, the principle of prioritizing smoothness and preventing a sharp decrease in clamping force is to trigger the smooth retraction mode.

[0144] An adaptive variable cycle number smooth back-off strategy is adopted to gradually reduce the actual clamping force. Specifically, the smooth back-off cycle and the gradient of clamping force change are calculated, and the actual clamping force is gradually reduced based on the gradient of clamping force change.

[0145] In this embodiment, the maximum design clamping force of the vehicle's electromechanical brake is first defined as... Define the transition period , Indicates from Beginning, then The target clamping force is updated for each control cycle. The minimum number of transition cycles is The maximum number of transition cycles is The clamping force target value in the previous control cycle was .

[0146] Establish using linear mapping method and The quantitative relationship ensures Follow Linearly increasing:

[0147] ;

[0148] in, The clamping force offset is calculated from this. It may be a non-integer, if For non-integer values, round up to the nearest integer. Integerization.

[0149] Due to the transition period Follow Adaptive change, therefore the clamping force changes gradient Synchronous adjustments are required to ensure the linear completion of the total transition amount:

[0150] ;

[0151] The above formula ensures that the target clamping force is from arrive exist Distribute evenly within each transition period. Determine the transition period. Gradient update with clamping force Then, initiate a smooth rollback:

[0152] Step S1: Set the counter The initial value of the current control cycle ;

[0153] Step S2: From start( ), which is the first control cycle to initiate rollback mode, then the first The target clamping force for each control cycle is ;

[0154] S3, when the counter At this point, the transition is complete, meaning that in this control cycle, the target clamping force is finally updated to... .

[0155] Before initiating rollback Each control cycle has a target clamping force command that is reduced compared to the previous cycle, and the actual clamping force during the retraction process is still controlled by the dual fuzzy sliding mode controller to track the target value.

[0156] During the smooth retraction process or at the end of the current control cycle's smooth retraction, the target clamping force applied by the driver is... The situation may change; the target clamping force command may increase, or it may decrease further, even becoming zero. If the monitored target clamping force... If the braking force increases, the system exits the smooth retraction mode and switches to the clamping force following braking mode, immediately responding to the intention to increase the braking force and directly updating the target clamping force to the increased target clamping force request; if the target clamping force... If the clamping force is further reduced (meaning reduced relative to the previous control cycle), the current mode remains a smooth rollback. The adaptive variable cycle number smooth rollback strategy is re-emerged to smoothly reduce the clamping force until it reaches the further reduced target clamping force. Then, the system returns to determine whether the target clamping force has decreased or become 0. If, during the smooth rollback process, the monitored target clamping force... If it becomes 0, it will switch from smooth rollback mode to fast rollback mode.

[0157] like Figure 2 As shown, a complete braking process is provided. Based on monitoring whether the target clamping force applied by the driver is 0, it is determined whether to start braking. That is, when the driver presses the brake pedal, if the target clamping force applied by the driver is greater than 0, braking will start; otherwise, braking will not start.

[0158] Once braking begins, the following modes are applied sequentially: free travel braking mode, clamping force following braking mode, and retraction mode.

[0159] When entering the no-travel braking mode, zero-force servo control braking is adopted, and micro-pulse excitation and equivalent stiffness are used to determine whether the friction pad and brake disc have reached the contact proximity point. If they have, the clamping force following braking mode is entered. If they have not, zero-force servo control braking is adopted again and the contact proximity point between the friction pad and brake disc is determined.

[0160] Once the clamping force following braking mode is entered, a dual fuzzy adaptive sliding mode control method is used to track the target clamping force so that the brake outputs the actual clamping force. During the tracking of the target clamping force, the target clamping force is constantly monitored, and when the target clamping force becomes 0 or decreases, the retraction mode is entered.

[0161] When entering the retraction mode, the system determines whether to use a smooth retraction mode or a rapid retraction mode based on the changes in clamping force. If in rapid retraction mode, the maximum torque control algorithm is used to release the brake. If in smooth retraction mode, the system calculates the smooth retraction period and the gradient of clamping force change based on the target clamping force of the current control cycle, and gradually reduces the actual clamping force based on the clamping force change gradient. During the operation of the retraction mode, the system continuously monitors whether the target clamping force increases. If it increases, the system directly returns to the clamping force following braking mode. If the target clamping force suddenly disappears to 0 during the smooth retraction mode, the system directly transitions from smooth retraction to rapid retraction mode.

[0162] Example 2:

[0163] The present invention provides a computer device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, it implements the steps of the multi-mode hierarchical control method for the electromechanical actuator described above.

[0164] For a more detailed explanation of the above method, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0165] Example 3:

[0166] The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, it implements the steps of the multi-mode hierarchical control method for the electromechanical actuator described above.

[0167] For a more detailed explanation of the above method, please refer to the relevant content disclosed in the foregoing embodiments, which will not be repeated here.

[0168] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The devices and storage media disclosed in the embodiments are described simply because they correspond to the methods disclosed in the embodiments; relevant parts can be referred to the method section.

[0169] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0170] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A multi-mode hierarchical control method for electromechanical brakes, characterized in that, include: The system continuously monitors the target clamping force applied by the driver and the actual clamping force output by the brake, and determines the current braking mode, which includes: free travel braking mode, clamping force following braking mode, and retraction mode. When in the no-travel braking mode, the zero-force servo control method is used for braking. The friction pads and brake discs are judged to be close to the contact point by micro-pulse excitation and equivalent stiffness. If they are, the clamping force following braking mode is entered; otherwise, the no-travel braking mode continues. When in clamping force following braking mode, based on the EMB dynamics model, a dual fuzzy adaptive sliding mode control method is used to track the target clamping force so that the brake outputs the actual clamping force. When in retraction mode, the intention to retract is identified based on changes in the target clamping force to release the brake; The determination of the current braking mode is as follows: when the target clamping force is not greater than 0, no braking operation is performed; when the target clamping force is greater than 0, it is determined whether the actual clamping force output by the brake is 0. If the actual clamping force output by the brake is 0, the no-travel braking mode is entered; if the actual clamping force is not 0, it is determined whether the target clamping force has decreased. If it has decreased, the retraction mode is entered; otherwise, the clamping force following braking mode is entered. The zero-force servo control method for braking specifically involves: under zero-force constraints, the target torque of the brake motor in the no-stroke braking mode is obtained using the following formula. ; The target torque of the brake motor The solution formula is: ； The zero-force constraint is: ; in, and The first The frictional torque and viscosity coefficient are estimated by the observer updated at the sampling time. and The first The angular velocity and angular displacement of the motor rotor measured at the sampling time. For the amplitude limiting function, For position servo gain, This is the prior value of the contact angle. This is a time-invariant constant, specifically the torque limiting during the positioning phase. The upper limit of near-zero force is set; The method of determining whether the friction pad and brake disc have reached the point of near contact through micropulse excitation and equivalent stiffness is as follows: In the At the sampling time, for the brake motor The brake motor is obtained by superimposing the shaft current with a micropulse excitation. The shaft current command is as follows: ； in, In the first The brake motor at the sampling time Shaft current command, This is the baseline current during the process of the friction pad approaching the brake disc. The set micro-pulse current amplitude, For symbolic functions, The frequency of the micropulse current. The sampling period of the brake motor. ,but Take 1, ,but Take 0, ,but Take -1; Within the defined sliding window, the amplitude of the angular displacement fluctuation caused by the superimposed micropulse excitation is defined. and torque fluctuation amplitude And calculate the equivalent contact stiffness. and torque change rate ; ； ； ； ； in, and These represent the maximum and minimum angular displacements within the time period corresponding to the sliding window. Indicates the definition symbol, The length of the defined sliding window, Indicates the time period Inside, This is the torque coefficient of the brake motor. For the first The torque of the brake motor at the sampling moment; If the following conditions are met during the duration: Then it enters the clamping force following braking mode; in, , This represents the median rate of change of torque within the time period corresponding to the sliding window. The first time period corresponding to the sliding window The rate of change of torque, A function that takes the median value. To achieve an adaptive threshold, it follows the angular velocity of the motor rotor. Adjustment, For stiffness threshold, The logical AND operator.

2. The multi-mode hierarchical control method for an electromechanical brake according to claim 1, characterized in that, The method based on the EMB dynamics model, employing a dual-fuzzy adaptive sliding mode control to track the target clamping force, enables the brake to output the actual clamping force. Specifically: Based on the EMB dynamics model, a linear sliding surface and a control law are defined. The control law includes an equivalent control term and a switching control term. The switching control term introduces a saturation function with respect to the boundary layer thickness. Using clamping force deviation and clamping force deviation change rate as input variables, the first fuzzy controller performs fuzzy inference and defuzzification through a preset fuzzy rule table to obtain the change in sliding mode gain, and updates the sliding mode gain based on the set initial gain. Using the actual clamping force change rate as the input variable, the second fuzzy controller performs fuzzy inference and defuzzification through a preset fuzzy rule table to obtain the change in boundary layer thickness, and updates the boundary layer thickness based on the set initial boundary layer thickness. Based on the updated sliding mode gain and boundary layer thickness, a sliding mode control method is used to track the target clamping force so that the brake can output the actual clamping force.

3. The multi-mode hierarchical control method for an electromechanical brake according to claim 2, characterized in that, The fuzzy sets corresponding to the input and output variables of the two fuzzy controllers are: ; The triangular membership functions for the input and output variables of both fuzzy controllers are: ； ； ； ； ； ； ； in, These are the input or output variables of the fuzzy controller. , , , , , , , , It is the universe of discourse of the input or output variables of the fuzzy controller. , , , , , and These are the input variables or output variables of the fuzzy controller. Belongs to negative large fuzzy set Membership degree, belonging to negative fuzzy set Membership degree, belonging to negative small fuzzy set Membership degree, belonging to zero fuzzy set Membership degree, belonging to positive small fuzzy set Membership degree, belonging to the central fuzzy set Membership degree and belonging to the large fuzzy set The degree of membership.

4. The multi-mode hierarchical control method for an electromechanical brake according to claim 2, characterized in that, When the first fuzzy controller obtains the change in sliding mode gain after fuzzy inference and defuzzification using a preset fuzzy rule table, it uses a product method to determine the activation intensity and outputs the change in sliding mode gain after defuzzification according to the following formula. ; ； in, For the first The activation strength of the rule, For effective fuzzy rules, This represents the change in sliding mode gain. Let the universe of discourse be the variable of sliding mode gain. In the first Under this rule, the change in sliding mode gain Belongs to fuzzy set The membership degree corresponding to time, fuzzy set It is one of the following: negative large fuzzy set, negative medium fuzzy set, negative small fuzzy set, zero fuzzy set, positive small fuzzy set, positive medium fuzzy set, and positive large fuzzy set. Indicates to Integrate the points.

5. The multi-mode hierarchical control method for an electromechanical brake according to claim 1, characterized in that, The retraction mode includes a rapid retraction mode and a smooth retraction mode. Specifically, the retraction intention is identified based on changes in the target clamping force to release the brake. like In this mode, the vehicle enters rapid reversal mode and uses a maximum torque control algorithm to release the brakes; among which, The target clamping force for the previous control cycle. The target clamping force for the current control cycle. To represent the actual clamping force before the brake is released, The logical AND operator; like If it is in smooth retraction mode, the smooth retraction cycle and the change gradient of clamping force are calculated using the following formula based on the target clamping force of the current control cycle, and the actual clamping force is gradually reduced based on the change gradient of clamping force. ； ； ； in, To smooth out the rollback cycle, and These are the maximum and minimum set smooth rollback periods, respectively. This is the clamping force offset. This is the maximum design clamping force of the brake. This represents the gradient of the clamping force during the current control cycle. This represents the actual clamping force output during the current control cycle.

6. A computer device, characterized in that, It includes a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the steps of the multi-mode hierarchical control method for the electromechanical actuator according to any one of claims 1-5.

7. A computer-readable storage medium, characterized in that, Used to store computer programs; when the computer programs are executed by a processor, they implement the steps of the multi-mode hierarchical control method for the electromechanical brake as described in any one of claims 1-5.

Images