An electromechanical brake system based on heterogeneous motor redundancy and a control method thereof

Abstract

This invention discloses an electromechanical braking system and its control method based on heterogeneous motor redundancy, relating to the field of brake-by-wire technology for new energy vehicles. The system includes the following steps: During vehicle braking, a permanent magnet synchronous motor (PMSM) acts as the main drive source, outputting braking torque, while a switched reluctance motor (SRM) operates in standby or low-power auxiliary mode, forming a heterogeneous motor redundant drive structure. This invention achieves complementary failure modes and reduces the risk of common-mode failures by employing a heterogeneous redundancy structure of a PMSM and a SRM; simultaneously, it optimizes system economy by utilizing the low cost and low standby energy consumption characteristics of the SRM. A deep learning health prediction model is introduced to achieve early fault prediction and proactive switching, and a rapid, seamless switching mechanism within 20ms ensures braking continuity and system safety.

Description

Technical Field

[0001] This invention relates to the field of brake-by-wire technology for new energy vehicles, specifically to an electromechanical braking system and its control method based on heterogeneous motor redundancy. Background Technology

[0002] With the development of vehicle electrification and intelligence, traditional hydraulic braking systems are gradually evolving into brake-by-wire systems. Among these, the Electronic Mechanical Brake (EMB) system directly drives the brake actuator via a motor, achieving precise control of braking force. It offers advantages such as fast response, high control accuracy, and ease of integration with autonomous driving systems. Current EMB systems typically employ a single-motor drive or a homogeneous dual-motor redundant structure, such as a permanent magnet synchronous motor (PMSM) or other types of motors, as the drive source to meet basic braking functions and a certain degree of redundancy requirements. However, as vehicle functional safety levels (such as ASIL-D in ISO 26262) continue to improve, higher demands are placed on the reliability, fault tolerance, and fault response speed of braking systems under complex operating conditions. The safety of traditional single-motor or homogeneous redundant solutions under extreme conditions is gradually becoming insufficient.

[0003] In existing technologies, single-motor driven electromechanical braking systems have a significant risk of single-point failure. If the drive motor demagnetizes, its windings are damaged, or its control fails, the braking function will be directly lost. Furthermore, homogeneous dual-motor redundant systems, using the same type of motor, exhibit consistent failure mechanisms, making them prone to common-mode failures under similar operating conditions or environments, thus reducing overall system reliability. In addition, existing redundant control strategies often employ passive fault-triggered mechanisms, lacking the ability to predict motor operating states in advance. Switching typically occurs only after a fault occurs, resulting in long switching response times, fluctuations or even interruptions in braking force output, making it difficult to meet the demands of high-dynamic braking scenarios. Moreover, existing systems still have shortcomings in multi-motor cooperative control, dynamic braking force distribution, and adaptive control strategies under different operating conditions, making it difficult to balance system safety and braking performance.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an electromechanical braking system and its control method based on heterogeneous motor redundancy, so as to solve the problems in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an electromechanical braking control method based on heterogeneous motor redundancy, comprising the following steps: During vehicle braking, the permanent magnet synchronous motor outputs braking torque as the main drive source, while the switched reluctance motor is in standby or low-power auxiliary operation state to form a heterogeneous motor redundant drive structure. The system acquires current signals, temperature change data, and vibration characteristic information during the operation of a permanent magnet synchronous motor, inputs the acquired data into a health prediction model for state assessment, and outputs the corresponding failure probability. When the failure probability reaches the first threshold, the electromagnetic clutch is triggered and the switched reluctance motor is driven to enter the pre-start operation state, which includes the speed increase and torque establishment process. When the failure probability reaches the second threshold or the permanent magnet synchronous motor malfunctions, the braking torque request is transferred from the permanent magnet synchronous motor to the switched reluctance motor within a preset time range, and the switched reluctance motor continuously outputs braking torque, wherein the preset time is within 20ms. After completing the braking torque transfer, a fault warning message is generated and sent to the vehicle display terminal. At the same time, speed limit control is implemented on the vehicle's operating status to ensure braking safety.

[0007] Preferably, a unified data input structure is constructed to support the health assessment process, focusing on the multi-source acquisition and feature representation of motor operating status information. The steps are as follows: The phase current signal during motor operation is acquired and discretely sampled. Frequency domain transformation is performed on the sampled signal to separate each harmonic component and extract the corresponding amplitude parameters. Temperature data of the motor windings and casing are collected and formed into a continuous temperature sequence. The temperature change rate curve is obtained by differential calculation of adjacent sampling points. Vibration signals during motor operation are collected and spectral analysis is performed to extract energy distribution data within a preset frequency band to form a vibration feature vector; The current harmonic characteristics, temperature change rate, and vibration spectrum data are time-aligned and encapsulated to form a multidimensional input matrix, which is then transmitted to the health prediction model for state calculation.

[0008] Preferably, taking into account the characteristics of time series data processing, the input format of the health prediction model is organized and expressed to enhance the ability to characterize state change trends. The steps are as follows: Establish a continuous sampling data buffer and store multi-source monitoring data in chronological order to form a sequence data set covering a preset time window; Extract the current harmonic amplitude variation trajectory from the buffer and construct a dynamic feature sequence according to the time dimension; The temperature change rate within the corresponding time range is serialized and a trend vector is generated, while maintaining data synchronization. The vibration spectrum features are concatenated with the above sequence data to form a unified input tensor, which is then input into a long short-term memory network to complete the failure probability calculation.

[0009] Preferably, based on the failure probability change process, a secondary motor intervention mechanism is constructed to improve the system's response continuity during the risk phase. The steps are as follows: Receive the failure probability value output by the health prediction model and continuously update the change status of the value, and compare it with the first threshold. When the comparison result meets the preset conditions, a pre-start trigger signal is generated and sent to the drive control unit; The auxiliary motor drive circuit is controlled to enter the operation preparation mode and execute the speed establishment process, so that the speed gradually increases to the preset value; Maintain the current driving state of the main motor, while applying a low-amplitude driving current to the auxiliary motor to form a basic torque output state.

[0010] Preferably, based on the pre-start trigger signal output process, the control method of the auxiliary motor operation preparation stage is further refined. After receiving the pre-start trigger signal, the auxiliary motor is driven into a controlled operation state and a target speed is established. At the same time, the drive current amplitude is adjusted according to the change of failure probability to form a gradual torque output, and the drive state coordination relationship between the main motor and the auxiliary motor is maintained to complete the intervention process.

[0011] Preferably, a tiered control logic is established to differentiate between early warning and takeover actions based on the differences in operational status under different risk ranges. The steps are as follows: Set a first threshold and a second threshold and define the numerical range relationship between them, while dividing the failure probability into multiple segments; Receive the output results of the health prediction model in real time and determine the range of the current value; When the value is between the first threshold and the second threshold, a pre-run control command is output and the auxiliary motor is driven into the preparation state. When the value exceeds the second threshold, a drive switching control command is triggered and a control transfer operation between the main motor and the auxiliary motor is executed.

[0012] Preferably, considering the dynamic characteristics during the drive source switching process, the torque transfer and time control are designed holistically to maintain output stability, as follows: Identify the main motor's operating status information and determine whether the switching trigger conditions are met, while generating a switching flag signal; A braking torque transfer command is generated and sent to the auxiliary motor control channel via the control bus to initiate drive response; During the switching process, the output current of the main motor is gradually reduced while the drive current of the auxiliary motor is simultaneously increased to complete the torque handover process. The entire switching process is controlled by time counting to ensure that the drive switching action is completed within a preset time range and the main motor drive state ends.

[0013] Preferably, a fault information generation and transmission mechanism is constructed to improve driving perception capabilities, focusing on the status feedback requirements after drive takeover. The steps are as follows: Detect the output status of the auxiliary motor and confirm that it has undertaken the task of outputting braking torque, and record the abnormal status of the main motor. The abnormal information of the main motor is encoded and processed to generate standardized fault prompt data; The fault indication data is sent to the instrument display terminal through the vehicle communication network, and the data transmission process is completed. The instrument display interface is driven to perform information update operations and display corresponding fault prompts.

[0014] Preferably, to address the vehicle's safety requirements under redundant operating conditions, an operational adjustment strategy is implemented to reduce system load risk. The specific steps are as follows: The system detects the operating status of the drive system and confirms that it has entered the redundant drive mode, while also acquiring the current vehicle operating status information. The vehicle speed limit control signal is generated and sent to the vehicle control unit via the communication interface; Adjusting the powertrain control parameters alters acceleration response characteristics and limits the output torque range; Maintain the vehicle's operating status within the set speed range and continuously execute the vehicle speed limit control logic.

[0015] An electromechanical braking system based on heterogeneous motor redundancy includes a brake pedal unit, an electronic control unit, and a brake actuator assembly. Brake pedal unit, used to generate brake request signal; The electronic control unit is used to receive braking request signals and output control commands; And brake actuator assemblies located at each wheel, used to output braking torque according to control commands; The brake actuator assembly includes: The main motor is a permanent magnet synchronous motor, used to provide braking drive force under normal operating conditions; The auxiliary motor, a switched reluctance motor, is connected to the main motor via a planetary gear set or clutch through the same lead screw transmission mechanism, and is used to provide braking drive force under redundant operating conditions. The health monitoring module includes current sensors, temperature sensors, and vibration sensors installed on the main motor and auxiliary motor, used to collect motor operating status data; The health status assessment unit, located in the electronic control unit, is used to receive operational status data and perform health status assessments. The health status assessment unit has a built-in health prediction model based on long short-term memory network. The health prediction model takes the motor current harmonic characteristics, temperature change rate and vibration spectrum as inputs, and outputs the predicted value of the motor's remaining life and failure probability. The mode switching unit is used to control the switching of the working modes between the main motor and the auxiliary motor according to the failure probability.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention employs a heterogeneous redundant structure using a permanent magnet synchronous motor (PMSM) and a switched reluctance motor (SRM), creating significant differences in their electromagnetic structures and operating mechanisms, thus exhibiting complementary characteristics in failure modes. PMSMs primarily rely on permanent magnets to provide the magnetic field, with typical failure modes manifesting as permanent magnet demagnetization or magnetic performance decay. SRMs, on the other hand, rely on winding excitation to form a magnetic flux path, with failures primarily concentrated in winding faults or drive anomalies. Because the failure mechanisms of the two types of motors are not identical, similar failures are unlikely to occur simultaneously under the same operating conditions or environmental conditions, fundamentally reducing the risk of common-mode failure. This achieves true heterogeneous fault tolerance, significantly improving the reliability and safety redundancy level of the braking system under complex operating conditions.

[0017] This invention employs a switched reluctance motor as the secondary drive source in redundant motor selection. This motor structure eliminates the need for permanent magnets to establish the magnetic field and requires no continuous power supply to maintain the magnetic field in standby mode, thus achieving near-zero energy consumption operation and effectively reducing energy loss in redundant mode. Furthermore, compared to the dependence of permanent magnet synchronous motors on rare-earth materials, switched reluctance motors have significant advantages in material and manufacturing costs, contributing to a reduction in the overall cost of the braking system. In addition, during long-term operation, the secondary motor experiences less thermal load and mechanical wear due to its standby or low-load state, which helps extend its service life, thereby ensuring system reliability while also considering economic efficiency and energy performance.

[0018] This invention introduces a deep learning-based health prediction model to continuously analyze and predict the operating status of motors, thus transforming control from traditional passive response to active predictive control. This model comprehensively utilizes multi-source data, including current harmonic characteristics, temperature change rate, and vibration spectrum, to identify potential motor failures in advance and output quantified failure probability indicators, providing a basis for control strategy decisions. In this way, redundancy mechanisms can be triggered before a fault occurs, allowing the auxiliary motor to enter a ready state in advance, avoiding braking function interruption due to sudden failures. This predictive drive mechanism significantly improves the system's response capability to abnormal operating conditions and enhances the overall control's foresight and stability.

[0019] This invention optimizes the control strategy to complete the drive source switching process within 20 milliseconds. This response time is far below the fault-tolerant response requirements of relevant functional safety standards, ensuring that braking torque can be quickly and continuously taken over by the auxiliary motor in the event of a malfunction in the main drive motor. In actual operation, because the auxiliary motor has completed pre-start preparation before switching, its speed and torque are close to its operating range, allowing for a smooth transition during the switch and avoiding sudden changes or interruptions in braking force. This rapid switching mechanism makes the driver almost imperceptible to changes in the drive source during operation, thus maintaining a good driving experience while ensuring driving safety, demonstrating high system response performance and engineering application value. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0021] Figure 1 This is a flowchart of an electromechanical braking control method based on heterogeneous motor redundancy according to the present invention.

[0022] Figure 2 This is a schematic diagram of a module of an electromechanical braking system based on heterogeneous motor redundancy according to the present invention. Detailed Implementation

[0023] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0024] This invention provides, for example Figure 1 The electromechanical braking control method based on heterogeneous motor redundancy, as shown, includes the following steps: During the operation of an electromechanical braking system, conventional operating conditions primarily correspond to scenarios where the vehicle is in normal driving condition and braking demand is within a predictable range, such as constant speed driving on urban roads, slight deceleration, and routine braking operations. Under these conditions, the permanent magnet synchronous motor (PMSM) serves as the primary drive source, responsible for outputting braking torque. It converts the motor's rotational motion into linear displacement of the brake actuator via a lead screw transmission mechanism, thereby driving the braking friction pair to generate braking force. Due to its fast response speed, high control precision, and high efficiency, the PMSM can precisely adjust the braking force under conventional operating conditions, meeting the driver's response requirements to the brake pedal input and ensuring the smoothness and stability of the braking process.

[0025] In this operating state, the switched reluctance motor does not participate in the main braking torque output, but is in standby or auxiliary operation. In standby mode, the switched reluctance motor maintains low-energy-consumption operation characteristics, with its current input at an extremely low level or even close to zero power consumption, thereby avoiding additional energy consumption and reducing the overall energy consumption level of the system. At the same time, when the control strategy allows, the switched reluctance motor can also enter auxiliary operation mode, that is, provide a small torque output under specific operating conditions to compensate for the insufficient transient response of the main drive motor under certain dynamic changes, such as when the braking force changes rapidly or the brake pedal input changes abruptly. The auxiliary torque participates in achieving a smooth transition of braking force, thereby further improving the system response performance.

[0026] This division of labor between the main motor and the auxiliary motor constitutes a typical heterogeneous motor redundant drive mode. The permanent magnet synchronous motor (PMSM) is responsible for the main functions, while the switched reluctance motor (SRM) plays a redundant support role. Due to fundamental differences in their electromagnetic structure and operating principles—PMSM relies on permanent magnets to generate a magnetic field, while SRM relies on electromagnetic excitation to form a magnetic flux path—the two motors exhibit significant differences in operating characteristics and failure mechanisms. Maintaining this main-operation + auxiliary-standby strategy under normal operating conditions allows for effective management of redundant resources without increasing control complexity. This ensures efficient execution of the main braking function while avoiding the energy consumption and thermal management problems caused by long-term high-load operation of redundant motors.

[0027] Furthermore, having only the permanent magnet synchronous motor handle braking under normal operating conditions helps reduce system control complexity. The control strategy only needs to revolve around a single motor, thereby reducing the algorithmic complexity and computational burden caused by multi-motor coordinated control and ensuring the real-time performance advantage of the electronic control unit. Simultaneously, since the auxiliary motor does not participate in the main drive, its control channel can remain simplified, maintaining only basic operational readiness capabilities, such as the drive circuit power-on status, the position detection function remaining active, and necessary communication link connections, thus enabling rapid response when subsequent intervention is needed.

[0028] In this operating mode, health monitoring information continues to be collected and updated. Although the auxiliary motor is in standby or auxiliary mode, relevant sensors continue to monitor the operating parameters of both the main and auxiliary motors, including current signals, temperature changes, and vibration characteristics. This data is continuously accumulated in the background, providing basic data support for subsequent health status assessments. Since the main motor undertakes all or most of the braking tasks under normal operating conditions, its operating data is more representative. Therefore, focusing on monitoring the operating status of the main motor helps to identify potential risks in advance.

[0029] Furthermore, this operating strategy also has positive implications for vehicle energy management. Switched reluctance motors do not require a continuous magnetic field to maintain energy in standby mode, significantly reducing static power consumption, while permanent magnet synchronous motors operating within their high-efficiency range further optimize energy utilization. This combination not only improves system reliability but also considers overall vehicle range performance, making it particularly suitable for new energy vehicle applications.

[0030] From a system safety perspective, this mode represents a pre-emptive redundancy state, where redundant resources are available but not fully activated. Compared to completely shutting down redundant motors, this strategy offers greater safety advantages because the secondary motor always maintains the ability to quickly enter a working state, thus rapidly participating in the control process when the primary motor's state changes. Simultaneously, since the secondary motor does not participate in continuous high-load operation, its mechanical wear and thermal load are lower, contributing to extended service life and improved overall system reliability.

[0031] In summary, the control method of using a permanent magnet synchronous motor (PMSM) as the main drive source and a switched reluctance motor (SRM) in standby or auxiliary operation under normal operating conditions not only fully leverages the advantages of PMSM in terms of efficiency and control precision, but also effectively improves system reliability by maintaining the redundancy availability of the SRM. Furthermore, this strategy achieves a balance between control complexity, energy consumption, and system safety, providing a stable operational foundation for subsequent health prediction and mode switching.

[0032] During the operation of the electromechanical braking system, health monitoring functions remain continuously active. Real-time acquisition and analysis of the main motor's operating parameters provide fundamental data support for subsequent control decisions. As the permanent magnet synchronous motor undertakes the primary braking and driving tasks, its operating status directly affects braking performance and system safety. Therefore, high-frequency, multi-dimensional data acquisition of the main motor is of great significance. During health monitoring, current signals, temperature changes, and vibration characteristics, as key physical quantities reflecting the motor's operating status, are continuously collected and processed. These parameters can characterize the motor's electromagnetic, thermal, and mechanical properties from different perspectives.

[0033] The acquisition of current signals primarily reflects the electromagnetic operation of the motor. Under normal operating conditions, the current waveform exhibits relatively stable periodic characteristics, and its amplitude and harmonic components can reflect load changes and internal electromagnetic states. When abnormalities occur inside the motor, such as local demagnetization, winding unevenness, or changes in electromagnetic coupling, the corresponding harmonic components in the current signal will change, especially exhibiting abnormal fluctuations at specific frequency components. Therefore, by continuously acquiring current signals and extracting harmonic characteristics, sensitive indicators of the motor's health status can be formed, providing an important basis for subsequent condition assessment.

[0034] Temperature variation data is also a crucial parameter for assessing the health of a motor. Under continuous braking or high-load operation, the temperature of the motor's internal windings and core gradually increases, and this trend reflects the motor's heat dissipation capacity and internal losses. By collecting temperature variation data and calculating the rate of temperature change, abnormal heating conditions, such as localized overheating or heat accumulation, can be identified. When the rate of temperature change shows an abnormally increasing trend, it usually indicates potential efficiency degradation or localized damage within the motor, which could potentially impact subsequent operation. Therefore, using temperature variation data as one of the input features helps to identify potential failure risks in advance.

[0035] Vibration characteristic information is primarily used to reflect the mechanical structural condition of the motor. During operation, motors may generate vibration signals within different frequency ranges due to rotor imbalance, bearing wear, or structural loosening. These vibration signals typically manifest as energy variations within specific frequency bands. By performing spectral analysis on the vibration signals, the energy distribution of key frequency bands can be extracted. When the motor is in a normal state, the vibration spectrum exhibits a stable distribution; however, when mechanical abnormalities occur, the vibration energy within a specific frequency range will increase significantly. Therefore, by continuously collecting vibration data and extracting spectral characteristics, the health status of the mechanical structure can be effectively reflected.

[0036] After being collected, the aforementioned current, temperature, and vibration data underwent a standardized data processing workflow before being input into the health prediction model. The data processing included basic filtering, feature extraction, and time series processing to ensure the stability and consistency of the input data. Because the motor's operating state exhibits significant time correlation, data from a single point in time cannot fully reflect the overall trend. Therefore, using continuous time series data as input allows for a more accurate depiction of the changes in operating state.

[0037] The health prediction model employs a long short-term memory (LSTM) network structure, which is suitable for processing time-series data and capturing long-term dependencies. At the input layer, current harmonic characteristics, temperature change rate, and vibration spectrum characteristics serve as multi-dimensional input variables, collectively forming the model input vector. Internally, the model updates historical data through multiple layers of memory units, thus forming a comprehensive assessment of the current state. At the output, the model provides a predicted remaining lifespan and failure probability of the motor, characterizing the current operational health status.

[0038] In actual operation, the health prediction model is not executed once, but continuously at fixed time intervals, such as updating the prediction results every certain sampling period, thus forming a dynamically updated health status assessment result. As time goes by, the model continuously receives new input data and updates its internal state, and its output results can reflect the changing trend of the motor's health status, rather than just the instantaneous state. This continuous prediction method helps to identify potential risks in advance and provides a lead time for subsequent control strategies.

[0039] Furthermore, multi-source data fusion plays a crucial role in health assessment. A single parameter often only reflects one aspect of the condition, while current, temperature, and vibration data provide information from different dimensions. Fusion processing can improve the accuracy of assessment results. For example, even if the current characteristics do not show obvious abnormalities, the temperature change rate or vibration spectrum may still indicate potential problems in advance. Comprehensive multi-parameter analysis can avoid the risk of misjudgment caused by the failure of a single indicator.

[0040] The failure probability output by the health prediction model serves as a crucial basis for subsequent control decisions, and its numerical changes reflect the evolution of the motor from a normal to an abnormal state. As operating time increases or operating conditions change, the failure probability may gradually rise. When it reaches a certain range, it indicates that the motor's operating state is approaching the risk boundary. Therefore, by acquiring this failure probability in real time, predictions can be made before a failure occurs, thus providing early intervention conditions for redundant control.

[0041] Through the aforementioned real-time data acquisition and model prediction process, a complete health monitoring and status assessment mechanism is formed. This mechanism not only provides a quantitative description of the motor's current state but also offers future trend predictions, thereby transforming the control strategy from traditional reactive response to proactive prediction. In electromechanical braking systems, this capability is crucial for improving system safety, as braking systems have extremely high requirements for response time and reliability; any delay or failure can pose a safety risk.

[0042] Overall, by continuously collecting the operating parameters of the main motor and inputting them into the health prediction model, dynamic assessment and trend prediction of the motor's health status can be achieved. This process introduces predictive capabilities into the system without changing the original control structure, thereby providing a reliable basis for subsequent redundant control and mode switching, while ensuring that the braking system still has stable operation capabilities and a high safety margin under complex operating conditions.

[0043] During electromechanical braking operation, when the failure probability output by the health prediction model gradually increases and approaches the risk range, early intervention is needed before complete failure occurs to avoid abrupt changes in braking function. This stage corresponds to the main drive motor not yet failing but having entered the performance degradation trend range. By introducing redundant drive capability in advance, a smooth transition and safety assurance can be achieved. The core of this process lies in converting the failure probability into executable control actions and realizing the gradual intervention of the secondary drive source through a mechanical coupling structure and motor control strategy.

[0044] First, the failure probability output by the health prediction model is denoted as... The first threshold is denoted as When satisfied When this occurs, it indicates that the main motor's operating status has deviated from the stable range and has entered a warning stage.

[0045] To avoid torque discontinuity caused by sudden switching, a pre-start intensity factor is needed to describe the degree of auxiliary motor intervention. This factor is defined as: in: It is the pre-start intensity factor, with a value ranging from 0 to 1, used to measure the degree of intervention of the auxiliary motor; This is the current failure probability; It is the first threshold.

[0046] The purpose of this expression is to normalize and map the portion of the failure probability that exceeds the first threshold, so that the degree of intervention of the auxiliary motor continuously increases with the increase of failure risk, rather than being triggered abruptly, thereby achieving gradual control.

[0047] After obtaining the pre-start intensity factor, it is necessary to determine the target operating state of the auxiliary motor. The target speed of the auxiliary motor can be expressed as: in: This is the target speed of the auxiliary motor; This is the current speed of the main motor; It is the pre-start intensity factor.

[0048] The purpose of this formula is to link the speed of the auxiliary motor with the current operating state of the main motor, so that the auxiliary motor gradually approaches the operating state of the main motor during the pre-start-up phase, thereby reducing the impact of the speed difference during the subsequent takeover process.

[0049] At the same time, the pre-start torque of the auxiliary motor needs to be established synchronously, which is expressed as: in: It is the output torque of the auxiliary motor; This is the current output torque of the main motor; It is the pre-start intensity factor.

[0050] The design feature of this formula is the use of a quadratic relationship, which means that when the failure probability just exceeds the first threshold, the auxiliary motor provides only a very small torque, while as the risk increases, the torque growth rate accelerates, thereby achieving the control characteristics of light intervention in the early stage and rapid enhancement in the later stage.

[0051] As the auxiliary motor gradually establishes its speed and torque, it is necessary to control the mechanical coupling structure to enter its working state. The degree of clutch engagement can be represented by a coupling activation coefficient: in: It is the clutch engagement coefficient; This is the target speed of the auxiliary motor; This is the current speed of the main motor; It is a very small positive number, used to avoid the denominator being zero.

[0052] The purpose of this formula is to dynamically adjust the clutch engagement state based on the proximity of the rotational speeds of the auxiliary motor and the main motor. As the rotational speeds of the two motors gradually become equal, the engagement degree gradually increases, thereby reducing mechanical shock.

[0053] Furthermore, to ensure the stability of the entire pre-start process, a torque smoothing transition function needs to be defined to describe the torque distribution relationship between the main motor and the auxiliary motor: in: It is the total braking torque output; It is the torque of the main motor; It is the torque of the auxiliary motor; It is the pre-start intensity factor.

[0054] The purpose of this formula is to gradually reduce the proportion of the main motor during the pre-start phase, while increasing the contribution of the auxiliary motor, so that the overall braking torque remains continuous and stable.

[0055] To further constrain the system's dynamic response, a pre-startup stability criterion needs to be introduced, defined as: in: It is the stability deviation coefficient; This is the actual output torque of the auxiliary motor; It is the torque of the main motor; It is a very small quantity that prevents the denominator from being zero.

[0056] The purpose of this formula is to measure whether the output of the auxiliary motor meets the expected control target. When the deviation is small, it indicates that the pre-start process is stable and the control strategy can be further advanced. If the deviation is large, it is necessary to maintain the current state to avoid switching too quickly.

[0057] In summary, the entire process can be understood as a continuous control chain: First, the failure probability is used to determine whether to enter the pre-start stage; then, the risk level is mapped to a control variable through the pre-start intensity factor; next, the auxiliary motor speed and torque are established based on the control variable; at the same time, the mechanical connection is gradually completed through the coupling activation coefficient; finally, the overall output is smoothly transitioned through the torque distribution relationship, and the reliability of the process is ensured through the stability criterion.

[0058] These interconnected computational relationships form a complete logic from state assessment, control mapping, and execution adjustment, enabling the auxiliary motor to gradually enter a working preparation state before the main motor completely fails, thus providing shock-free conditions for subsequent switching and significantly improving the safety and stability of the system during the fault transition phase.

[0059] During the operation of the electromechanical braking system, when the health prediction results deteriorate further and reach a higher risk range, or when the main drive motor experiences actual functional abnormalities, the system needs to transition from the early warning control phase to the substantive redundancy takeover phase. In this phase, the control objective shifts from advance preparation to functional maintenance, with the core objective being to ensure continuous, stable, and uninterrupted braking torque output, while simultaneously completing the rapid transfer of the drive source from the main motor to the auxiliary motor.

[0060] When the failure probability reaches the second threshold, it means that the main motor's operating state has entered a high-risk range, and continuing to perform braking tasks may lead to a significant performance degradation or even complete failure. On the other hand, in some extreme cases, even if the failure probability has not yet fully reached the threshold, once an actual fault phenomenon is detected, such as abnormal current surges, loss of drive response, or abnormal speed deviations, the switching mechanism still needs to be triggered immediately. Therefore, the control logic considers both predicted result triggering and actual state triggering conditions at this stage to ensure that the system has a rapid response capability.

[0061] After the switching is triggered, the control core shifts to the redirection process of the braking torque request. The braking torque request originates from the driver's input and is converted into a drive execution command by the electronic control unit. Under normal operating conditions, this command is executed by the main motor. When entering the switching phase, this control command needs to be transferred from the main motor channel to the auxiliary motor channel in a very short time, and the auxiliary motor takes over all braking output tasks. This process involves not only the redistribution of control signals but also the dynamic adaptation of the actuators and the stable transition of the mechanical transmission path.

[0062] At the control signal level, the electronic control unit remaps the drive commands originally assigned to the main motor to the auxiliary motor control channel. Since the auxiliary motor has already established its base speed and partially loaded torque during the pre-start phase, it has the ability to immediately output braking force at the time of switching, thus avoiding response delay. This advance preparation mechanism ensures that the switching action no longer relies on starting from zero, but rather on a transition based on the existing operating state, effectively shortening the response time.

[0063] At the mechanical transmission level, the main motor and auxiliary motor are connected to the same lead screw transmission mechanism via planetary gear sets or a clutch. During the switching process, the mechanical structure needs to ensure the continuity of the power transmission path. When the auxiliary motor takes over the driving task, its output torque acts directly on the braking actuator through the existing coupling path, while the main motor output gradually decreases until it exits the driving state. During this process, the mechanical connection remains stable, without disconnection or impact, thus ensuring continuous braking force output.

[0064] Time constraints are a key indicator in this process. The entire switching process is limited to within 20 milliseconds, a timescale far smaller than the driver's perception threshold, while simultaneously meeting the response speed requirements of high-safety-level systems. Within this time window, multiple actions need to be completed, including control signal transfer, torque adjustment, and mechanical output stabilization. Since the auxiliary motor has already established a partial operating state during the pre-start phase, its output capacity can be rapidly increased after the switching is triggered, ensuring that the braking torque curve remains continuously changing rather than abruptly changing.

[0065] During the switching process, there is a brief coordinated transition phase between the main motor and the auxiliary motor. In this phase, the output of the main motor gradually decreases, while the output of the auxiliary motor gradually increases, forming a dynamic handover process. The goal of this process is to ensure that the total braking torque remains near the target value, without significant fluctuations due to changes in the drive source. This gradual handover method effectively avoids braking shocks or sudden deceleration changes, thereby improving the overall vehicle driving stability.

[0066] After taking over the braking task, the auxiliary motor enters the main drive state. Its control strategy remains consistent with the original main motor's operating mode, that is, it outputs the corresponding torque according to the braking request and converts it into braking force through the lead screw mechanism. Since the switched reluctance motor does not rely on permanent magnets in its structure, it has stronger stability under high temperature or complex operating conditions. Therefore, it can provide reliable protection for taking over the braking function after the main motor fails.

[0067] Meanwhile, even after the auxiliary motor has fully taken over, the system continues to monitor its operational status to ensure stability after the takeover process. Although the main motor has been removed from the main control channel, its operating status can still be recorded as auxiliary reference information, thus providing a basis for subsequent system maintenance or fault analysis.

[0068] The key to the entire switching process lies in achieving seamless switching, meaning the driver will not perceive any abnormal changes in braking response during operation. This relies on the synergy of three aspects: first, health prediction provides advance notice, allowing the auxiliary motor to prepare; second, control commands are rapidly redirected, achieving millisecond-level response; and third, the mechanical structure ensures stable power transmission, preventing power interruption. These three elements work together to ensure the braking system maintains normal function even if the primary drive source fails or is about to fail.

[0069] Furthermore, this switching mechanism is adaptable to different operating conditions. While braking torque requirements vary in different scenarios such as low-speed braking, high-speed braking, and emergency braking, the switching logic remains consistent: once the triggering condition is met, the drive source transfer is completed within a limited time. This unified control strategy simplifies system design while ensuring consistent safety performance under various operating environments.

[0070] From an overall perspective, the control strategy at this stage achieves a smooth transition from risk warning to functional takeover. Through advance pre-start, rapid response of the auxiliary motor, and strict time control, the braking system can maintain stable operation when the main motor fails or is close to failure, thereby avoiding the loss of braking performance caused by sudden failures in traditional single-motor structures. Furthermore, the entire process requires no additional complex structures, relying solely on existing heterogeneous motor configurations and control strategies, demonstrating good engineering feasibility.

[0071] In summary, when the failure probability reaches the high-risk range or an actual fault occurs, by transferring the braking torque request to the auxiliary motor within 20 milliseconds and having the auxiliary motor undertake all braking output tasks, the braking function can be ensured to be uninterrupted, with continuous and stable response, thereby significantly improving the safety and reliability of the electromechanical braking system under fault conditions.

[0072] After the electromechanical braking system completes the drive source switch, the system's operating state transitions from normal operation mode to fault-tolerant mode. In this state, although the auxiliary motor has successfully taken over the braking function and can maintain basic braking capability, the overall system reliability boundary has changed. Therefore, it is necessary to further constrain the vehicle's operating state through a combination of information prompts and operational restrictions to reduce potential risks and ensure driving safety.

[0073] After the drive source switch is completed, the electronic control unit immediately enters the fault information processing stage. The core task of this stage is to identify, confirm, and output the current system status externally, enabling the driver to obtain system operation information in a timely manner. The generation of fault indication information is based on the health status assessment results obtained in the previous stage and the drive switch status, mainly including the main motor abnormal status indicator, the redundancy takeover status indicator, and system operation level change information. This information is integrated within the electronic control unit and forms a unified fault indication signal.

[0074] During signal output, the electronic control unit (ECU) sends fault warning information to the vehicle's instrument cluster display terminal via the vehicle communication network. This communication process typically relies on the vehicle's internal bus to ensure that information is transmitted to the display interface in a very short time. In terms of display format, fault warnings can manifest as illuminated warning lights, text messages, or changes in graphic icons, allowing the driver to intuitively identify the current system status. The key message conveys information about a primary drive source malfunction or the intervention of redundant drives, thus reminding the driver that the braking system is currently operating in a degraded state.

[0075] In addition to visual cues, audible cues, such as buzzer alarms or alert tones, can be used in some cases to enhance the effectiveness of information delivery. In complex driving environments, single visual information may be overlooked, while multi-channel cues can improve the driver's perception of changes in system status, thereby prompting them to take appropriate driving strategies, such as reducing speed or increasing following distance.

[0076] Upon outputting the fault indication, the electronic control unit enters the operational limitation control phase. Since the system is already in redundant operation mode, although the auxiliary motor can maintain braking function, the overall system's fault tolerance has decreased. Continuing to operate at the original performance limits could trigger risks again in extreme situations. Therefore, it is necessary to constrain the vehicle's operating state, and the most direct and effective way is to limit the vehicle's maximum speed.

[0077] The speed limiting strategy is implemented by sending constraint signals to the vehicle's powertrain control system. The electronic control unit generates a speed limit command based on the current system state and transmits it to the vehicle control unit or powertrain control module via the in-vehicle network, thereby limiting the output capability of the drive system. This limitation does not directly control the braking system itself, but rather restricts the vehicle's acceleration capability, allowing the vehicle to operate within a safer speed range, thus reducing the performance requirements on the braking system.

[0078] In practice, speed limits typically manifest as reduced power output or decreased acceleration response. For example, when the driver presses the accelerator pedal, the powertrain no longer fully responds to the input request but instead outputs controlled torque according to the limiting strategy, preventing the vehicle from reaching higher speeds. This method effectively controls operational risks without affecting the vehicle's basic driving capabilities.

[0079] Furthermore, speed limits indirectly improve braking safety. At lower speeds, braking distances are shorter, reducing the required output torque of the braking system and thus alleviating the load on the auxiliary motor, thereby improving system stability. Simultaneously, even slight changes in braking response at low speeds are less likely to cause serious safety issues; therefore, limiting speed can improve the overall system's fault tolerance.

[0080] Throughout the fault indication and operational limitation process, the electronic control unit continuously monitors the system status. Although the main motor has been identified as abnormal, its status changes are still recorded and updated for subsequent diagnostic and maintenance analysis. Simultaneously, after the auxiliary motor takes over the braking task, its operating parameters are also continuously collected to ensure stable operation in redundant mode. This continuous monitoring mechanism maintains the system's closed-loop control capability even in fault conditions.

[0081] From a driving experience perspective, the control strategy at this stage emphasizes perceptibility without abrupt changes. Drivers can clearly understand the current status through instrument prompts, but the vehicle's operation will not undergo drastic changes, such as sudden deceleration or control failure, thus ensuring a smooth ride. This design strikes a balance between safety and comfort, allowing the system to maintain good controllability even after entering degraded mode.

[0082] Further analysis reveals that the control measures at this stage fall under the degraded operation strategy within the functional safety system. When the system cannot maintain full functionality, the goal of safe operation is achieved by preserving core functions and limiting operating conditions. This strategy is widely used in high-safety-level systems, and its core idea is to improve reliability by reducing performance requirements. In electromechanical braking systems, by having a secondary motor take over braking and coordinating with vehicle speed limits, basic braking capability can be maintained even if the main motor fails, thereby preventing serious safety accidents.

[0083] From the perspective of the overall control process, drive switching, fault indication, and speed limiting form a complete closed loop: first, risks are identified through health assessment; second, redundant drives enable functional takeover; and finally, information prompts and operational restrictions control the spread of risks. This multi-level control mechanism not only improves system robustness but also enhances the driver's awareness of the system's status, making human-machine collaboration more effective.

[0084] In summary, after the drive source switch is completed, the electronic control unit sends a fault warning to the vehicle's instrument panel and limits the vehicle's maximum speed, thus enabling effective management of the system's operating status. This ensures that the driver is promptly informed of the current status and can take appropriate action, while simultaneously reducing system load by limiting vehicle operating conditions, thereby maintaining stable and safe braking capability in redundant operation mode. This strategy, in conjunction with the aforementioned early warning control and rapid switching mechanisms, collectively constructs a comprehensive safety assurance system.

[0085] Example: In this embodiment, the electromechanical braking system employs a heterogeneous motor redundancy structure. The main drive motor is an 800W permanent magnet synchronous motor, while the auxiliary drive motor is a 600W switched reluctance motor. The two motors differ in their electromagnetic structure and operating mechanism. The permanent magnet synchronous motor relies on permanent magnets to provide the excitation magnetic field, while the switched reluctance motor forms a magnetic flux path through winding excitation, thus creating a complementary relationship in terms of failure mechanism. The main and auxiliary motors are connected to the same lead screw transmission mechanism via a planetary gear set. The planetary gear set is used to transmit and superimpose the torque of the two motors, while ensuring the continuity and stability of the power path under different operating conditions. The lead screw mechanism converts the rotational motion into the axial force required for braking, thereby driving the brake to complete the braking action.

[0086] During operation, the health monitoring function continuously collects and updates the motor status. The health monitoring module collects the main motor operating parameters at a fixed sampling period. In this embodiment, the sampling period is set to 10ms, meaning that current signals, temperature data, and vibration information are acquired every 10ms, thus forming continuous time-series data. This sampling frequency can balance data real-time performance and computational load, ensuring that status changes are captured in a timely manner without placing excessive computational pressure on the control unit.

[0087] The acquired current signals are used to extract harmonic features, with a focus on the 5th and 7th harmonic components. This harmonic information reflects changes in the motor's electromagnetic state; the harmonic amplitude will change under abnormal conditions such as demagnetization or winding imbalance. Temperature data is used to calculate the rate of temperature change, i.e., the temperature change per unit time, to reflect the trend of the motor's thermal state. An abnormally high rate of temperature change indicates that the motor's internal losses or heat dissipation conditions may have changed. Vibration data is analyzed using spectrum analysis to extract the energy distribution in the 200Hz to 500Hz frequency band. This band corresponds to the common vibration response range in the motor's mechanical structure and can be used to identify problems such as bearing wear or structural instability.

[0088] The aforementioned current harmonic characteristics, temperature change rate, and vibration spectrum energy are input features into the health prediction model. In this embodiment, the health prediction model is constructed using a three-layer long short-term memory network structure, which can process time-series data and capture the impact of historical state changes on the current state. The model input data is selected from continuous sampling data within the past 5-second time window, which is input into the network in a time-series manner to achieve dynamic evaluation of the motor's operating state. The model output includes the predicted remaining life of the motor and the failure probability, where the failure probability characterizes how close the current operating state is to the failure state.

[0089] The training data for the health prediction model comes from an accelerated aging experiment dataset of electric motors. In this embodiment, a total of 100 sets of experimental data were collected. By simulating the gradual degradation process of the motor under different operating conditions, multiple failure development paths were obtained, thereby improving the model's ability to identify different abnormal situations. After offline training, the model is deployed in the electronic control unit for real-time inference operations.

[0090] Regarding the control strategy, two different failure probability thresholds are set to achieve hierarchical control. The first threshold is set at 30%. When the failure probability reaches this threshold, it indicates that the main motor has entered a potential risk range, at which point the auxiliary motor is triggered to enter a pre-start state. In this embodiment, the auxiliary motor's pre-start state is characterized by increasing its speed to 100 rpm while maintaining low torque output to ensure its ability to participate in driving at any time. During this stage, the main motor still undertakes the primary braking task, and the auxiliary motor exists only as an auxiliary preparation unit.

[0091] The second threshold is set at 80%. When the failure probability reaches this threshold, it indicates that the main motor's operating state is close to the failure boundary, and the system enters a high-risk state. Under this condition, the control logic triggers a drive switching process, transferring the braking torque request from the main motor to the auxiliary motor. In this embodiment, the switching process is limited to completion within 20ms. Through the advance preparation of the pre-start mechanism, the auxiliary motor can increase its output capability in a very short time, thereby ensuring the continuity of braking torque output.

[0092] During the switching process, the output of the main motor gradually decreases while the output of the auxiliary motor gradually increases, creating a brief transition phase that keeps the total braking torque stable and avoids significant fluctuations. Since the auxiliary motor has already established a base speed during the pre-start phase, it does not need to start from a standstill when taking over, thus significantly shortening the response time and reducing impact.

[0093] Bench tests verified that the system can still maintain complete braking function under single motor failure conditions, meaning that the braking efficiency reaches 100%. During the switching process, the fluctuation of braking deceleration was controlled within 0.1g, indicating that the vehicle deceleration changes smoothly during the drive source switching process and will not have a significant impact on driving stability.

[0094] Furthermore, after the switchover is complete, the system enters a redundant operation mode, with the auxiliary motor continuously undertaking the braking task. Simultaneously, the control unit outputs fault warnings and limits vehicle speed to reduce system operational risks. Through this control strategy, the braking function remains available even when the main motor malfunctions or is nearing failure, thereby improving the overall system safety and reliability.

[0095] This invention employs a heterogeneous redundant structure using a permanent magnet synchronous motor (PMSM) and a switched reluctance motor (SRM), creating significant differences in their electromagnetic structures and operating mechanisms, thus exhibiting complementary characteristics in failure modes. PMSMs primarily rely on permanent magnets to provide the magnetic field, with typical failure modes manifesting as permanent magnet demagnetization or magnetic performance decay. SRMs, on the other hand, rely on winding excitation to form a magnetic flux path, with failures primarily concentrated in winding faults or drive anomalies. Because the failure mechanisms of the two types of motors are not identical, similar failures are unlikely to occur simultaneously under the same operating conditions or environmental conditions, fundamentally reducing the risk of common-mode failure. This achieves true heterogeneous fault tolerance, significantly improving the reliability and safety redundancy level of the braking system under complex operating conditions.

[0096] This invention employs a switched reluctance motor as the secondary drive source in redundant motor selection. This motor structure eliminates the need for permanent magnets to establish the magnetic field and requires no continuous power supply to maintain the magnetic field in standby mode, thus achieving near-zero energy consumption operation and effectively reducing energy loss in redundant mode. Furthermore, compared to the dependence of permanent magnet synchronous motors on rare-earth materials, switched reluctance motors have significant advantages in material and manufacturing costs, contributing to a reduction in the overall cost of the braking system. In addition, during long-term operation, the secondary motor experiences less thermal load and mechanical wear due to its standby or low-load state, which helps extend its service life, thereby ensuring system reliability while also considering economic efficiency and energy performance.

[0097] This invention introduces a deep learning-based health prediction model to continuously analyze and predict the operating status of motors, thus transforming control from traditional passive response to active predictive control. This model comprehensively utilizes multi-source data, including current harmonic characteristics, temperature change rate, and vibration spectrum, to identify potential motor failures in advance and output quantified failure probability indicators, providing a basis for control strategy decisions. In this way, redundancy mechanisms can be triggered before a fault occurs, allowing the auxiliary motor to enter a ready state in advance, avoiding braking function interruption due to sudden failures. This predictive drive mechanism significantly improves the system's response capability to abnormal operating conditions and enhances the overall control's foresight and stability.

[0098] This invention optimizes the control strategy to complete the drive source switching process within 20 milliseconds. This response time is far below the fault-tolerant response requirements of relevant functional safety standards, ensuring that braking torque can be quickly and continuously taken over by the auxiliary motor in the event of a malfunction in the main drive motor. In actual operation, because the auxiliary motor has completed pre-start preparation before switching, its speed and torque are close to its operating range, allowing for a smooth transition during the switch and avoiding sudden changes or interruptions in braking force. This rapid switching mechanism makes the driver almost imperceptible to changes in the drive source during operation, thus maintaining a good driving experience while ensuring driving safety, demonstrating high system response performance and engineering application value.

[0099] This invention provides, for example Figure 2 The illustrated electromechanical braking system based on heterogeneous motor redundancy includes a brake pedal unit, an electronic control unit, and a brake actuator assembly. Brake pedal unit, used to generate brake request signal; The electronic control unit is used to receive braking request signals and output control commands; And brake actuator assemblies located at each wheel, used to output braking torque according to control commands; The brake actuator assembly includes: The main motor is a permanent magnet synchronous motor, used to provide braking drive force under normal operating conditions; The auxiliary motor, a switched reluctance motor, is connected to the main motor via a planetary gear set or clutch through the same lead screw transmission mechanism, and is used to provide braking drive force under redundant operating conditions. The health monitoring module includes current sensors, temperature sensors, and vibration sensors installed on the main motor and auxiliary motor, used to collect motor operating status data; The health status assessment unit, located in the electronic control unit, is used to receive operational status data and perform health status assessments. The health status assessment unit has a built-in health prediction model based on long short-term memory network. The health prediction model takes the motor current harmonic characteristics, temperature change rate and vibration spectrum as inputs, and outputs the predicted value of the motor's remaining life and failure probability. The mode switching unit is used to control the switching of the working modes between the main motor and the auxiliary motor according to the failure probability.

[0100] The present invention provides an electromechanical braking control method based on heterogeneous motor redundancy, which is implemented by the above-mentioned electromechanical braking system based on heterogeneous motor redundancy. For details of the specific method and process of the electromechanical braking system based on heterogeneous motor redundancy, please refer to the above-mentioned embodiment of the electromechanical braking control method based on heterogeneous motor redundancy, which will not be repeated here.

[0101] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method of electromechanical brake control based on heterogeneous motor redundancy, characterized in that, Includes the following steps: During vehicle braking, the permanent magnet synchronous motor outputs braking torque as the main drive source, while the switched reluctance motor is in standby or low-power auxiliary operation state to form a heterogeneous motor redundant drive structure. The system acquires current signals, temperature change data, and vibration characteristic information during the operation of a permanent magnet synchronous motor, inputs the acquired data into a health prediction model for state assessment, and outputs the corresponding failure probability. When the failure probability reaches the first threshold, the electromagnetic clutch is triggered and the switched reluctance motor is driven into the pre-start operation state. When the failure probability reaches the second threshold or the permanent magnet synchronous motor malfunctions, the braking torque request is transferred from the permanent magnet synchronous motor to the switched reluctance motor within a preset time range, and the switched reluctance motor continuously outputs braking torque. After completing the braking torque transfer, a fault warning message is generated and sent to the vehicle display terminal, while speed limit control is implemented on the vehicle's operating status.

2. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 1, characterized in that, To support the health assessment process, a unified data input structure is constructed based on the multi-source acquisition and feature representation of motor operating status information. The steps are as follows: The phase current signal during motor operation is acquired and discretely sampled. Frequency domain transformation is performed on the sampled signal to separate each harmonic component and extract the corresponding amplitude parameters. Temperature data of the motor windings and casing are collected and formed into a continuous temperature sequence. The temperature change rate curve is obtained by differential calculation of adjacent sampling points. Vibration signals during motor operation are collected and spectral analysis is performed to extract energy distribution data within a preset frequency band to form a vibration feature vector; The current harmonic characteristics, temperature change rate, and vibration spectrum data are time-aligned and encapsulated to form a multidimensional input matrix, which is then transmitted to the health prediction model for state calculation.

3. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 2, characterized in that, Based on the characteristics of time series data processing, the input format of the health prediction model is organized and expressed in the following steps: Establish a continuous sampling data buffer and store multi-source monitoring data in chronological order to form a sequence data set covering a preset time window; Extract the current harmonic amplitude variation trajectory from the buffer and construct a dynamic feature sequence according to the time dimension; The temperature change rate within the corresponding time range is serialized and a trend vector is generated, while maintaining data synchronization. The vibration spectrum features are concatenated with the above sequence data to form a unified input tensor, which is then input into a long short-term memory network to complete the failure probability calculation.

4. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 3, characterized in that, Based on the failure probability change process, a secondary motor intervention mechanism is constructed to improve the system's response continuity during the risk phase. The steps are as follows: Receive the failure probability value output by the health prediction model and continuously update the change status of the value, and compare it with the first threshold. When the comparison result meets the preset conditions, a pre-start trigger signal is generated and sent to the drive control unit; The auxiliary motor drive circuit is controlled to enter the operation preparation mode and execute the speed establishment process, so that the speed gradually increases to the preset value; Maintain the current driving state of the main motor, while applying a low-amplitude driving current to the auxiliary motor to form a basic torque output state.

5. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 4, characterized in that, Based on the pre-start trigger signal output process, the control method for the auxiliary motor operation preparation stage is further refined. After receiving the pre-start trigger signal, the auxiliary motor is driven into a controlled operating state and a target speed is established. At the same time, the drive current amplitude is adjusted according to the change in failure probability to form a gradual torque output, and the drive state coordination relationship between the main motor and the auxiliary motor is maintained to complete the intervention process.

6. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 4, characterized in that, To address the differences in operational status across different risk ranges, a tiered control logic is established to differentiate between early warning and takeover actions. The steps are as follows: Set a first threshold and a second threshold and define the numerical range relationship between them, while dividing the failure probability into multiple segments; Receive the output results of the health prediction model in real time and determine the range of the current value; When the value is between the first threshold and the second threshold, a pre-run control command is output and the auxiliary motor is driven into the preparation state. When the value exceeds the second threshold, a drive switching control command is triggered and a control transfer operation between the main motor and the auxiliary motor is executed.

7. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 6, characterized in that, Based on the dynamic characteristics during the drive source switching process, a holistic design for torque transfer and time control is implemented to maintain output stability. The steps are as follows: Identify the main motor's operating status information and determine whether the switching trigger conditions are met, while generating a switching flag signal; A braking torque transfer command is generated and sent to the auxiliary motor control channel via the control bus to initiate drive response; During the switching process, the output current of the main motor is gradually reduced while the drive current of the auxiliary motor is simultaneously increased to complete the torque handover process. The entire switching process is controlled by time counting to ensure that the drive switching action is completed within a preset time range and the main motor drive state ends.

8. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 7, characterized in that, To improve driving perception capabilities, a fault information generation and transmission mechanism is constructed based on the status feedback requirements after drive takeover. The steps are as follows: Detect the output status of the auxiliary motor and confirm that it has undertaken the task of outputting braking torque, and record the abnormal status of the main motor. The abnormal information of the main motor is encoded and processed to generate standardized fault prompt data; The fault indication data is sent to the instrument display terminal through the vehicle communication network, and the data transmission process is completed. The instrument display interface is driven to perform information update operations and display corresponding fault prompts.

9. The electromechanical braking control method based on heterogeneous motor redundancy according to claim 8, characterized in that, To address the vehicle safety requirements under redundant operation, operational adjustment strategies are implemented to reduce system load risks. The specific steps are as follows: The system detects the operating status of the drive system and confirms that it has entered the redundant drive mode, while also acquiring the current vehicle operating status information. The vehicle speed limit control signal is generated and sent to the vehicle control unit via the communication interface; Adjusting the powertrain control parameters alters acceleration response characteristics and limits the output torque range; Maintain the vehicle's operating status within the set speed range and continuously execute the vehicle speed limit control logic.

10. An electromechanical braking system based on heterogeneous motor redundancy, used to implement the electromechanical braking control method based on heterogeneous motor redundancy as described in any one of claims 1-9, characterized in that, Includes the brake pedal unit, electronic control unit, and brake actuator assembly: Brake pedal unit, used to generate brake request signal; The electronic control unit is used to receive braking request signals and output control commands; And brake actuator assemblies located at each wheel, used to output braking torque according to control commands; The brake actuator assembly includes: The main motor is a permanent magnet synchronous motor, used to provide braking drive force under normal operating conditions; The auxiliary motor, a switched reluctance motor, is connected to the main motor via a planetary gear set or clutch through the same lead screw transmission mechanism, and is used to provide braking drive force under redundant operating conditions. The health monitoring module includes current sensors, temperature sensors, and vibration sensors installed on the main motor and auxiliary motor, used to collect motor operating status data; The health status assessment unit, located in the electronic control unit, is used to receive operational status data and perform health status assessments. The health status assessment unit has a built-in health prediction model based on long short-term memory network. The health prediction model takes the motor current harmonic characteristics, temperature change rate and vibration spectrum as inputs, and outputs the predicted value of the motor's remaining life and failure probability. The mode switching unit is used to control the switching of the working modes between the main motor and the auxiliary motor according to the failure probability.

Images