A method for multi-dimensional state control and fault hierarchical treatment of rotating machines based on effect space trajectory features

Abstract

The application discloses a kind of based on effect space trajectory feature Rotating machine multi-dimensional state control and fault hierarchical disposal method.This method is based on equilateral triangle effect coordinate system, receives the standardized control signal output by diagnostic system, calls dynamic overflow and quantization truncation mechanism to execute control.For motor and generator, through reconfiguration remaining phase space, balance point is realized to resist and balance smooth derating and fault tolerance;For pure mechanical rotating equipment, through active magnetic bearing or automatic balancing head, mechanical compensation is executed.The application unifies the space control logic of electrical and mechanical rotating equipment, realizes the seamless closed loop from "perception diagnosis" to "execution control", and improves the fault-tolerant operation and active defense capability of high-value rotating equipment in extreme working conditions.

Description

Technical Field

[0001] This invention relates to the fields of rotating equipment condition monitoring, closed-loop control and fault handling technology, and in particular to a method for multi-dimensional condition control, smooth derating and fault classification handling of rotating machines (including electric motors, generators and purely mechanical rotating equipment) based on multi-dimensional spatial feature vectors of an equilateral triangle effect coordinate system. Background Technology

[0002] Current fault control and handling solutions for various rotating equipment such as industrial motors, generators, fans, and compressors suffer from multiple inherent technical shortcomings. Existing technologies do not analyze the spatial imbalance patterns of electromagnetic fields and rotor mechanics based on spatial geometric quantification, making it difficult to achieve flexible, graded control of progressive faults. First, traditional closed-loop control relies solely on one-dimensional time-domain error signals for adjustment, failing to capture the spatial evolution characteristics of three-phase electromagnetic field and rotor mass imbalances. When equipment enters the nascent, developing, and progressively deteriorating stages, the controller continues to operate at full load, continuously accelerating insulation aging and bearing wear, significantly shortening the overall lifespan of the machine, and lacking gradient buffer control methods. Second, existing controllers and protection devices employ binary handling logic, setting only normal operation and emergency shutdown modes. Even slight imbalances or distortions directly cut off power, frequently causing unplanned shutdowns and resulting in significant production losses. Simultaneously, the deep hardware and software coupling between the diagnostic unit and control unit makes misdiagnosis prone to causing unintended shutdowns; conventional alarms only output binary switch alarm signals, lacking structured information on fault location and degradation gradient, leaving the controller without a basis for graded adjustment.

[0003] Commercially available electronic motor protectors rely solely on a single current amplitude threshold to determine faults, making them only suitable for constant power frequency conditions. They are prone to false tripping under variable frequency speed control and variable load conditions. They can only disconnect power after a fault exceeds the limit, lacking early warning of degradation and smooth derating fault-tolerant operation capabilities. These protectors require separate purchase and secondary wiring, increasing the hardware and construction costs of the complete equipment. For purely mechanical rotating equipment such as fans and compressors, traditional vibration control methods can only perform shutdown correction and passive vibration reduction, lacking active balancing compensation methods for full-speed operation. Rotor imbalance faults can only be repaired by shutting down the machine, resulting in high maintenance costs and a significant reduction in effective equipment operating time.

[0004] Conventional industrial controllers only support on / off alarm inputs and are incompatible with multi-dimensional spatial vector diagnostic data, failing to identify fine features such as fault offset location and distortion degradation gradient. Existing control systems lack standardized spatial geometric judgment boundaries, resulting in direct shutdown for minor imbalances and a lack of reliable braking criteria for severe over-limit faults, ultimately failing to balance equipment operational safety and production continuity.

[0005] Existing technologies rely solely on time-domain parameters and single-amplitude thresholds for protection and control, lacking a unified approach to spatial balance control for electrical and mechanical rotating equipment, making it difficult to achieve progressive fault-tolerant operation. To address this, this invention introduces an efficacious spatial geometric analysis system to construct a universally applicable electromechanical hierarchical closed-loop control architecture. Based on spatial trajectory distortion boundaries, it designs multi-level smoothing strategies to achieve decoupling of diagnosis and control, flexible derating fault tolerance in the early stages of faults, and reliable braking during critical faults. Simultaneously, it can replace traditional single-threshold motor protectors, reducing equipment matching and construction costs.

[0006] This application is one of a series of spatial trajectory analysis patents filed on the same day by the same applicant. The entire series of patents is divided into two major sections: the principle family (family 1) and the implementation family (family 2). The entire series of patents uses unified terminology, calculation rules, and data interaction interfaces, and cannot be separated. They constitute a complete technical chain from underlying spatial modeling, quantization mapping rules, feature acquisition, fault diagnosis to hierarchical control.

[0007] Principle family (family 1): Builds a complete set of underlying geometric and quantitative algorithms and feature index extraction methods, including patent 1, patent 2 and patent 3, which are built on a dedicated spatial coordinate system quantization system; The implementation family (family 2) is geared towards engineering implementation and covers three major execution modules: feature extraction hardware acquisition, fault diagnosis, and equipment closed-loop control. It includes patents 4, 5, and this patent 6, and only reuses the geometric rules of the principle family without involving any proprietary theoretical system.

[0008] The patents within the family have clear functional boundaries and complementary technical links: the principle family only completes the construction of the spatial coordinate system (patent 1), the design of hierarchical quantification rules (patent 2), and the method for extracting anti-interference feature indicators (patent 3), while the implementation family further implements the principle. Patent 4 provides a hardware acquisition carrier, and patent 5 only completes the fault classification and judgment and outputs standardized signals, without having the ability to control and execute equipment. The first five patents can only output raw diagnostic data such as fault vectors and fault evolution stages, and lack a practical control scheme to convert diagnostic signals into differentiated equipment adjustment actions.

[0009] Meanwhile, the geometric judgment rules proposed in Patent 1, such as dynamic overflow and quantized truncation of the inscribed circle, are only used for fault degree identification. They do not form a hierarchical control logic, nor do they have a unified spatial balance control method that can be adapted to both electric rotating equipment and 120° distributed sensing pure mechanical rotating equipment. They cannot rely on the same coordinate system to realize motor phase space balance adjustment and rotor online counterweight compensation respectively. Summary of the Invention

[0010] The purpose of this invention is to provide a multi-dimensional state control and fault classification method for rotating machines based on the spatial trajectory characteristics of effector. With a unified spatial control logic, it enables smooth derating and fault-tolerant operation of various rotating machines under progressive faults, and triggers emergency handling based on geometric boundaries during the critical period, thus creating a closed loop of integrated diagnosis and control.

[0011] To achieve the above objectives, the present invention adopts the following technical solution: S1: Multidimensional feature and signal acquisition: Acquire multidimensional physical signals through motion state sensors and extract multidimensional spatial feature vectors; S2: Control preparation signal reception: Receives standardized control preparation signals output by the progressive fault diagnosis system; S3: Multi-dimensional state control and fault classification handling: Invoke the dynamic overflow mechanism and quantization truncation mechanism to execute the corresponding physical control actions; S4: Closed-loop feedback and trajectory reconstruction: By reconstructing the remaining phase space to counteract the equilibrium point or adjusting the mechanical counterweight, the centroid of the spatial trajectory is forced to return to the origin.

[0012] Compared with the prior art, the present invention has the following significant technical effects: Unify the spatial control logic of electrical and mechanical systems to overcome the shortcomings of existing control schemes that separate electromechanical designs: This invention is not limited to a specific machine type. Whether it is an electromagnetic field imbalance of an electric motor or a rotor mass distribution imbalance of a purely mechanical system, it is unified in the equilateral triangular effect coordinate system. Fault-tolerant control is achieved by reconstructing the spatial counterbalancing equilibrium point.

[0013] For electric motors, smooth derating and fault-tolerant operation under progressive stator / rotor degradation can be achieved: Compared with the shortcomings of existing control algorithms that can only passively shut down or cannot effectively suppress torque ripple under progressive faults, this invention realizes smooth derating and fault-tolerant operation of electric motors under progressive stator / rotor degradation by reconstructing the remaining phase space to counteract the balance point, which significantly suppresses torque ripple and overheating risk during the fault development period.

[0014] For generators, this invention optimizes excitation regulation and grid stability control under imbalance conditions: Compared with the shortcomings of existing excitation and grid connection control, which are slow to respond and prone to grid fluctuations during gradual imbalance, this invention, based on strict hierarchical handling of geometric boundaries, enables the generator to fine-tune excitation or limit prime mover power in advance when the internal electromagnetic field gradually deteriorates, and triggers emergency disconnection during the critical period, thereby improving the quality of output power and the stability of the grid.

[0015] For purely mechanical rotating equipment, this invention achieves full-speed online active balancing without the need for shutdown maintenance: Compared with the shortcomings of existing mechanical vibration control that relies on post-event maintenance or passive vibration reduction, this invention maps mechanical motion parameters to the effect space, realizing active electromagnetic compensation and online dynamic counterweight under the imbalance of pure mechanical rotor mass distribution, breaking through the physical bottleneck that traditional mechanical faults must be dealt with by stopping the machine.

[0016] The rigorous hierarchical handling based on geometric boundaries completely breaks the traditional black-and-white control logic. Utilizing the dynamic overflow mechanism described in Patent 1, the physical quantity of the fault-corresponding effect axis is allowed to exceed the limit without being truncated in the early stage of control. The inscribed circle / determination circle built into the coordinate system of Patent 1 is used as the absolute boundary for determining the loss of effective functioning capability of the equipment, realizing a continuous hierarchical handling transition from fine-tuning, hierarchical smooth derating to emergency locking.

[0017] Decoupling and seamless coordination of diagnosis and control: As the execution end, this invention does not repeatedly perform complex diagnostic calculations, but directly receives the structured control preparation signal output by Patent 5, and achieves millisecond-level closed-loop response through hardware-level acceleration, which greatly improves the real-time performance and reliability of the system. Attached Figure Description

[0018] The present invention will now be described in further detail with reference to the accompanying drawings and specific implementation methods.

[0019] Figure 1 This is a schematic diagram of the overall control flow of the present invention; Figure 2 This is a schematic diagram of the closed-loop feedback control principle of the present invention. Detailed Implementation

[0020] 1. Prerequisite Basics 1.1 Overall Structure and Division of Labor for the Series of Patents: This patent belongs to the family of control execution terminals. All underlying geometric and quantitative judgment logic reuses the principles of patents 1 and 2. The feature index extraction method comes from patent 3. The hardware relies on the edge acquisition device of patent 4. The control input source is the output preparation signal of the fault diagnosis system of patent 5. This patent only protects the engineering implementation method of mechatronics hierarchical closed-loop control and fault hierarchical handling of rotating equipment.

[0021] To clearly illustrate the position and synergistic relationship of this patent within the entire series of inventions, it is listed below:

[0022] This patent is the last in the family of patents implementing the principles of spatial electricity in the field of effects. With this, the logical loop from principle to implementation in this patent family is complete. Subsequent patents based on the principles of spatial electricity in the field of effects are referred to as the application family.

[0023] 1.2 Technical call logic and data interaction rules for this layer This patent does not require the independent completion of low-level operations such as spatial mapping, feature extraction, and static centroid calibration. All multi-dimensional spatial feature vectors are preprocessed by the preceding patent. It only receives standardized control quasi-signals to carry out hierarchical control. The calling relationship between each low-level module is defined as follows: The reference steady-state operating condition spatial reference is based on: The equilateral triangle coordinate system of Patent 1 and the layered quantization mapping rule of Patent 2 are uniformly adopted. The inscribed circle and dynamic overflow geometric datum are taken from Patent 1.

[0024] Feature data source: Core four-dimensional space feature vector (|δ) _rel |、ε、η、H _d The calculation is performed using the composite trajectory geometric moment algorithm of patent 3. Hardware carrier: Signal acquisition, synchronous sampling, and hardware geometric acceleration rely on the edge computing device of patent 4; Control input: The standardized control preparation signal is output by the patent 5 fault diagnosis system, and the signal carries the relative centroid offset modulus |δ. _rel | eccentricity ε, petal distortion rate η, spatial topological distortion H _d The four-dimensional spatial feature vector contains three types of control instruction fields: fault spatial location vector, fault evolution stage code, and recommended control mode. This patent does not require repeated calculation of spatial geometric features and directly reads the quantization results preprocessed by Patent 5 to carry out hierarchical treatment.

[0025] For details of the pre-operations such as equipment static centroid calibration, equivalent mapping of electrical / mechanical physical quantities, and sensor matching scheme, please refer to the implementation content of Patent 5. This patent directly reads the pre-processed feature vector through the standard data bus, without repeating complex geometric calculations at the control end, which greatly reduces the local computing power requirement and achieves millisecond-level closed-loop response.

[0026] 1.3 Four-level hierarchical processing logic (based on the geometric boundaries of patents 1 and 2) This solution, based on the dynamic overflow mechanism of Patent 1 and the quantized truncation benchmark of the inscribed circle, divides the process into four standardized treatment intervals. These four intervals correspond to the germination, development, and critical fault stages output by Patent 5. Level 1 Normal Maintenance Zone (corresponding to the budding stage): If the relative centroid offset modulus is |δ _rel | or spatial topological distortion degree H _d The system exhibits a slow, unidirectional drift, which is identified as early-stage progressive degradation. The system maintains the equipment at its rated power, synchronously increases the signal sampling frequency, generates predictive maintenance prompts, and does not perform derating or shutdown operations.

[0027] Level 2 Fine-tuning Zone (corresponding to the early stage of development): The centroid offset rate increases slightly. The system calls the basic balance adjustment logic to carry out small-scale parameter fine-tuning to suppress the continuous expansion of distortion.

[0028] Level 3 derating zone (corresponding to the mid-to-late stage of development): If the eccentricity ε and petal distortion rate η show a periodic and continuous increase, the fault is determined to have entered a rapid development stage. The system calls the dynamic overflow mechanism of Patent 1, allowing the physical quantities of the fault phase to exceed the limit for a short period of time without truncation, and simultaneously performs smooth power derating or active balancing compensation to control the speed of fault deterioration.

[0029] Level 4 Emergency Lockout Zone (corresponding to the critical period): When the high-frequency fluctuation amplitude of the feature vector exceeds the safety threshold, or the trajectory envelope area is less than the reference area of ​​the inscribed circle / judgment circle specified in Patent 1, the equipment is determined to have reached the critical failure boundary, and the quantization cutoff mechanism is immediately activated to execute the equipment emergency lockout or braking protection and terminate the closed-loop adjustment process. 2. Specific Implementation Examples 2.1 Example 1: Application scenarios of large and medium-sized electric motors and generators (closed-loop control and fault-tolerant handling of electrical signals) Taking new energy vehicle drive motors as an example, and large and medium-sized generators as examples, the system directly collects three-phase electrical parameters and receives the control preparation signal output by Patent 5. When Patent 5 determines that the equipment is in the "development period" (such as the deterioration of the stator inter-turn insulation of the motor) and the spatial positioning vector points to phase A, this patent receives a structured control preparation signal containing "fault phase (phase A), evolution stage code (development period), and suggested control mode (smooth derating)". The system calls the dynamic overflow mechanism of Patent 1, allowing the voltage / current of phase A to exceed the limit without being cut off in the initial stage of control. Subsequently, the system executes the "reconstruction of the remaining phase spatial counterbalance point" algorithm, dynamically adjusting the current amplitude and phase of phases B and C through the FOC control algorithm, forcing the centroid of the composite trajectory to return to the origin O, and realizing smooth derating operation with faults. When Patent 5 determines that the device is in a "critical period" (e.g., the trajectory area is less than the area of ​​the inscribed circle defined in Patent 1), a differentiated quantitative cutoff mechanism is triggered according to the device type: For large wind turbines, the system cuts off the inverter PWM output and performs emergency locking and feathering braking within milliseconds to prevent the unit from burning out or running away; For new energy vehicle drive motors, in order to ensure driving safety, the system does not perform instantaneous power cutoff, but immediately forces the output power to be limited to a safe limp-home mode and sends the highest level emergency stop request to the vehicle controller (VCU). After the vehicle has come to a safe stop or has left the high-risk operating condition, the system finally locks off the protection.

[0031] 2.2 Example 2: Application scenario of purely mechanical rotating equipment (mechanical signal closed-loop control and active compensation) This embodiment targets purely mechanical rotating equipment such as centrifugal compressors and industrial fans. The system collects mechanical vibration signals through externally mounted 120° uniformly distributed mechanical motion state sensors and strictly implements the "static centroid calibration subtraction method" to eliminate installation errors. In the electrical scenario defined in Patent 3, the "spatial anchor point" is the current trajectory with system inertia. However, in the purely mechanical scenario of this embodiment, due to the lack of electrical inertial reference, the system redefines the "mechanical spatial anchor point." Specifically, the mechanical spatial anchor point is: the standard reference trajectory point set of the equipment in a healthy steady state or the theoretical geometric axis of the rotor. When extracting multidimensional mechanical features, the system uses this "mechanical spatial anchor point" as a reference to calculate the relative centroid offset and topological distortion of the actual mechanical vibration trajectory relative to the anchor point. This mechanism is equivalent to "using stability as an anchor to filter instability" in the electrical scenario, which can effectively filter out high-frequency background noise from mechanical sensors and base resonance interference, ensuring the absolute authenticity of the extracted mechanical spatial feature vectors (such as centroid drift caused by progressive rotor imbalance). When Patent 5 determines that the equipment is in the "development stage" (such as progressive imbalance of rotor mass distribution, H...), the system can effectively filter out high-frequency background noise from mechanical sensors and base resonance interference, ensuring the absolute authenticity of the extracted mechanical spatial feature vectors (such as centroid drift caused by progressive rotor imbalance). _d (Significantly increased), this patent receives a control preparation signal and sends instructions to the active magnetic bearing controller or automatic balancing head. During full-speed operation, the actuator dynamically generates an electromagnetic compensation force of equal magnitude and opposite direction to the offset vector, counteracting mechanical imbalance in real time. When Patent 5 determines it to be in the "emergence stage" (e.g., uneven scaling on the blades causing slow unidirectional drift of the centroid), this patent outputs a "predictive maintenance" instruction, suggesting that the control system adjust the mechanical counterweight within the planned shutdown window to avoid unplanned shutdowns.

[0032] 2.3 Example 3: A four-level hierarchical handling mechanism based on the geometric boundary of an equilateral triangle This embodiment sets up four levels of geometric boundary handling intervals, which are matched one by one with the fault evolution stages output by Patent 5: Level 1 normal maintenance zone corresponds to the budding stage, Level 2 fine-tuning zone and Level 3 depreciation zone correspond to the development stage, and Level 4 emergency zone corresponds to the critical period.

[0033] Level 1 (Normal Zone): Trajectory eccentricity ε < 0.02, centroid offset < 2%. The system automatically records the characteristics and maintains rated power output.

[0034] Level 2 (Fine-tuning zone): 2% ≤ centroid offset rate < 4%. The system invokes the anti-interference feature of Patent 3 to output a fine-tuning control signal to drive the intelligent commutation or reactive power compensation device.

[0035] Level 3 (Derating Zone): 4% ≤ centroid offset rate < 10%, or the trajectory reaches the dynamic overflow boundary of Patent 1. The system performs smooth derating control, limits new load, and forces the spatial trajectory to return to the safe envelope.

[0036] Level 4 (Emergency Zone): Centroid offset rate ≥ 10%, or trajectory area less than the inscribed circle reference area specified in Patent 1. Relying on the inscribed circle / judgment circle boundary trigger cutoff mechanism of Patent 1, the relay protection device or mechanical brake is directly linked to immediately disconnect the load or perform emergency braking.

[0037] 2.4 Example 4: Mid-to-low-end electronic motor protection and control strategy and system based on this method This embodiment provides a motor protection control method based on effects spatial trajectory characteristics. This method can be deployed in a conventional microcontroller (MCU) or edge gateway, enabling the general controller to replace the function of traditional independent electronic motor protectors.

[0038] The system relies on patent 4 to synchronously acquire three-phase voltage and current data. It calls patent 1 to map electrical quantities to a 120° equilateral triangle coordinate system to generate spatial trajectories, and uses patent 2's voltage potential method to perform layered conversion of electrical quantities. Then, based on the geometric boundaries of the inscribed circle / judgment circle in patent 1, it determines the failure threshold. Patent 3 is used to extract trajectory eccentricity and distortion features, serving as a general low-computing-power configuration for low-to-mid-range motor protectors, saving hardware and software resources. Patent 5 then identifies various faults such as phase loss, locked rotor, three-phase imbalance, overload, undervoltage, and overvoltage. When the fault judgment threshold is reached, this patent outputs a switch signal to cut off the motor contactor. The entire algorithm achieves all the protection functions of a traditional electronic motor protector. The main circuit retains air switches and fuses for short-circuit hardware protection; this solution does not replace this type of hardware.

[0039] Traditional electronic motor protectors rely solely on a single threshold value for three-phase current amplitude to determine faults, making them only suitable for fixed-frequency operating conditions. In variable-frequency speed control and equipment with frequent speed fluctuations, they are prone to false tripping and can only directly disconnect power after a fault exceeds limits, failing to quantify the slow degradation trend of the equipment. This solution encodes voltage and current information into spatial geometric features, decoupling feature extraction from frequency disturbances in a layered manner. This results in higher fault identification accuracy in variable-frequency unsteady-state scenarios and allows for flexible configuration of graded warnings based on trajectory distortion levels or early maintenance alerts. In terms of cost, existing solutions require the purchase of separate motor protectors and associated secondary wiring; this solution reuses existing voltage and current acquisition hardware, relying solely on built-in algorithms to achieve a complete set of protection functions, eliminating the cost of separate protector components and wiring, significantly reducing overall hardware and construction costs. This invention, when used as a low- to mid-range motor protector, not only reduces false alarms but also provides progressive fault prediction, offering better cost-effectiveness.

[0040] The above description, through specific examples, illustrates the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

Claims

1. A method for multi-dimensional state control and fault classification of rotating machines based on the spatial trajectory characteristics of effectors, characterized in that, The rotating machine includes an electric motor, a generator, and purely mechanical rotating equipment; based on an equilateral triangular effect coordinate system containing a centroid O, three effect axes that differ by 120°, and a dynamic spillover mapping mechanism, the following steps are included: S1: Multidimensional Feature and Signal Acquisition: Acquire physical signals and complete the benchmark steady-state static centroid calibration. Extract multidimensional spatial feature vectors from the mapped coordinate system. The signals include the spatial trajectory geometric features of the rotating machine in the equilateral triangular effect coordinate system. When the rotating machine is a motor or generator, the signals are generated based on three-phase electrical parameters. When the rotating machine is a purely mechanical rotating device, the signals are generated based on mechanical motion parameters distributed in 120° space and physical installation deviations are eliminated. All electrical and mechanical acquired signals are uniformly subjected to the benchmark steady-state static centroid calibration. _static Calibration compensation: The static reference offset is completely deducted from the dynamic trajectory. S2: Control preparation signal reception: Receive the standardized control preparation signal output by the fault classification and diagnosis system. The standardized control preparation signal includes the fault spatial location vector, the fault evolution stage code, and the suggested control mode. The fault evolution stage code distinguishes between two types of operating conditions: slow-change progressive faults and instantaneous sudden faults. S3: Multidimensional state control and fault classification handling: Based on the fault evolution stage code, the dynamic overflow mechanism and quantization truncation mechanism of the equilateral triangle effect coordinate system are invoked to execute the corresponding physical control actions; S4: Closed-loop feedback and trajectory reconstruction: After executing control actions in the budding and development stages, the changes in multi-dimensional spatial feature vectors are monitored in real time. By reconstructing the remaining phase space to counteract the equilibrium point or adjusting the mechanical counterweight, the centroid of the spatial trajectory is forced to return to the origin. Critical faults trigger quantization cutoff and emergency braking, terminating the real-time closed-loop adjustment process.

2. The method according to claim 1, characterized in that, In step S1, the static centroid calibration subtraction method recorded by the composite spatial trajectory geometric moment algorithm is adopted. The specific rule is: calculate and record the static centroid O of the trajectory under the reference steady-state operating condition of the equipment. _static In subsequent actual operation, all dynamic centroid offsets are subtracted from the static centroid O. _static This is to eliminate fixed errors caused by physical installation deviations of the sensor.

3. The method according to claim 1, characterized in that, In step S1, the specific physical acquisition rule for obtaining the multidimensional spatial feature vector signal is as follows: the multidimensional physical signal is directly acquired by the motion state sensor configured on the rotating machine and mapped to the equilateral triangle effect coordinate system for local extraction; or the multidimensional spatial feature vector signal pre-calculated by the external diagnostic system or the underlying edge computing gateway is directly received through the communication interface.

4. The method according to claim 1, characterized in that, In step S3, the specific rules for invoking the quantization truncation mechanism are as follows: using the preset inscribed circle / determination circle of the equilateral triangle effect coordinate system of Patent 1 as the minimum effective work benchmark, when the envelope area of ​​the single-cycle spatial trajectory is detected to be less than the area of ​​the inscribed circle / determination circle, the determination system loses the effective work benchmark and directly triggers emergency locking or braking control.

5. The method according to claim 1, characterized in that, In step S4, the specific rule for reconstructing the remaining phase space counterbalance point is as follows: when the fault space positioning vector points to a specific effect axis, the amplitude and phase of the physical quantities corresponding to the remaining effect axes are dynamically adjusted through the underlying control algorithm to force the centroid of the composite trajectory to return to the origin O.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 5.

Images