Quantum fuzzy control method and system for rail transit vehicle door controller

Abstract

The invention relates to the technical field of rail transit vehicle control, in particular to a quantum fuzzy control method and system for a rail transit vehicle door controller, and the method comprises the steps: collecting the position, speed and other four-dimensional operation data of a door body, and carrying out the Kalman filtering and EMC compensation preprocessing; mapping and coding the four-dimensional variable quantum bits, and constructing a quantum superposition state containing disturbance compensation; a membership function is optimized based on quantum probability distribution, and a control instruction is generated through quantum annealing algorithm iterative reasoning; and the brushless motor is driven to execute and form closed-loop feedback. The system is developed based on an FPGA and comprises a data acquisition module, a preprocessing module, a quantum fuzzy calculation module, an instruction execution module and a closed-loop feedback module. According to the invention, the problems of poor dynamic load adaptation, weak anti-interference and the like of the existing gate controller can be solved, and accurate and stable control of the gate controller under dynamic load and strong electromagnetic interference scenes can be realized.

Description

Technical Field

[0001] This invention relates to the field of rail transit vehicle control technology, and in particular to a quantum fuzzy control method and system for rail transit vehicle door controllers, applicable to high-speed rail, subway, light rail vehicle door and platform screen door systems. Background Technology

[0002] For high-speed rail, subway, and light rail vehicle door and platform screen door systems, the rail transit vehicle door controller is a core component ensuring the safe and reliable operation of the vehicle door system. Its control accuracy directly affects the smoothness of door opening and closing, passenger safety, and equipment lifespan. Existing door controllers generally employ traditional PID algorithms or conventional fuzzy control algorithms, which have the following prominent problems: 1) When vehicle doors are in operation, they face dynamic load fluctuations such as changes in sealing resistance and passenger collision interference. Traditional PID algorithms process parameters such as door position, speed, and current serially, resulting in large adjustment response delays, which can easily lead to door opening and closing shocks and shorten the lifespan of motors and mechanical transmission components. 2) Weak electromagnetic interference resistance. The strong electromagnetic noise generated by the rail transit traction system (signal-to-noise ratio as low as 28dB) causes the fixed threshold parameters of conventional fuzzy control to drift, the bit error rate of gate controller communication to increase, and even causes door opening and closing logic disorder. 3) The door controller needs to be adapted to multiple variables such as door position, running speed, drive current, and ambient temperature. Existing algorithms mostly use single-variable independent adjustment, which makes it difficult to balance the efficiency and stability of opening and closing the door, resulting in obvious shaking when the door starts and stops, causing a poor passenger experience. 4) Existing control algorithms lack a rapid emergency adjustment mechanism for abnormal operating conditions such as emergency braking and power fluctuations, resulting in a large delay in emergency door closing response and potential safety hazards.

[0003] Therefore, how to achieve precise and stable control has become an urgent technical problem to be solved, given the special operating conditions of rail transit gate controllers, such as large dynamic load fluctuations, strong electromagnetic interference, and high safety requirements. Summary of the Invention

[0004] The purpose of this invention is to provide a quantum fuzzy control method and system for gate controllers of rail transit vehicles, which solves the problems of poor dynamic load adaptation, weak anti-interference, lack of multi-parameter coordination and insufficient safety redundancy of existing gate controllers. It can realize accurate and stable control of the gate controller under dynamic load and strong electromagnetic interference scenarios, and is applicable to high-speed rail, subway, light rail vehicle door and platform screen door systems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: According to one aspect of the present invention, a quantum fuzzy control method for a gate controller of a rail transit vehicle is provided, comprising the following steps: S1: Real-time acquisition of the gate controller's operating data, including gate position, operating speed, drive current, ambient temperature, and power supply voltage; S2: Perform preprocessing on the running data, including Kalman filtering and EMC anti-interference compensation; S3: Perform four-dimensional variable qubit mapping encoding on the preprocessed gate position, running speed, driving current, and supply voltage to construct a quantum superposition state including perturbation compensation; S4: Optimize the fuzzy membership function based on quantum probability distribution, call quantum fuzzy rules to iteratively execute quantum fuzzy inference, and generate gating control instructions; S5: The brushless motor drive module of the gate controller executes the control command; S6: Collect the gate operation data after the instruction adjustment, and calculate the deviation between the actual state and the target state; if the deviation meets the preset threshold, maintain the current parameters; otherwise, return to step S3 to re-execute quantum mapping encoding and quantum fuzzy inference.

[0006] According to an embodiment of the present invention, in step S2, the filtering equation of the Kalman filter is: , Where A = 0.92, B = 0.08, For filter gain, when signal-to-noise ratio ≥ 30dB =0.75, signal-to-noise ratio <30dB =0.55; This is the estimated state value at the current moment. This is the estimated state value from the previous moment; To control the input quantity, specifically the drive control signal of the gate controller; This represents the observed value at the current moment.

[0007] According to an embodiment of the present invention, in step S3, the specific rule for the qubit mapping encoding is: gate position Pos Mapped to 6 qubits, operating speed Vel mapped to 4 qubits, driving current Cur The power supply voltage Volt is mapped to 4 qubits; the expression for the quantum superposition state is: , Where α and β are quantum probability amplitudes and α² + β² = 1, | Pos >、∣ Vel >、∣ Cur >、|Volt> are the reference quantum states, | Pos ′>、∣ Vel ′>、∣ Cur′>、|Volt'> are perturbation-compensated quantum states.

[0008] According to an embodiment of the present invention, the expression of the fuzzy membership function in step S4 is: , in, , It is an 18-dimensional quantum probability density matrix; The variance of the fuzzy membership function is dynamically adjusted according to the application scenario. When the driving current is ≥10A or the ambient temperature is ≥60℃ / ≤-30℃, the variance is set to 0.25; otherwise, it is set to 0.08.

[0009] According to one embodiment of the present invention, in step S4, the quantum fuzzy rule base includes more than 150 rules, and the core rules include at least: Rule 1: If the door position is ≤500mm, the speed is ≤0.3m / s, the current is ≥10A, and the voltage is ≥120VDC, then the driving voltage increment is +5V; Rule 2: If the door position is ≥1500mm, the speed is ≥0.6m / s, the current is ≤3A, and the voltage is ≤90VDC, then the driving voltage increment is -3V; Rule 3: If a sudden current change of ≥3A indicates a passenger collision has been detected, a stop command will be immediately output and the machine will be finely adjusted in the opposite direction by 20mm.

[0010] According to one embodiment of the present invention, in step S4, the iterative execution of quantum fuzzy inference uses the quantum annealing algorithm for iterative calculation, the initial temperature is set to 9, the temperature decreases by 0.4 for each iteration, and the iteration termination condition is that the rule matching error is ≤0.009.

[0011] According to one embodiment of the present invention, the door position ranges from 0 to 2000 mm, the running speed ranges from 0 to 0.8 m / s, the driving current ranges from 0 to 15 A, the ambient temperature ranges from -40℃ to +85℃, the power supply voltage ranges from DC77V to DC137.5V, and the acquisition period is 5 ms.

[0012] According to one embodiment of the present invention, the interval division of the qubit mapping in step S3 includes: the interval corresponding to each quantum state of the gate position is approximately 31.25 mm, the interval corresponding to each quantum state of the running speed is 0.05 m / s, the interval corresponding to each quantum state of the driving current is 0.9375 A, and the interval corresponding to each quantum state of the supply voltage is 3.78 VDC; the compensation amount of the disturbance compensation quantum state is 8% of the interval of the corresponding variable quantum state.

[0013] According to one embodiment of the present invention, in step S2, the EMC anti-interference compensation coefficient is set according to the noise intensity, and the compensation coefficient is set to 0.13~0.16 when the signal-to-noise ratio is 28dB.

[0014] On the other hand, the present invention also provides a quantum fuzzy control system for a gate controller of a rail transit vehicle, the system comprising a data acquisition module, a preprocessing module, a quantum fuzzy calculation module, an instruction execution module, and a closed-loop feedback module; wherein: The data acquisition module includes a displacement sensor, a Hall speed sensor, a current sensor, a temperature sensor, and a voltage sampling interface, which respectively collect data on the door position, running speed, drive current, ambient temperature, and power supply voltage. The preprocessing module, integrated into the FPGA chip, uses Kalman filtering and EMC anti-interference compensation algorithms to preprocess the acquired data and transmits it to the quantum fuzzy computing module via the MVB bus; The quantum fuzzy computing module, developed based on an FPGA chip, includes a quantum encoding unit, a quantum superposition state construction unit, a quantum fuzzy membership function optimization unit, and a quantum fuzzy inference unit, which are used to realize quantum mapping encoding, superposition state construction, membership function optimization, and control instruction generation. The instruction execution module, including a brushless motor drive module, a CPLD logic control unit, and a power amplifier circuit, is used to receive control instructions and drive the door motor to run. The closed-loop feedback module includes a data feedback interface and a deviation judgment unit, which is used to collect the door's operating status data, judge the deviation, and trigger closed-loop control.

[0015] The quantum fuzzy control method and system for a gate controller of a rail transit vehicle, as disclosed in this invention, have the following advantages compared to existing technologies: 1. Significantly improved dynamic load adaptability. By processing four-dimensional variables in parallel using quantum superposition states, the adjustment response delay is reduced to ≤15ms, which reduces the inrush current when opening and closing the door and the amplitude of door start-stop jitter, effectively ensuring the extended lifespan of the door controller motor and mechanical components.

[0016] 2. Through quantum compensation mechanism and EMC anti-interference compensation, the parameter drift rate is significantly reduced. The adjustment pass rate is ≥99% in a strong interference environment with a signal-to-noise ratio of 25dB, meeting the EU EN50155 Class A certification requirements.

[0017] 3. Based on quantum fuzzy membership function, multi-dimensional variable coordinated regulation is realized, which significantly improves the efficiency of door opening and closing while taking into account stability, thus optimizing the passenger experience.

[0018] 4. Sufficient safety redundancy, emergency response delay under abnormal operating conditions ≤120ms, and safety mechanisms such as passenger collision detection and enhanced emergency door closing are provided.

[0019] 5. Strong hardware compatibility and low modification cost: The technical solution of this invention can be developed on existing gate controller FPGA hardware platforms (such as the Bosman BSM gate controller hardware architecture) without replacing the core hardware, resulting in low modification cost and direct application to upgrade existing projects. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of a quantum fuzzy control method for a gate controller of a rail transit vehicle according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a quantum fuzzy control system for a gate controller of a rail transit vehicle according to an embodiment of the present invention; Figure 3 This is a comparison chart of the univariate membership function curves of embodiments of the present invention; Figure 4 This is a contour plot of the two-dimensional variable membership function according to an embodiment of the present invention; Figure 5 This is the iterative error curve of the quantum annealing algorithm in an embodiment of the present invention. Detailed Implementation

[0021] To facilitate a clear description of the technical solutions in the embodiments of the present invention, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. For example, the first threshold and the second threshold are merely used to distinguish different thresholds and do not limit their order. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" are not necessarily different.

[0022] It should be noted that in this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0023] In this invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one" or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, "at least one of a, b, or c" can represent: a, b, c, a combination of a and b, a combination of a and c, a combination of b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0024] like Figure 1 The diagram shows a flowchart of a quantum fuzzy control method for a gate controller of a rail transit vehicle. The method includes the following steps: S101: Real-time acquisition of gate controller operation data; The operating data includes five core categories: door position (Pos), operating speed (Vel), drive current (Cur), ambient temperature (Temp), and power supply voltage (Volt). The door position ranges from 0 to 2000 mm (corresponding to the door being fully closed to fully open, with 0 mm being fully closed and 2000 mm being fully open), the operating speed is 0 to 0.8 m / s (rated door opening and closing speed is 0.6 m / s), the drive current is 0 to 15 A (rated operating current is 3 to 8 A), the ambient temperature is -40℃ to +85℃, and the power supply voltage is DC77V to DC137.5V.

[0025] The data acquisition hardware includes a displacement sensor, a Hall velocity sensor (sampling frequency 1kHz), a current sensor (bandwidth 10kHz), an NTC temperature sensor, and a voltage sampling interface (integrated with a TVS transient suppression diode for surge protection). The acquisition period is 5ms, and synchronous acquisition is triggered by the timer interrupt of the FPGA chip to ensure data timestamp consistency. The acquired data is transmitted to the preprocessing module via the SPI interface.

[0026] S102: Preprocessing the acquired data by performing Kalman filtering for noise reduction and EMC anti-interference compensation; Kalman filter equation: , The current state estimate (optimal value after filtering) is presented in vector form, taking into account the multi-parameter characteristics of the gate controller: ; The state estimate from the previous moment (filtered result 5ms ago) is compared with... The dimensions are consistent and stored in the FPGA's RAM cache; A is the state transition matrix (coefficients), fixed at 0.92, which characterizes the temporal correlation of the gate parameters (parameter changes are gradual within a short period of 5ms, and the influence of historical states is significant); B is the control input matrix (coefficients), fixed at 0.08, which characterizes the sensitivity of the driving signal to the state adjustment (the gate is an inertial system, and the control input takes effect slowly). For Kalman filter gain, dynamic adjustment strategy: when signal-to-noise ratio ≥ 30dB =0.75 (weak interference, dependent on observation), signal-to-noise ratio <30dB =0.55 (strong interference, dependent on predicted value); The input quantity, i.e. the gate controller drive control signal (such as PWM duty cycle, with a value range of 0~100%), is generated by the previous round of quantum fuzzy inference; The current observation value (raw data collected by the sensor), and... The dimensions are consistent, but noise components are present. The filtering algorithm is implemented using VerilogHDL on the FPGA.

[0027] The compensation coefficient is 0.13~0.16 (ripple amplitude 0.3~0.5V) when the signal-to-noise ratio is 28dB, and the compensation coefficient is 0.08~0.11 (ripple amplitude 0.1~0.3V) when the signal-to-noise ratio is 35dB. Filtered data , This is the filtered data. This is the compensation coefficient.

[0028] S103: Quantum bit mapping encoding; The preprocessed four-dimensional variables (Pos, Vel, Cur, Volt) are encoded using qubit mapping, and the specific encoding rules are shown in Table 1.

[0029] Table 1: Quantum Bit Mapping Rules For example, if the gate position is 320mm, then 320 ÷ 31.25 = 10.24, then round down to 10, resulting in the binary code |01010> (padding with zeros to make 6 bits, the resulting quantum code is |001010>). For example, if the running speed is 0.25m / s, then 0.25÷0.05=5, which is the binary code |0101>, resulting in the quantum code |0101>.

[0030] For the perturbation compensation quantum state, the compensation amount is set to 8% of the corresponding variable quantum state range. For example, the compensation amount of Pos is 31.25mm × 8% ≈ 2.5mm, and the compensation amount of Vel is 0.05m / s × 8% ≈ 0.004m / s. The reference quantum state is fine-tuned through the XOR gate of the FPGA. For example, the compensation state of the reference state |001010> is |001001> (leaving the least significant bit).

[0031] S104: Construction of quantum superposition states; The quantum superposition state is constructed as follows: , The probability amplitude values ​​are: α=0.85 (base quantum state weight) and β=0.52 (perturbation compensation quantum state weight), satisfying α²+β²=1 (0.85²+0.52²≈0.72+0.27=0.99≈1); the superposition state is implemented through the tensor product operation unit of the FPGA, using a parallel logic architecture.

[0032] S105: Based on quantum probability distribution optimization of fuzzy membership function, iterative reasoning through quantum annealing algorithm generates gate control instructions; The quantum fuzzy membership function is defined as: ,in, Let represent the mode length of the quantum superposition state. Since α² + β² = 1, therefore... =1; It is an 18-dimensional quantum probability density matrix (6+4+4+4=18), derived from the sum of the number of qubits in the four-dimensional variables.

[0033] The 18-dimensional matrix is ​​decomposed into Pos submatrix (6-dimensional), Vel submatrix (4-dimensional), Cur submatrix (4-dimensional), and Volt submatrix (4-dimensional) through dimensionality decomposition. Each submatrix follows a Gaussian distribution, such as the Pos submatrix. 1000mm is the middle position of the door.

[0034] Joint probability fusion is achieved by implementing tensor product operations using a multiplier array on an FPGA. The fusion formula is as follows: The simplified expression is: .

[0035] The dynamic variance σ is dynamically adjusted for different scenarios: For normal scenarios (Cur < 10A and -30℃ < Temp < 60℃): σ = 0.08, the function curve is steep, which is suitable for high-precision control; For special scenarios (Cur ≥ 10A or Temp ≥ 60℃ / ≤ -30℃): σ = 0.25, the function curve is flat, which enhances the anti-interference ability.

[0036] like Figure 3 As shown, a comparison chart of the quantum fuzzy membership function curves for a single variable (gate position) is presented.

[0037] The curve is for a typical scenario, corresponding to a drive current <10A and an ambient temperature of -30℃ to 60℃ (no heavy load, non-extreme temperature); the curve shape is a steep Gaussian distribution, with the core peak at 1000mm (middle of the gate), and a membership degree of 1.0; the critical interval... A value of ≥0.8 corresponds to a narrow range of 850~1150mm, demonstrating high-precision control and adapting to the smoothness requirements of conventional door opening and closing. Special scenario curves correspond to drive currents ≥10A (heavy load) or ambient temperatures ≥60℃ / ≤-30℃ (extreme temperatures); the curve shape exhibits a gentle Gaussian distribution, with a peak value also around 1000mm and a membership degree of 1.0; among these, the critical range... A value of ≥0.8 corresponds to a wide range of 600~1400mm. By expanding the effective membership range, it can cope with parameter fluctuations caused by heavy loads or extreme temperatures and avoid frequent adjustment shocks. In normal scenarios, the variance is small (σ=0.08), pursuing accuracy; in special scenarios, the variance is large (σ=0.25), enhancing anti-interference.

[0038] Membership function calculation is implemented through the floating-point unit of the FPGA. The quantum fuzzy rule base contains 156 rules, covering 12 operating conditions including conventional door opening and closing, heavy load, light load, strong interference, and emergency braking. The rule base is stored in the FPGA's ROM, and the core rules (including triggering logic) include: Rule 1: If Pos≤500mm (initial closing), Vel≤0.3m / s, Cur≥10A (high sealing resistance), and Volt≥120VDC, then the driving voltage increment is +5V (enhancing driving force); Triggering condition: all four parameters must be met simultaneously for a duration ≥2 acquisition cycles (10ms). Rule 2: If Pos≥1500mm (end of door opening), Vel≥0.6m / s, Cur≤3A, Volt≤90VDC, then the drive voltage increment is -3V (deceleration buffer); Trigger condition: the first three parameters are met, and the voltage parameter tolerance is ±5VDC; Rule 3: If the Cur sudden change is ≥3A (passenger collision characteristic) and the duration is ≥1 acquisition cycle (5ms), then immediately output a stop command (PWM duty cycle 0%), and output a reverse fine adjustment command 10ms later (drive voltage +2V, lasting 4ms, door moves 20mm in the opposite direction).

[0039] like Figure 4As shown, a two-dimensional quantum fuzzy membership contour plot of gate position versus driving current is presented. The horizontal axis represents the gate position (0~2000mm), and the vertical axis represents the driving current (0~15A), covering the entire operating range of the gate controller; it is divided into four regions according to the membership value. Optimal fit region ( ≥0.8): Corresponding to a position of 700~1300mm and a current of 4~7A, this is the normal stable operating range of the gate controller; slight adjustment range (0.5≤ <0.8): Corresponding positions 500~700mm / 1300~1500mm, current 3~4A / 7~9A, requiring small-amplitude voltage adjustment; key adjustment area (0.2≤ <0.5): Corresponding positions 300~500mm / 1500~1700mm, current 1~3A / 9~12A, requires moderate adjustment, matching core rules 1 and 2; emergency adaptation zone ( <0.2): Corresponding position <300mm / >1700mm, current <1A / >12A, triggering the emergency strategy and matching rule 3 (passenger collision). The heavy load threshold line is marked with a dashed line indicating a current of 10A, distinguishing between normal load and heavy load conditions; the core section line of the door body is marked at a position of 500~1500mm, which is the main working range of the door controller; three core rules are marked in the corresponding area, intuitively showing the membership conditions for rule triggering.

[0040] Rule matching is achieved through parallel comparison and priority encoding. Iterative inference is performed using the quantum annealing algorithm. Initial temperature. =9 (dimensionless), iteration step size =0.4, maximum number of iterations =20, the termination condition is that the rule matching error ≤0.009; the iterative steps based on FPGA are as follows: T1, Initialization: Read the current membership degree Values ​​and rule base standard membership Calculate the initial error ; T2, Temperature Update: k is the current iteration number; T3, Calculation of probability for rule adjustment: ,in This is the current error. This refers to the adjacent rule error; T4, Random Number Generation: Generates a random number r between 0 and 1 through the FPGA's Linear Feedback Shift Register (LFSR) at a clock frequency of 10MHz; T5, Rule Update: If If so, update the candidate rule to the adjacent rule and recalculate. Otherwise, the current rules will remain in effect. T6, Termination of judgment: If the error... ≤0.009 or k≥ Output the control command corresponding to the current rule.

[0041] like Figure 5 The diagram illustrates the error variation of the simulated quantum annealing algorithm. In iteration 0 (initial state), the error is 0.015; after 10 iterations, the error decreases to 0.007 (below the threshold of 0.009), achieving convergence; from iterations 10 to 20, the error stabilizes between 0.006 and 0.008, exhibiting slight fluctuations, demonstrating the low-temperature fine-tuning characteristics of quantum annealing. The initial temperature is 9, decreasing by 0.4°C per iteration: 8.2°C for the 2nd iteration, 6.6°C for the 6th, 5.4°C for the 10th, and 1.0°C for the 20th.

[0042] The control commands generated by quantum fuzzy inference include drive voltage increments (-8V to +8V, step size 0.5V), stop commands, and reverse fine-tuning commands. The commands are transmitted to the execution module via the CAN bus.

[0043] S106: The brushless motor drive module that drives the door controller executes control commands to achieve precise door operation.

[0044] The execution module hardware includes a brushless motor drive module, a CPLD logic control unit, a power amplifier circuit (using SiC MOSFETs), and an overcurrent / overvoltage protection circuit. It adopts a combined control mode of ARM processor and CPLD. The ARM is responsible for instruction parsing and motor speed closed-loop control, while the CPLD is responsible for power device driving and hardware protection. Drive voltage regulation is achieved by adjusting the duty cycle of the PWM signal. The PWM frequency is 20kHz, the duty cycle resolution is 0.1%, and the voltage regulation accuracy is ±0.2V.

[0045] Overcurrent protection: When the current is ≥13A, the CPLD immediately cuts off the drive signal with a response time ≤1μs; Overvoltage protection: When the supply voltage is ≥140VDC, the TVS diode breaks down and clamps, and the CPLD triggers a power-off command; Phase loss protection: The motor's three-phase current is detected, and the machine stops within 10ms when a phase is lost. The motor control mode adopts the FOC (Field Oriented Control) algorithm, and the ARM processor calculates the motor rotor position in real time to control the gate speed fluctuation rate.

[0046] S107: Collects adjusted gate operation data to form closed-loop control and handles abnormal operating conditions.

[0047] The sensor in S101 synchronously collects the door position, speed, and current data with a collection period of 5ms, which is synchronized with the control command execution period. Positional deviation The target location is sent by the vehicle's TCMS system; speed deviation The target speed is preset according to the trapezoidal speed curve (start acceleration → constant speed → deceleration and stop).

[0048] Normal operating conditions: If ≤±2mm and ≤±0.02m / s, maintain the current control parameters; Deviation exceeds limit: If >±2mm or If the value is greater than ±0.02 m / s, return to step S103 to re-execute quantum encoding and inference.

[0049] Abnormal operating conditions include: current ≥13A (heavy load jamming), passenger collision (current sudden change ≥3A), power supply voltage ≤77VDC / ≥137.5V, door position deviation ≥10mm (lasting for 2 cycles); the corresponding emergency strategies include: variance adjustment: the variance of the quantum fuzzy membership function is increased from the usual 0.08 / special 0.25 to 0.35, expanding the effective membership interval; Algorithm parameter adjustments: The initial temperature of quantum annealing is increased to 11, the iteration step size is shortened to 0.3, and the target inference time is ≤4ms; the feedback cycle is shortened: from 5ms to 3ms, and the door status is monitored in real time; emergency command: when closing the door in an emergency, the driving voltage increments by +6V, and the response delay is ≤120ms; when a passenger collides, the machine stops immediately and is finely adjusted in the reverse direction by 20mm; recovery logic: after the abnormality is cleared (current ≤10A, deviation ≤3mm), the normal control parameters are restored within 3 feedback cycles (9ms).

[0050] like Figure 2 The diagram shows a quantum fuzzy control system for a gate controller in a rail transit vehicle. This system implements the aforementioned method and includes a data acquisition module, a preprocessing module, a quantum fuzzy calculation module, an instruction execution module, and a closed-loop feedback module. Data acquisition module: includes displacement sensor, Hall speed sensor, current sensor, temperature sensor and voltage sampling interface (including TVS transient suppression diode), which collects door position, running speed, drive current, ambient temperature and power supply voltage data respectively, with an acquisition period of 5ms; The preprocessing module, integrated into the FPGA chip, includes a Kalman filter operation unit, an EMC compensation unit, and an MVB bus interface unit. It uses Kalman filtering and EMC anti-interference compensation algorithms to preprocess the acquired data and transmits it to the quantum fuzzy computing module via the MVB bus. The quantum fuzzy computing module, developed based on an FPGA chip, includes a quantum encoding unit, a quantum superposition state construction unit, a quantum fuzzy membership function optimization unit, a quantum fuzzy inference unit, and a rule base storage unit. It realizes quantum mapping encoding, superposition state construction, membership function optimization, and generation of control instructions; the rule base supports dynamic updates of 156 rules.

[0051] The instruction execution module includes a brushless motor drive module, a CPLD logic control unit, a SiC power amplifier circuit and a protection circuit. It receives the generated control instructions and drives the door motor to run. It has hardware-level protection functions for overcurrent, overvoltage and phase loss. Closed-loop feedback module: including data feedback interface, deviation judgment unit, and emergency control unit, integrated into FPGA chip, to collect door operation status data in real time, judge deviation and trigger closed-loop control or emergency strategy to ensure control accuracy.

[0052] Example 1: Subway door controller case This embodiment is applied to the Bosman BSM series subway door controller. The hardware configuration includes mature commercial components such as FPGA chips (100K logic gates), brushless motor drive modules, and MVB communication interfaces. Core logic such as quantum superposition state construction and quantum annealing algorithm can be implemented through the hardware logic unit of FPGA. The number of iterations (≤20 times) and the time consumption (≤8ms) of the quantum annealing algorithm are within the computing power range of FPGA, and there is no technical bottleneck.

[0053] Data acquisition includes door position ranging from 0 to 2000 mm (corresponding to the door being fully closed to fully open), operating speed ranging from 0 to 0.8 m / s, drive current ranging from 0 to 15 A, ambient temperature ranging from -40℃ to +85℃, and power supply voltage ranging from DC77V to DC137.5V (compatible with DC110V power supply systems). The system acquires door position data at 320 mm via a displacement sensor, operating speed data at 0.25 m / s via a Hall effect speed sensor, drive current data at 11.2 A via a current sensor, ambient temperature data at 28℃ via a temperature sensor, and power supply voltage data at 125VDC via a voltage sampling interface. The acquisition period is 5 ms.

[0054] The collected operational data undergoes preprocessing, including Kalman filtering and EMC anti-interference compensation. The Kalman filter equation is: Where A=0.92, representing the state transition matrix (coefficients); B=0.08, representing the control input matrix (coefficients). For filter gain, when signal-to-noise ratio ≥ 30dB =0.75, signal-to-noise ratio <30dB =0.55; This is the estimated state value at the current moment. This is the estimated state value from the previous moment; To control the input quantity, specifically the drive control signal of the gate controller; These are the observed values ​​at the current moment. The design uses fixed values ​​for A and B to accommodate system inertia. To adapt the dynamic values ​​to the short-cycle, low-dynamic, and highly volatile characteristics of the gate controller, this ensures that the filtered data is both stable and accurate. Dynamic adjustment The weight of observations can be reduced under strong disturbances to minimize the impact of noise, while the weight of observations can be increased under weak disturbances to ensure the real-time performance of state estimation.

[0055] The Kalman filter algorithm (A=0.92, B=0.08) was used. Electromagnetic noise is filtered out using a coefficient of 0.75, and EMC anti-interference compensation (compensation coefficient 0.14) is added. The corrected data is then transmitted to the quantum fuzzy computing module via the MVB bus with a transmission delay of 1.2ms.

[0056] Quantum encoding is performed according to the mapping rules: 320mm corresponds to quantum state |001000>, 0.25m / s corresponds to |0101>, 11.2A corresponds to |1101>, and 125VDC corresponds to |1011>. Superposition state construction based on quantum encoding: Next, quantum fuzzy inference is performed. Since the driving current is 11.2A ≥ 10A, the variance is set to 0.25, and the membership degree is calculated. =0.83, matching rule 1 generates a "+5V" drive voltage adjustment command, and the error of the quantum annealing algorithm drops to 0.007 after 10 iterations.

[0057] The drive module adopts ARM+CPLD combined control (two-level protection of software and hardware). After receiving the command, it drives the brushless motor to run, realizing precise adjustment of the opening and closing speed and force. When executing the control command, the drive module driving the brushless motor adjusts the voltage from the original 35V to 40V to enhance the closing driving force.

[0058] Closed-loop feedback is used to collect and adjust the gate position (280mm), speed (0.28m / s), and current (10.8A). If the deviations meet the threshold requirements, the current parameters are maintained; otherwise, data is collected again to continue the gate control adjustment.

[0059] The deviation judgment is as follows: after collecting and adjusting the data, the parameters are maintained when the position deviation is ≤ ±2mm and the speed deviation is ≤ ±0.02m / s. When the deviation exceeds the limit or the current is ≥13A, the process returns to the encoding step and re-encodes and infers the newly collected data; in emergency situations (such as passenger collision), an emergency strategy is triggered (variance 0.35, feedback period 3ms, response delay ≤120ms).

[0060] Throughout the quantum fuzzy control process of the gating controller, multiple parameters are processed in parallel. Experimental results show significant improvements in response delay, inrush current, and parameter drift rate, making it suitable for scenarios with strong interference and dynamic loads in rail transit. The performance indicators of the gating controller under different operating conditions are shown in Table 2 below.

[0061] Table 2: Performance Indicators of Gate Controllers under Different Operating Conditions All the indicators listed in Table 1 meet the requirements of standards such as EN50155 and IEC61508. Compared with the traditional PID control algorithm, the dynamic load adaptability, anti-interference performance and safety reliability are significantly improved.

[0062] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, disclosure, and other materials. In this specification, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple components. A single processor or other unit can implement several functions listed in the specification. While certain measures are described in different embodiments, this does not mean that these measures cannot be combined to produce good results.

[0063] Although the invention has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made therein without departing from the spirit and scope of the invention. Accordingly, this specification and drawings are merely illustrative of the invention and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if such modifications and modifications fall within the scope of the invention and its equivalents, the invention is also intended to include such modifications and modifications.

Claims

1. A quantum fuzzy control method for a gate controller of a rail transit vehicle, characterized in that, Includes the following steps: S1: Real-time acquisition of the gate controller's operating data, including gate position, operating speed, drive current, ambient temperature, and power supply voltage; S2: Perform preprocessing on the running data, including Kalman filtering and EMC anti-interference compensation; S3: Perform four-dimensional variable qubit mapping encoding on the preprocessed gate position, running speed, driving current, and supply voltage to construct a quantum superposition state including perturbation compensation; S4: Optimize the fuzzy membership function based on quantum probability distribution, call quantum fuzzy rules to iteratively execute quantum fuzzy inference, and generate gating control instructions; S5: The brushless motor drive module of the gate controller executes the control command; S6: Collect the gate operation data after the instruction adjustment, and calculate the deviation between the actual state and the target state; if the deviation meets the preset threshold, maintain the current parameters; otherwise, return to step S3 to re-execute quantum mapping encoding and quantum fuzzy inference.

2. The method according to claim 1, characterized in that, In step S2, the Kalman filter equation is: ， Where A = 0.92, B = 0.08, For filter gain, when signal-to-noise ratio ≥ 30dB =0.75, signal-to-noise ratio <30dB =0.55; This is the estimated state value at the current moment. This is the estimated state value from the previous moment; To control the input quantity, specifically the drive control signal of the gate controller; This represents the observed value at the current moment.

3. The method according to claim 1, characterized in that, In step S3, the specific rule for the qubit mapping encoding is: gate position Pos Mapped to 6 qubits, operating speed Vel mapped to 4 qubits, driving current Cur The power supply voltage Volt is mapped to 4 qubits; the expression for the quantum superposition state is: ， Where α and β are quantum probability amplitudes and α² + β² = 1, | Pos >、∣ Vel >、∣ Cur >、|Volt> are the reference quantum states, | Pos ′>、∣ Vel ′>、∣ Cur ′>、|Volt'> are perturbation-compensated quantum states.

4. The method according to claim 3, characterized in that, The expression for the fuzzy membership function in step S4 is: ， in, , It is an 18-dimensional quantum probability density matrix; The variance of the fuzzy membership function is dynamically adjusted according to the application scenario. When the driving current is ≥10A or the ambient temperature is ≥60℃ / ≤-30℃, the variance is set to 0.25; otherwise, it is set to 0.

08.

5. The method according to claim 1, characterized in that, In step S4, the quantum fuzzy rule base includes more than 150 rules, with core rules including at least: Rule 1: If the door position is ≤500mm, the speed is ≤0.3m / s, the current is ≥10A, and the voltage is ≥120VDC, then the driving voltage increment is +5V; Rule 2: If the door position is ≥1500mm, the speed is ≥0.6m / s, the current is ≤3A, and the voltage is ≤90VDC, then the driving voltage increment is -3V; Rule 3: If a sudden current change of ≥3A indicates a passenger collision has been detected, a stop command will be immediately output and the machine will be finely adjusted in the opposite direction by 20mm.

6. The method according to claim 1, characterized in that, In step S4, the iterative execution of quantum fuzzy inference uses the quantum annealing algorithm for iterative calculation. The initial temperature is set to 9, and the temperature decreases by 0.4 with each iteration. The iteration terminates when the rule matching error is ≤0.

009.

7. The method according to claim 3, characterized in that, The range of gate position is 0~2000mm, the range of running speed is 0~0.8m / s, the range of driving current is 0~15A, the range of ambient temperature is -40℃~+85℃, the range of power supply voltage is DC77V~DC137.5V, and the acquisition period is 5ms.

8. The method according to claim 7, characterized in that, The interval division of the qubit mapping in step S3 includes: the interval corresponding to each quantum state of the gate position is about 31.25 mm, the interval corresponding to each quantum state of the running speed is 0.05 m / s, the interval corresponding to each quantum state of the driving current is 0.9375 A, and the interval corresponding to each quantum state of the supply voltage is 3.78 VDC; the compensation amount of the disturbance compensation quantum state is 8% of the interval of the corresponding variable quantum state.

9. The method according to claim 1, characterized in that, In step S2, the EMC anti-interference compensation coefficient is set according to the noise intensity, and the compensation coefficient is set to 0.13~0.16 when the signal-to-noise ratio is 28dB.

10. A quantum fuzzy control system for a gate controller of a rail transit vehicle, used to implement the method described in any one of claims 1 to 9, characterized in that, The system includes a data acquisition module, a preprocessing module, a quantum fuzzy computing module, an instruction execution module, and a closed-loop feedback module; wherein: The data acquisition module includes a displacement sensor, a Hall speed sensor, a current sensor, a temperature sensor, and a voltage sampling interface, which respectively collect data on the door position, running speed, drive current, ambient temperature, and power supply voltage. The preprocessing module, integrated into the FPGA chip, uses Kalman filtering and EMC anti-interference compensation algorithms to preprocess the acquired data and transmits it to the quantum fuzzy computing module via the MVB bus; The quantum fuzzy computing module, developed based on an FPGA chip, includes a quantum encoding unit, a quantum superposition state construction unit, a quantum fuzzy membership function optimization unit, and a quantum fuzzy inference unit, which are used to realize quantum mapping encoding, superposition state construction, membership function optimization, and control instruction generation. The instruction execution module, including a brushless motor drive module, a CPLD logic control unit, and a power amplifier circuit, is used to receive control instructions and drive the door motor to run. The closed-loop feedback module includes a data feedback interface and a deviation judgment unit, which is used to collect the door's operating status data, judge the deviation, and trigger closed-loop control.

Images