Quantum fuzzy control method and system for rail transit vehicle-mounted air conditioner

Abstract

The invention relates to the technical field of intelligent control of rail transit vehicle-mounted equipment, in particular to a quantum fuzzy control method and system for a rail transit vehicle-mounted air conditioner, and the method comprises the steps: collecting the operation data of the vehicle-mounted air conditioner in real time, and carrying out the Kalman filtering noise reduction and EMC anti-interference compensation preprocessing; performing four-dimensional variable quantum bit mapping coding on the preprocessed operation data, and constructing a quantum superposition state including disturbance compensation; a fuzzy membership function is optimized based on quantum probability distribution, a quantum fuzzy rule base is called, and an air conditioner control instruction is generated through quantum annealing algorithm iterative reasoning; an execution mechanism is driven to execute the air conditioner control instruction, the adjusted operation data, the comfort index and the state of the solid state contactor are collected, and deviation is calculated; and if the deviation meets a preset threshold value, maintaining the current parameter, otherwise, returning to re-execute quantum mapping coding and quantum fuzzy reasoning. According to the invention, accurate control of the vehicle-mounted air conditioner of the rail transit vehicle can be realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for rail transit onboard equipment, and in particular to a quantum fuzzy control method and system for rail transit onboard air conditioning, applicable to the precise control of onboard air conditioning in rail transit vehicles such as subways, urban rail transit, and high-speed trains. Background Technology

[0002] As a core energy-consuming device in rail transit vehicles, air conditioning accounts for 30%-40% of total energy consumption, and its control performance directly affects operating costs and passenger experience. During rail transit vehicle operation, passenger density and indoor / outdoor temperature and humidity frequently change. Traditional PID algorithms process multiple parameters serially, resulting in high adjustment response delays, leading to air conditioning overcooling / overheating, significant compressor start-up and shutdown shocks, and severely impacting equipment lifespan. The rail transit traction system generates strong electromagnetic noise (signal-to-noise ratio as low as 25dB), causing drift in the fixed threshold parameters of conventional fuzzy control, affecting the adjustment effect. Existing algorithms often focus on optimizing a single parameter, making it difficult to balance passenger comfort (PMV index), energy efficiency (COP value), and equipment protection, resulting in contradictions such as high energy consumption and low comfort or high comfort and high energy consumption. Especially in special scenarios such as vehicle start-up and shutdown, tunnel entry and exit, and sudden changes in passenger flow, there is a lack of rapid emergency adjustment mechanisms for abnormal operating conditions, posing risks of sudden drops in comfort or equipment overload.

[0003] There is an urgent need for an intelligent control method and system for air conditioning in rail transit vehicles that can adapt to dynamic operating environments. Summary of the Invention

[0004] The purpose of this invention is to provide a quantum fuzzy control method and system for rail transit vehicle air conditioning, which solves the problems of slow response, weak anti-interference, insufficient multi-parameter coordination, poor adaptability to abnormal working conditions and low hardware reliability of existing rail transit vehicle air conditioning control. It can achieve a dynamic balance of precise comfort control, energy consumption optimization, equipment safety and convenient operation and maintenance, and is compatible with various types of rail transit vehicles such as subways, urban rail, and EMU trains.

[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 an onboard air conditioner in a rail transit vehicle is provided, comprising the following steps: S1: Real-time collection of vehicle air conditioning operation data, including interior temperature, interior relative humidity, passenger density, and outdoor temperature; S2: Perform Kalman filtering noise reduction and EMC anti-interference compensation preprocessing on the running data; S3: The pre-processed temperature inside the carriage, relative humidity inside the carriage, passenger density, and outdoor temperature are encoded by four-dimensional variable qubit mapping to construct a quantum superposition state including perturbation compensation; S4: Based on the quantum probability distribution, optimize the fuzzy membership function, call the quantum fuzzy rule base, and generate air conditioning control instructions through iterative reasoning using the quantum annealing algorithm; the control instructions include compressor frequency, fan speed, electronic expansion valve opening, fresh air ratio, and solid-state contactor drive signal; S5: Drive the actuator to execute the air conditioning control command, the actuator including a SiC power module and a solid-state contactor; S6: Collect the adjusted operating data, comfort index and solid contactor status, and calculate the deviation; 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 = 0.94, B = 0.06, For filter gain, when signal-to-noise ratio ≥ 30dB =0.78, signal-to-noise ratio <30dB =0.58; This is the estimated state value at the current moment. This is the estimated state value from the previous moment; To control the input amount, This represents the observed value at the current moment.

[0007] According to one embodiment of the present invention, in step S3, the specific rule for the quantum bit mapping encoding is: temperature inside the carriage. Tin Mapping 6-bit qubits, relative humidity Hin Mapping 4-bit qubits, passenger density P Mapping 4 qubits, outdoor temperature Tout Mapping 4 qubits; the expression for the quantum superposition state is: , Where α and β are quantum probability amplitudes and α² + β² = 1, | Tin >、∣ Hin >、∣ P >、∣ Tout > represents the baseline quantum state, | Tin ′>、∣ Hin ′>、∣ P ′>、| Tout '> is the perturbation compensation quantum state.

[0008] According to an embodiment of the present invention, the expression for the quantum fuzzy membership function in step S4 is: , in, , It is an 18-dimensional quantum probability density matrix; The variance of the quantum fuzzy membership function is dynamically adjusted according to the application scenario. It is set to 0.1 for normal scenarios and 0.3 for special scenarios. The special scenarios include: passenger density > 8 people / m² or < 2 people / m², and temperature inside the carriage > 32℃ or < 14℃.

[0009] According to one embodiment of the present invention, in step S4, the quantum fuzzy rule base includes more than 152 rules, covering all operating conditions of cooling, heating, ventilation, and dehumidification, and the core rules include at least: Rule 1: If Tin≥28℃, Hin≥60%, P≥7 people / m², and Tout≥30℃, then the compressor frequency increment is +5Hz, the fresh air ratio is +10%, and the solid contactor remains stably conductive. Rule 2: If Tin≤18℃, Hin≤40%, P≤3 people / m², and Tout≤5℃, then the compressor frequency increment is -3Hz, the fresh air ratio is -8%, and the fan drive current is reduced. Rule 3: If the passenger density suddenly increases by ≥2 people / m² (for 2 consecutive collection cycles), immediately maintain the current compressor frequency, temporarily increase the fresh air ratio by 5%, and start solid contactor status monitoring; Rule 4: If the current of the solid-state contactor suddenly changes by ≥5A or the temperature by ≥110℃, the overcurrent / overtemperature protection will be triggered, the compressor frequency will drop by 10Hz, and the fault log will be recorded.

[0010] According to one embodiment of the present invention, in step S4, the initial temperature of the quantum annealing algorithm is set to 9.5, the temperature decreases by 0.45 with 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 air conditioning control command includes compressor frequency, fan speed, electronic expansion valve opening, fresh air ratio, and solid-state contactor drive signal.

[0012] According to one embodiment of the present invention, the interval division of the qubit mapping in step S3 includes: the temperature inside the carriage is mapped to 6 qubits, with each quantum state corresponding to an interval of approximately 0.39°C; the relative humidity is mapped to 4 qubits, with each quantum state corresponding to an interval of 3.125%; the passenger density is mapped to 4 qubits, with each quantum state corresponding to an interval of 0.625 people / m²; the outdoor temperature is mapped to 4 qubits, with each quantum state corresponding to an interval of 6.875°C; and the compensation amount of the disturbance compensation quantum state is 10% of the corresponding variable quantum state interval.

[0013] According to one embodiment of the present invention, step S6 further includes triggering an emergency adjustment strategy if the membership degree is <0.2 or the solid-state contactor fails, and recording fault data at the same time; wherein, the emergency adjustment strategy includes: increasing the variance to 0.4, increasing the initial temperature of quantum annealing to 11.5, shortening the iteration step size to 0.35, strengthening the adjustment command, and shortening the feedback cycle to 5ms; when the solid-state contactor fails, switching to the backup channel and recording fault data.

[0014] On the other hand, the present invention also provides a quantum fuzzy control system for an air conditioner in 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 is used to collect the operating data of the vehicle air conditioner in real time, including the temperature inside the vehicle, the relative humidity inside the vehicle, the passenger density, and the outdoor temperature. 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 membership function optimization unit, a quantum fuzzy inference unit, and a rule base storage unit. It is used to realize quantum mapping encoding, superposition state construction, membership function optimization, storage of fuzzy rules, and generation of control instructions. The instruction execution module includes a SiC power module, a compressor drive circuit, a fan control unit, an electronic expansion valve drive interface, a solid-state contactor, and a protection circuit; the solid-state contactor uses semiconductor switching devices and has hardware-level protection functions for overcurrent, overvoltage, and overtemperature. The closed-loop feedback module includes a data feedback interface and a deviation judgment unit, which is used to collect air conditioning operation data, judge deviations and trigger closed-loop control, and also integrates a fault recording unit to store fault data.

[0015] The quantum fuzzy control method and system for onboard air conditioning in rail transit vehicles of the present invention have the following advantages compared with the prior art: 1. Significantly improved dynamic load adaptability: By processing four-dimensional variables in parallel through quantum superposition, the adjustment response delay is reduced, the impact of compressor start-up and shutdown is reduced, the equipment life is extended, and it is adapted to different passenger flow fluctuation scenarios in subways, urban rail transit, and high-speed trains. 2. By combining quantum compensation mechanism with EMC anti-interference compensation, the parameter drift rate is reduced, and the adjustment pass rate is ≥98.5% in a strong interference environment with a signal-to-noise ratio of 25dB. 3. Multi-parameter collaborative optimization was achieved: the PMV index was stabilized in the comfort range of -0.5 to +0.5, and the COP of the air conditioner was improved from 3.2 to 3.8; 4. It can efficiently handle abnormal operating conditions and quickly adapt to special scenarios such as sudden changes in passenger flow, sudden changes in environment, and failure of solid contactors; 5. Based on existing FPGA chips, CAN interfaces and SiC power modules, no core hardware replacement is required, resulting in low modification costs; solid-state contactors replace traditional mechanical parts, eliminating contact wear and arcing risks, reducing failure rates, and fault recording and remote diagnostic functions improve maintenance efficiency. Attached Figure Description

[0016] 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 an onboard air conditioner in rail transit according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a quantum fuzzy control system for an onboard air conditioner in a rail transit vehicle according to an embodiment of the present invention; Figure 3 This is a contour plot of the two-dimensional variable membership function according to an embodiment of the present invention; Figure 4 This is the iterative error curve of the quantum annealing algorithm in an embodiment of the present invention; Figure 5 This is a schematic diagram of the solid-state contactor control logic according to an embodiment of the present invention. Detailed Implementation

[0017] 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.

[0018] 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.

[0019] 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.

[0020] Quantum computing's superposition and parallel processing capabilities can significantly improve the computational efficiency of complex systems, while fuzzy control excels at handling uncertainty problems. Combining the two can effectively compensate for the shortcomings of traditional control methods. Simultaneously, the contactless nature of solid-state power control technology can address the reliability issues of traditional mechanical components. Currently, there is no mature solution that combines quantum fuzzy control and solid-state control technology for targeted application in all types of rail transit vehicle air conditioning systems. There is an urgent need for an intelligent control method and system adapted to dynamic operating environments. This application implements quantum logic operations based on FPGA rather than a large-scale quantum computer. Its core principle is to utilize the quantum superposition property to improve parallel processing efficiency, belonging to quantum-inspired classical computing.

[0021] like Figure 1 As shown, a flowchart of a quantum fuzzy control method for onboard air conditioning in rail transit vehicles is presented. The method includes the following steps: S101: Real-time acquisition of core operating data of vehicle air conditioning, including four-dimensional control variables and auxiliary monitoring parameters; The operational data includes four-dimensional control variables: interior temperature (Tin), interior relative humidity (Hin), passenger density (P), and outdoor temperature (Tout); the range of interior temperature acquisition parameters is 10℃~35℃, the range of outdoor temperature acquisition parameters is -40℃~+70℃; the range of relative humidity acquisition parameters is 30%~80%; and the range of passenger density acquisition parameters is 0~10 people / m².

[0022] Auxiliary monitoring parameters include: air conditioner compressor current (Ic), fan speed (N), power supply voltage (U), ambient temperature (Te), and solid contactor status (including operating current and surface temperature); compressor current: 0~50A, power supply voltage: AC380V.

[0023] The data acquisition period is set to 10ms. Synchronous acquisition by multiple sensors is triggered by an FPGA timer interrupt to ensure data timestamp consistency. Acquired data is transmitted to the preprocessing module via an SPI interface. The acquisition equipment includes a high-precision temperature sensor, humidity sensor, infrared passenger flow sensor, Hall current sensor, and voltage sampling interface (integrated with a TVS transient voltage suppressor diode for surge protection).

[0024] 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 represented in vector form using four-dimensional variables, taking into account the multi-parameter characteristics of the air conditioner. 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; where A=0.94 (state transition matrix coefficients) and B=0.06 (control input matrix coefficients). For Kalman filter gain, dynamic adjustment strategy: when signal-to-noise ratio ≥ 30dB =0.78 (weak interference, dependent on observation), signal-to-noise ratio <30dB =0.58 (strong interference, dependent on predicted value); The input quantity is generated from the previous round of quantum fuzzy reasoning to control the input quantity; The current observation value (raw data collected by the sensor), and... The dimensions are consistent, but it contains noise components. The filtering algorithm is implemented using VerilogHDL on the FPGA.

[0025] EMC interference immunity compensation sets the compensation coefficient according to the noise intensity. The compensation coefficient is set to 0.13~0.16 when the signal-to-noise ratio is 25dB, and to 0.09~0.12 when the signal-to-noise ratio is 35dB; the compensated data... , This is the filtered data. The compensation coefficient is used. The preprocessed data is transmitted to the quantum fuzzy computing module via the CAN bus.

[0026] S103: Quantum bit mapping encoding; The preprocessed four-dimensional variables were encoded using qubit mapping. The mapping rules included: the temperature inside the carriage (10℃~35℃) was mapped to 6 qubits, with each quantum state corresponding to a range of approximately 0.39℃; the relative humidity (30%~80%) was mapped to 4 qubits, with each quantum state corresponding to a range of 3.125%; the passenger density (0~10 people / m²) was mapped to 4 qubits, with each quantum state corresponding to a range of 0.625 people / m²; and the outdoor temperature (-40℃~+70℃) was mapped to 4 qubits, with each quantum state corresponding to a range of 6.875℃. The specific qubit mapping rules are shown in Table 1, which clearly defines the range division and encoding logic.

[0027] Table 1: Quantum Bit Mapping Rules S104: Construction of quantum superposition state; the quantum superposition state is constructed as follows: , Where α and β are quantum probability amplitudes and α² + β² = 1, α = 0.85 (baseline quantum state weight), β = 0.52 (perturbation compensation quantum state weight); | Tin >、∣ Hin >、∣ P >、∣ Tout > represents the baseline quantum state, | Tin ′>、∣ Hin ′>、∣ P ′>、| Tout The superposition state is a perturbation-compensated quantum state, with the compensation amount being 10% of the corresponding variable quantum state range. For example, the compensation amount for Tin is approximately 0.039℃, and the compensation amount for Hin is approximately 0.3125%. The superposition state is implemented through the tensor product operation unit of the FPGA, using a parallel logic architecture.

[0028] S105: Based on the optimization of fuzzy membership function using quantum probability distribution, air conditioning control commands are generated through iterative reasoning using quantum annealing algorithm; 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, derived from the sum of the number of qubits in the four-dimensional variables.

[0029] The 18-dimensional matrix is ​​decomposed into its components by dimensionality decomposition. Tin Submatrix (6-dimensional) Hin Submatrix (4D), P-submatrix (4D) ToutSubmatrices (4-dimensional), each following a Gaussian distribution, are merged into submatrix P through tensor product operation: .

[0030] The membership function employs a dynamic variance adjustment strategy: In typical scenarios, the variance σ = 0.1, the passenger density is 3~6 people / m², and the temperature inside the carriage is 18~28℃. The membership function curve is steep, which is used to improve control accuracy. In special scenarios, the variance σ = 0.3, the passenger density is >8 people / m² or <2 people / m², the temperature inside the carriage is >32℃ or <14℃, and the membership function curve is flat to enhance the anti-interference capability.

[0031] The quantum fuzzy rule base includes 152 rules, covering all operating conditions including cooling, heating, ventilation, and dehumidification. Core rules include: Rule 1: If Tin≥28℃, Hin≥60%, P≥7 people / m², and Tout≥30℃, then the compressor frequency increment is +5Hz, the fresh air ratio is +10%, and the solid contactor remains stably conductive. Rule 2: If Tin≤18℃, Hin≤40%, P≤3 people / m², and Tout≤5℃, then the compressor frequency increment is -3Hz, the fresh air ratio is -8%, and the fan drive current is reduced. Rule 3: If the passenger density suddenly increases by ≥2 people / m² (for 2 consecutive collection cycles), immediately maintain the current compressor frequency, temporarily increase the fresh air ratio by 5%, and start solid contactor status monitoring; Rule 4: If the current of the solid-state contactor suddenly changes by ≥5A or the temperature by ≥110℃, the overcurrent / overtemperature protection will be triggered, the compressor frequency will drop by 10Hz, and the fault log will be recorded. Rule 5: If the power supply voltage is ≤342V or ≥418V and continues for one sampling cycle, the compressor frequency will be locked at the current value ±2Hz, the fresh air ratio will be fixed at 30%, and a voltage abnormality alarm will be triggered.

[0032] The initial temperature for quantum annealing algorithm inference is set to 9.5°C, and the temperature decreases by 0.45°C with each iteration. The iteration terminates when the rule matching error is ≤0.009. The rule matching probability is adjusted by simulating the quantum tunneling effect to quickly lock in the globally optimal control strategy. Specific iterative steps include: T1, Initialization: Read the current membership degree Membership degree with the rule base standard Calculate the initial error ; T2, Temperature Update: (where 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 using the FPGA's Linear Feedback Shift Register (LFSR); 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... If k reaches the maximum number of iterations, output a control command.

[0033] Control command types include: compressor frequency (30Hz~100Hz, step size 0.5Hz), fan speed (500rpm~1500rpm, step size 50rpm), fresh air ratio (10%~80%, step size 5%), electronic expansion valve opening (20%~80%, step size 5%), and solid-state contactor drive signal (PWM duty cycle 0%~100%).

[0034] Step S106: Control Command Execution. The precise control command generated by the inference is output to the actuator. The actuator uses a combination of SiC power module and solid-state contactor, and achieves precise adjustment through PWM pulse width modulation, with a voltage adjustment accuracy of ±0.2V. It has three levels of protection functions: software, hardware, and fuse. More specifically, the actuator includes a SiC power module (1200V / 450A), a compressor drive circuit, a fan control unit, an electronic expansion valve drive interface, a solid-state contactor (rated current 450A, AC380V compatible), and protection circuit.

[0035] The compressor frequency is regulated by controlling the SiC power module to turn on and off via PWM pulse width modulation (frequency 20kHz, duty cycle resolution 0.1%); the fan speed is regulated by outputting an adjustable voltage (0~380V) through the frequency converter drive circuit; solid-state contactor control is implemented using opto-isolation drive to ensure electrical isolation between the control terminal and the power terminal; the three-level protection of software, hardware, and fuses is as follows: in case of overcurrent (current ≥50A), overvoltage (voltage ≥450V), or overtemperature (SiC module temperature ≥120℃), the protection circuit response time is ≤1μs, and the drive signal is immediately cut off.

[0036] Step S107: Closed-loop feedback and dynamic optimization: Collect adjusted comfort index (PMV), energy consumption data, equipment operating parameters, and solid-state contactor status; calculate the deviation between the actual state and the target state. Under normal operating conditions, if the PMV value is in the range of -0.5 to +0.5 and the compressor current deviation is ≤ ±5%, and the solid-state contactor is normal, then the current control parameters shall be maintained. Deviation exceeds limit: If PMV exceeds the comfort range or current deviation is > ±5%, return to step S103 to re-execute quantum mapping encoding and quantum fuzzy inference; Emergency adjustment under abnormal operating conditions: When the quantum fuzzy membership degree is <0.2 (corresponding to Tin <14℃ or >32℃, P <2 people / m² or >8 people / m²) or the solid-state contactor fails, an emergency strategy is triggered, which includes the following strategies: Variance adjustment: Increase the variance of the membership function to 0.4 to expand the effective membership interval; Inference parameter adjustments: The initial temperature of quantum annealing was increased to 11.5°C, and the iteration step size was shortened to 0.35, accelerating the rule matching speed; Enhanced adjustment instructions: If Tin > 32℃, add an additional 3Hz compressor frequency on top of core rule 1; if P > 8 people / m², add an 8% compensation to the fresh air ratio; if the solid contactor fails, immediately switch to the backup channel and issue an alarm signal. High-frequency feedback: The closed-loop feedback cycle is shortened from 10ms to 5ms until the deviation is ≤±3% or the membership degree is ≥0.2, and the normal parameters are restored; at the same time, fault data is recorded to support remote diagnosis.

[0037] like Figure 2 As shown, a schematic diagram of a quantum fuzzy control system for an air conditioning system in a rail transit vehicle is presented.

[0038] The system is developed based on an FPGA chip and includes a data acquisition module, a preprocessing module, a quantum fuzzy computing module, an instruction execution module, and a closed-loop feedback module, among which: The data acquisition module includes temperature sensors (inside and outside the carriage), humidity sensors, infrared passenger flow sensors, current sensors, voltage sampling interfaces (including TVS transient suppression diodes), speed sensors, and solid-state contactor status monitoring units. The sensors are deployed inside and outside the carriage and in the air conditioning unit, supporting simultaneous acquisition of multiple data sources at a sampling frequency of 100Hz, and transmitting the data to the preprocessing module via the SPI interface.

[0039] The preprocessing module, integrated into the FPGA chip, includes a Kalman filter operation unit and an EMC compensation unit. It uses hardware logic to implement the filtering and compensation algorithms, with a processing delay of ≤2ms to ensure data real-time performance. The preprocessed data is transmitted to the quantum fuzzy computing module via the CAN bus, and it also supports preliminary data interaction with the vehicle's TCMS system.

[0040] The quantum fuzzy computing module, developed based on an FPGA chip, includes a quantum encoding unit, a quantum superposition state construction unit, a membership function optimization unit, a quantum fuzzy inference unit, and a rule base storage unit. Quantum coding unit: realizes the quantum bit mapping and encoding of four-dimensional variables, and constructs the coding circuit through logic gate units; Quantum superposition state building unit: Based on the encoding results, a quantum superposition state with perturbation compensation is generated, and a parallel logic architecture is used to improve computational efficiency; Membership function optimization unit: dynamically adjusts the function variance to adapt to different working conditions, and achieves high-precision calculation through floating-point arithmetic unit; Quantum fuzzy reasoning unit: Iterative reasoning is performed using the quantum annealing algorithm, with ≤20 iterations; Rule base storage unit: Stores 152 quantum fuzzy rules, supports dynamic expansion through software updates, and includes dedicated rules for solid-state contactor protection.

[0041] The instruction execution module includes a SiC power module, a compressor drive circuit, a fan control unit, an electronic expansion valve drive interface, a solid-state contactor, and a protection circuit. It is used to receive control instructions output by the quantum fuzzy computing module to achieve precise adjustment of compressor frequency, fan speed, and valve opening. The solid-state contactor uses semiconductor switching devices to replace traditional mechanical contacts, eliminating the risk of oxidation and jamming, and has hardware-level protection functions for overcurrent, overvoltage, and overtemperature.

[0042] The closed-loop feedback module, integrated into the FPGA chip, includes a data feedback interface, a deviation judgment unit, an emergency control unit, and a fault recording unit. It collects adjusted operating data, comfort indicators, and solid-state contactor status in real time, calculates deviations, and triggers closed-loop control or emergency strategies. The fault recording unit can store more than 1,000 fault data entries, supports historical fault queries and abnormal data time-series analysis, and adapts to the needs of intelligent vehicle operation and maintenance.

[0043] like Figure 3 As shown, a two-dimensional quantum fuzzy membership contour plot of passenger density and in-car temperature is presented. The relative humidity is fixed at 50%, and the outdoor temperature is 25℃, which is typical for rail transit. The horizontal axis (X-axis) represents the in-car temperature (Tin), covering the entire operating temperature range of rail transit air conditioning from 10℃ to 35℃, with a scale interval of 5℃, and the core operating range is 22℃ to 26℃ (passenger comfort temperature range); the vertical axis (Y-axis) represents the passenger density (P), ranging from 0 to 10 people / m², which can cover the entire passenger flow scenario of subway / urban rail / high-speed train, with a scale interval of 1 person / m², and the core operating range corresponds to a typical passenger flow density of 3 to 6 people / m².

[0044] Figure 3The shades of color and the density of contour lines represent the quantum fuzzy membership degree (μq), ranging from 0 to 1: Membership degree = 1 indicates that the air conditioning system is in its optimal adaptation state, requiring no significant adjustments, with the lowest energy consumption and best comfort; membership degree 0.8~1 indicates the optimal adaptation zone, corresponding to normal operating conditions, with stable control parameters; membership degree 0.5~0.8 corresponds to the slight adjustment zone, requiring minor adjustments to the compressor frequency / fresh air ratio; membership degree 0.2~0.5 indicates the key adjustment zone, requiring the execution of core rules (such as rule 1 / rule 2); membership degree <0.2 corresponds to the emergency adaptation zone, requiring the triggering of emergency adjustment strategies (variance expansion, high-frequency feedback, etc.). The core range (22~26℃, 3~6 people / m²) has a membership degree close to 1, which aligns with the goal of "comfort first, energy consumption optimal".

[0045] like Figure 4 As shown, a schematic diagram illustrating the error variation of the simulated quantum annealing algorithm is presented, visually demonstrating the dynamic relationship between the number of iterations and the rule matching error, verifying the algorithm's ability to quickly converge to the target error within a finite number of iterations. The horizontal axis (X-axis) represents the number of iterations (k) of the quantum annealing algorithm, ranging from 1 to 20 (with a maximum iteration count N_max=20), with a scale interval of 2, representing the number of iterations from initialization to termination. The vertical axis (Y-axis) represents the rule matching error (Ek), ranging from 0 to 0.12 (dimensionless), representing the matching deviation between the current quantum fuzzy rule and the optimal rule. The smaller the error, the more precise the control commands output by the rule. The iteration termination condition is set to a termination error threshold (0.009). When the rule matching error ≤ 0.009, the algorithm stops iterating and outputs control commands.

[0046] During the iteration process, the initial temperature T0 = 9.5°C, and the temperature remained relatively high for the first 5 iterations (Tk ≥ 7.7°C), resulting in a large Padj. The algorithm could quickly try adjacent rules, reducing the error from 0.12 to approximately 0.03, demonstrating the characteristic of quantum annealing: "rapidly exploring the optimal solution at high temperatures." After 12 iterations of rule adjustment probability, the error decreased to 0.007 (below the termination threshold), verifying that the algorithm can converge earlier in practical applications, meeting real-time control requirements. The quantum annealing algorithm solves the problems of excessive iterations and slow response in traditional fuzzy algorithms, verifying that the algorithm meets the core requirement of "real-time control." Furthermore, Figure 4 The iteration parameters (such as initial temperature and step size) can be adjusted using curves as a basis for parameter optimization: if the error decreases too slowly, the initial temperature can be increased; if the error fluctuates greatly after convergence, the iteration step size can be decreased, providing a quantitative basis for algorithm adaptation to different vehicle types (metro / high-speed trains). The quantum annealing algorithm demonstrates fast convergence and high accuracy in the scenario of air conditioning control for rail transit vehicles.

[0047] Figure 5The full lifecycle control logic based on solid-state contactors is presented, forming a hardware architecture of instruction-driven, state monitoring, fault protection, and feedback closed loop. Those skilled in the art can implement hardware wiring and logic programming based on this diagram.

[0048] The quantum fuzzy computing module serves as the control command source, outputting the PWM drive signal (duty cycle 0%~100%) of the solid-state contactor. The control commands include the solid-state contactor drive signal. The status monitoring unit collects the core parameters of the solid-state contactor in real time: operating current, surface temperature, and supply voltage. The acquisition period is consistent with the system control period (10ms), corresponding to the solid-state contactor status in the auxiliary monitoring parameters.

[0049] The opto-isolated drive unit uses opto-isolation to achieve electrical isolation between the control terminal (FPGA) and the power terminal (solid-state contactor). The opto-isolated drive method is used to avoid drive signal distortion caused by electromagnetic interference.

[0050] The fault diagnosis logic is as follows: Normal operating conditions: operating current ≤50A, surface temperature ≤110℃, power supply voltage 342~418V (AC380V±10%). Fault conditions: Operating current ≥ 50A (overcurrent), surface temperature ≥ 110℃ (overtemperature), power supply voltage < 342V or > 418V (abnormal voltage), triggering protection command.

[0051] Solid-state contactors use semiconductor switching devices to replace traditional mechanical contactors, eliminating the risk of contact oxidation and jamming. The specific action is to achieve "on / off" based on the drive signal: when on, it provides AC380V power to loads such as air conditioning compressors and fans; when off, it cuts off the power circuit and stops the load from running (during normal adjustment or fault protection).

[0052] The three-level protection circuit includes software, hardware, and fuse protection. Software protection outputs shutdown commands through a quantum fuzzy computing module; hardware protection suppresses surge voltage through a voltage clamping circuit; and fuse protection prevents electrical fires by rapidly blowing the fuse during a short circuit.

[0053] The fault recording unit stores ≥1000 fault data entries, including fault type (overcurrent / overtemperature / voltage anomaly), occurrence timestamp, and operating parameters at the time. It supports remote diagnostics and meets the requirements for storing fault data. In the dual feedback link, local feedback transmits status to the closed-loop feedback module for parameter adjustment of the quantum fuzzy algorithm. Remote feedback uploads fault information to the vehicle's TCMS system, supporting intelligent train operation and maintenance. Maintenance personnel can locate fault points according to the logical link (e.g., when a solid-state contactor is not conducting, check the command output, drive unit, and monitoring parameters sequentially), improving maintenance efficiency. Figure 5This not only demonstrates the hardware advantages of contactless solid-state contactors, but also showcases the linkage mechanism with quantum fuzzy algorithms, providing intuitive visual guidance for the system's engineering implementation, debugging, and maintenance.

[0054] Example 1: Quantum fuzzy control of air conditioning in urban subway vehicles.

[0055] The air conditioning has a rated cooling capacity of 36kW (metro / light rail). The control targets are: PMV value -0.5 to +0.5, COP ≥ 3.6. The range of the four-dimensional control variables is: interior temperature 10℃ to 35℃ (core working area 22℃ to 26℃), relative humidity 30% to 80% (core working area 40% to 60%), passenger density 0 to 10 people / m² (metro / light rail core working area 3 to 6 people / m²), and outdoor temperature -40℃ to +70℃ (core working area -10℃ to 35℃).

[0056] The FPGA chip adopts a 100K logic gate architecture and is model Xilinx Spartan-7. The quantum fuzzy computing module is integrated into the FPGA, the rule base is stored in the FPGA's on-chip ROM, and the solid-state contactor has a rated current of 450A and is compatible with AC380V power supply systems.

[0057] The sensor collects real-time data on the following parameters: interior temperature 28.5℃, relative humidity 65%, passenger density 7.2 people / m², and outdoor temperature 32℃. Auxiliary parameters include compressor current 28A, power supply voltage 380V, ambient temperature 32℃, and solid-state contactor temperature 85℃. The data acquisition period is 10ms.

[0058] Data preprocessing involved Kalman filtering (A=0.94, B=0.06). Electromagnetic noise is filtered out using a coefficient of 0.78, 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 CAN bus.

[0059] Quantum encoding is performed according to the mapping rules: 28.5℃ corresponds to quantum state |101001>, 65% humidity corresponds to |1011>, 7.2 people / m² corresponds to |1011>, and 32℃ outdoor temperature corresponds to |1001>; quantum superposition states are constructed: .

[0060] In the quantum fuzzy reasoning process, since the passenger density of 7.2 people / m² is greater than 6 people / m², the variance of the membership function is set to 0.3, and the membership degree is calculated. =0.82, matching rule 1; after 12 iterations of the quantum annealing algorithm, the error decreased to 0.007, generating control commands: compressor frequency +5Hz (current frequency 50Hz adjusted to 55Hz), fresh air ratio +10% (current 30% adjusted to 40%), electronic expansion valve opening +5%, and the solid-state contactor maintained stable conduction. The execution module receives the control commands, adjusts the compressor frequency and fan speed through the SiC power module, maintains reliable conduction of the solid-state contactor, and the protection circuit monitors the working status in real time.

[0061] During the closed-loop feedback control process, the following parameters are collected and adjusted: passenger compartment temperature 26.3℃, PMV value 0.2, compressor current 26.5A, and solid-state contactor temperature 88℃. All deviations meet the threshold requirements, and the current parameters are maintained. If a sudden change in passenger density to 9 people / m² (membership < 0.2) is detected, an emergency strategy is triggered: the variance increases to 0.4, the initial quantum annealing temperature is increased to 11.5℃, the compressor frequency is increased by an additional 3Hz, and the feedback cycle is shortened to 5ms until the PMV stabilizes at 0.3. If a sudden change in solid-state contactor current to 40A is detected, overcurrent protection is immediately triggered, the compressor frequency drops sharply by 10Hz, fault data is recorded, and an alarm is triggered.

[0062] This embodiment demonstrates that the air conditioning quantum fuzzy control method of the present invention is significantly superior to the traditional control method, while solving the hardware reliability problem of the traditional air conditioning control panel and adapting to the needs of various types of rail transit vehicles.

[0063] 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.

[0064] 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 onboard air conditioning in rail transit vehicles, characterized in that, Includes the following steps: S1: Real-time collection of vehicle air conditioning operation data, including interior temperature, interior relative humidity, passenger density, and outdoor temperature; S2: Perform Kalman filtering noise reduction and EMC anti-interference compensation preprocessing on the running data; S3: The pre-processed temperature inside the carriage, relative humidity inside the carriage, passenger density, and outdoor temperature are encoded by four-dimensional variable qubit mapping to construct a quantum superposition state including perturbation compensation; S4: Based on the quantum probability distribution, optimize the fuzzy membership function, call the quantum fuzzy rule base, and generate air conditioning control instructions through iterative reasoning using the quantum annealing algorithm; the control instructions include compressor frequency, fan speed, electronic expansion valve opening, fresh air ratio, and solid-state contactor drive signal; S5: Drive the actuator to execute the air conditioning control command, the actuator including a SiC power module and a solid-state contactor; S6: Collect the adjusted operating data, comfort index and solid contactor status, and calculate the deviation; 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 = 0.94, B = 0.06, For filter gain, when signal-to-noise ratio ≥ 30dB =0.78, signal-to-noise ratio <30dB =0.58; This is the estimated state value at the current moment. This is the estimated state value from the previous moment; To control the input amount, 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 quantum bit mapping encoding is: temperature inside the carriage. Tin Mapping 6-bit qubits, relative humidity Hin Mapping 4-bit qubits, passenger density P Mapping 4 qubits, outdoor temperature Tout Mapping 4 qubits; the expression for the quantum superposition state is: ， Where α and β are quantum probability amplitudes and α² + β² = 1, | Tin >、∣ Hin >、∣ P >、∣ Tout > represents the baseline quantum state, | Tin ′>、∣ Hin ′>、∣ P ′>、| Tout '> is the perturbation compensation quantum state.

4. The method according to claim 3, characterized in that, The expression for the quantum fuzzy membership function in step S4 is: ， in, , It is an 18-dimensional quantum probability density matrix; The variance of the quantum fuzzy membership function is dynamically adjusted according to the application scenario. It is set to 0.1 for normal scenarios and 0.3 for special scenarios. The special scenarios include: passenger density > 8 people / m² or < 2 people / m², and temperature inside the carriage > 32℃ or < 14℃.

5. The method according to claim 1, characterized in that, In step S4, the quantum fuzzy rule base includes more than 152 rules, covering all operating conditions of cooling, heating, ventilation, and dehumidification. The core rules include at least: Rule 1: If Tin≥28℃, Hin≥60%, P≥7 people / m², and Tout≥30℃, then the compressor frequency increment is +5Hz, the fresh air ratio is +10%, and the solid contactor remains stably conductive. Rule 2: If Tin≤18℃, Hin≤40%, P≤3 people / m², and Tout≤5℃, then the compressor frequency increment is -3Hz, the fresh air ratio is -8%, and the fan drive current is reduced. Rule 3: If the passenger density changes by ≥2 people / m² for two consecutive collection cycles, immediately maintain the current compressor frequency, temporarily increase the fresh air ratio by 5%, and start solid contactor status monitoring; Rule 4: If the current of the solid-state contactor suddenly changes by ≥5A or the temperature by ≥110℃, the overcurrent / overtemperature protection will be triggered, the compressor frequency will drop by 10Hz, and the fault log will be recorded.

6. The method according to claim 1, characterized in that, In step S4, the initial temperature of the quantum annealing algorithm is set to 9.5, and the temperature decreases by 0.45 with each iteration. The iteration terminates when the rule matching error is ≤0.

009.

7. The method according to claim 1, characterized in that, The air conditioning control commands include compressor frequency, fan speed, electronic expansion valve opening, fresh air ratio, and solid-state contactor drive signal.

8. The method according to claim 3, characterized in that, The interval division of the qubit mapping in step S3 includes: the temperature inside the carriage is mapped to 6 qubits, with each quantum state corresponding to an interval of approximately 0.39℃; the relative humidity is mapped to 4 qubits, with each quantum state corresponding to an interval of 3.125%; the passenger density is mapped to 4 qubits, with each quantum state corresponding to an interval of 0.625 people / m²; the outdoor temperature is mapped to 4 qubits, with each quantum state corresponding to an interval of 6.875℃; and the compensation amount of the disturbance compensation quantum state is 10% of the corresponding variable quantum state interval.

9. The method according to claim 1, characterized in that, In step S6, an emergency adjustment strategy is triggered if the membership degree is less than 0.2 or the solid-state contactor fails, and fault data is recorded. The emergency adjustment strategy includes: increasing the variance to 0.4, increasing the initial quantum annealing temperature to 11.5, shortening the iteration step size to 0.35, strengthening the adjustment command, and shortening the feedback cycle to 5ms. When the solid-state contactor fails, the backup channel is switched and fault data is recorded.

10. A quantum fuzzy control system for an onboard air conditioner in 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 is used to collect the operating data of the vehicle air conditioner in real time, including the temperature inside the vehicle, the relative humidity inside the vehicle, the passenger density, and the outdoor temperature. 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 membership function optimization unit, a quantum fuzzy inference unit, and a rule base storage unit. It is used to realize quantum mapping encoding, superposition state construction, membership function optimization, storage of fuzzy rules, and generation of control instructions. The instruction execution module includes a SiC power module, a compressor drive circuit, a fan control unit, an electronic expansion valve drive interface, a solid-state contactor, and a protection circuit; the solid-state contactor uses semiconductor switching devices and has hardware-level protection functions for overcurrent, overvoltage, and overtemperature. The closed-loop feedback module includes a data feedback interface and a deviation judgment unit, which is used to collect air conditioning operation data, judge deviations and trigger closed-loop control, and also integrates a fault recording unit to store fault data.

Images