A control method and system for a high-speed air compressor for fuel cells

Abstract

The application provides a control method and system of a high-speed air compressor for a fuel cell, which doubles the angle resolution of the air compressor within the same control frequency, is beneficial to more accurately control the operation of the air compressor, and improves the control performance, and comprises the following steps: a first processor samples signals including current, voltage and temperature, the first processor processes data and calculates to obtain a rotor position angle of the air compressor; while the first processor operates, when a set misalignment time arrives, a second processor is triggered to sample signals including current, voltage and temperature, the second processor processes the sampled data and calculates to obtain a rotor position angle of the air compressor, and the rotor position angles of the air compressor obtained by the first processor and the second processor are combined to control the air compressor.

Description

Technical Field

[0001] This invention relates to the field of fuel cell air compressor technology, and specifically to a control method and system for a high-speed air compressor for fuel cells. Background Technology

[0002] When a fuel cell system is working, fuel (hydrogen, methanol, etc.) and oxygen (usually air) are fed into the core reaction unit stack. After chemical reaction, it outputs electrical and thermal energy and discharges the final products.

[0003] To improve the operating efficiency of fuel cell systems, the reactants need to be pressurized, and air compressors are generally used for this purpose.

[0004] Centrifugal air compressors are commonly used in fuel cells, operating at very high speeds. Because these compressors typically employ air bearings, their structure is not suitable for mounting position sensors; therefore, sensorless control strategies are commonly used. These requirements place certain demands on the air compressor controller.

[0005] (a) Control frequency: The operating speed of air compressors is mostly above 100,000 rpm, and for low-power scenarios it may even reach 200,000 rpm, requiring the controller's control frequency to be at least 30kHz.

[0006] (b) Sensorless control strategy: High-performance algorithms are required to obtain accurate air compressor rotor position;

[0007] In order to meet the above requirements, the following solutions are commonly used in existing technologies:

[0008] (1) Use high-performance chips: Use high-performance chips to meet the requirements of air compressor algorithms by increasing computing power;

[0009] (2) Reduce control frequency: By reserving more calculation time, the controller's algorithm can be completed;

[0010] (3) Optimize software execution time: Optimize the code in the software to reduce the software execution time;

[0011] While the above three strategies can meet the requirements of air compressor controllers to some extent, they all have certain drawbacks:

[0012] (1) Using high-performance chips physically improves the chip's computing power, but this also increases the cost.

[0013] (2) Reducing the control frequency also reduces the calculation frequency of the air compressor rotor angle, thereby reducing the rotor position resolution, which is essentially a compromise. The drawback of the reduced rotor position accuracy is the decrease in control accuracy, which corresponds to an increase in the power of the air compressor, resulting in a decrease in the overall efficiency of the fuel cell system;

[0014] (3) Optimizing software execution time can bring some improvement, but the improvement is generally no more than 20%, and the overall effect is limited. Summary of the Invention

[0015] To address the aforementioned problems, this invention provides a control method and system for a high-speed air compressor used in fuel cells. Within the same control frequency, the angular resolution of the air compressor is doubled, which facilitates more precise control of the air compressor's operation and improves control performance.

[0016] The technical solution is as follows: a control method for a high-speed air compressor for fuel cells, characterized by comprising the following steps:

[0017] The first processor samples the signals, including current, voltage, and temperature. The first processor then processes and calculates the data to obtain the rotor position angle of the air compressor.

[0018] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to sample. The sampled signals include current, voltage, and temperature. The second processor processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angles of the air compressor obtained by the first processor and the second processor.

[0019] Furthermore, the misalignment time is half of the air compressor drive clock cycle.

[0020] Furthermore, the first processor and the second processor are different cores of a multi-core processor.

[0021] Furthermore, when performing sampling, the sampling methods include oversampling and single-point sampling.

[0022] Further data processing and calculations are performed, including:

[0023] Process the sampled data;

[0024] The air compressor control algorithm is executed to calculate the output voltage signal of the air compressor controller;

[0025] Perform modulation calculations to calculate the PWM signal comparison value;

[0026] The sensorless algorithm for the air compressor is executed to calculate the rotor position angle of the air compressor.

[0027] Furthermore, the first processor further includes a first coprocessor, and the second processor further includes a second coprocessor, wherein the first coprocessor and the second coprocessor assist the first processor and the second processor in performing data processing and calculation, respectively.

[0028] Furthermore, the plan will be implemented according to the following steps:

[0029] The first processor is triggered by the sampling clock to oversample the current and voltage.

[0030] When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the voltage and current and switches to single-point sampling of the temperature.

[0031] The first coprocessor is triggered to process the oversampling results;

[0032] After completing single-point temperature sampling, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0033] After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0034] After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0035] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current and voltage.

[0036] When the air compressor drive clock signal reaches its highest point, the second processor stops rapidly sampling the voltage and current, and instead performs single-point sampling of the temperature.

[0037] The second coprocessor is triggered to process the oversampling results;

[0038] After completing single-point temperature sampling, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0039] After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0040] After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0041] Furthermore, the plan will be implemented according to the following steps:

[0042] The first processor is triggered by the sampling clock to oversample the current, voltage, and temperature;

[0043] When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling;

[0044] The first coprocessor is triggered to process the sampling results;

[0045] The first processor executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller.

[0046] After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0047] After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0048] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current, voltage, and temperature.

[0049] When the air compressor drive clock signal reaches its highest point, the second processor stops oversampling;

[0050] The second coprocessor is triggered to process the sampling results;

[0051] The second processor executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller.

[0052] After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0053] After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0054] Furthermore, the plan will be implemented according to the following steps:

[0055] The first processor is triggered by the sampling clock to perform single-point sampling of current, voltage, and temperature;

[0056] After sampling is completed, the first processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0057] After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0058] After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0059] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to perform single-point sampling of current, voltage, and temperature.

[0060] After sampling is completed, the second processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0061] After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0062] After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0063] Furthermore, the plan will be implemented according to the following steps:

[0064] The first processor is triggered by the sampling clock to oversample the current.

[0065] When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the current and switches to single-point sampling of the voltage and temperature.

[0066] The first coprocessor is triggered to process the oversampling results;

[0067] After completing single-point sampling of voltage and temperature, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0068] After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0069] After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0070] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current.

[0071] When the air compressor drive clock signal reaches its highest point, the second processor stops rapidly sampling the current and switches to single-point sampling of voltage and temperature.

[0072] The second coprocessor is triggered to process the oversampling results;

[0073] After completing single-point sampling of voltage and temperature, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0074] After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value.

[0075] After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0076] Furthermore, the plan will be implemented according to the following steps:

[0077] The first processor is triggered by the sampling clock to oversample the current and voltage.

[0078] When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the voltage and current and switches to single-point sampling of the temperature.

[0079] The first coprocessor is triggered to process the oversampling results;

[0080] After completing single-point temperature sampling, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0081] After the first processor calculates the output voltage signal of the air compressor controller, the first processor performs modulation calculation to calculate the PWM signal comparison value;

[0082] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0083] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current and voltage.

[0084] When the air compressor drive clock signal reaches its highest point, the second processor stops rapidly sampling the voltage and current, and instead performs single-point sampling of the temperature.

[0085] The second coprocessor is triggered to process the oversampling results;

[0086] After completing single-point temperature sampling, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0087] After the second processor calculates the output voltage signal of the air compressor controller, the second processor performs modulation calculation to calculate the PWM signal comparison value;

[0088] After completing the modulation calculation, the second processor executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0089] Furthermore, the plan will be implemented according to the following steps:

[0090] The first processor performs single-point sampling of current, voltage, and temperature;

[0091] After sampling is completed, the first processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0092] After calculating the output voltage signal of the air compressor controller, the first processor performs modulation calculation to calculate the PWM signal comparison value;

[0093] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0094] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to perform single-point sampling of current, voltage, and temperature.

[0095] After sampling is completed, the second processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0096] After calculating the output voltage signal of the air compressor controller, the second processor performs modulation calculation to calculate the PWM signal comparison value;

[0097] After completing the modulation calculation, the second processor executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0098] A control system for a high-speed air compressor for fuel cells, characterized in that it comprises: a first processor and a second processor.

[0099] The first processor samples signals including current, voltage, and temperature. The first processor then processes and calculates the data to obtain the rotor position angle of the air compressor.

[0100] While the first processor is running, when the set misalignment time is reached, the second processor is triggered to sample. The sampled signals include current, voltage, and temperature. The second processor processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angles of the air compressor obtained by the first processor and the second processor.

[0101] The control method and system for a high-speed air compressor for fuel cells of the present invention can achieve a frequency multiplication effect by utilizing the staggered execution of multiple CPUs. Within the same control frequency, the number of air compressor control executions is doubled, thereby doubling the angular resolution of the air compressor. This facilitates more precise control of the air compressor's operation and improves control performance. Without increasing the computing power of the chip, the performance of the air compressor controller is doubled, which can improve the utilization efficiency of hardware resources and avoid increased costs and chip supply risks. Through parallel computing of the processor and coprocessor, the single-cycle execution time can be compressed, effectively improving computing efficiency and enhancing control performance. Attached Figure Description

[0102] Figure 1 This is a schematic diagram of the control system module for a high-speed air compressor for fuel cells, as shown in the embodiment.

[0103] Figure 2 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 1;

[0104] Figure 3 A comparison of the efficiency of traditional air compressor architecture execution and the method proposed in Example 1 in air compressor execution;

[0105] Figure 4 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 2;

[0106] Figure 5 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 3;

[0107] Figure 6 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 4;

[0108] Figure 7 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 5;

[0109] Figure 8 This is a schematic diagram of the control method and timing sequence for a high-speed air compressor for a fuel cell in Example 6. Detailed Implementation

[0110] The present invention provides a control method for a high-speed air compressor for fuel cells, comprising the following steps:

[0111] The first processor samples signals including current, voltage, and temperature. Sampling methods include oversampling and single-point sampling. The first processor then processes and calculates the data to obtain the rotor position angle of the air compressor. In this embodiment, the data processing and calculation include: processing the sampled data; executing an air compressor control algorithm to calculate the output voltage signal of the air compressor controller; performing modulation calculations to calculate the PWM signal comparison value; and executing a sensorless air compressor algorithm to calculate the rotor position angle of the air compressor.

[0112] While the first processor is running, the second processor is triggered to sample signals, including current, voltage, and temperature, once the set misalignment time is reached. The second processor processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angles obtained by the first and second processors. In this embodiment, by utilizing the misalignment execution of the first and second processors, the number of air compressor control executions can be doubled within the same control frequency, thus doubling the air compressor angle resolution. This can significantly improve the accuracy of sensorless control, reduce air compressor power loss, and help optimize the overall efficiency of the fuel cell system. The control frequency doubling effect is achieved without requiring additional chip computing power.

[0113] Meanwhile, the dual processors implement backup redundancy and can verify each other during operation, which can improve the reliability of the software. In the event of any software calculation failure, it can quickly switch to single CPU operation and has fault-tolerant control capabilities.

[0114] In the embodiments, the first processor and the second processor can be separate chips or different cores of at least a dual-core chip. Whether using a single chip or a multi-core chip, staggered execution can be achieved using lower-cost chips, avoiding reliance on high-end chips, significantly reducing hardware costs, and mitigating supply chain risks. The method of this invention does not require high-end, high-performance chips. Through staggered execution of multiple CPUs and parallel computation of multiple CPUs and coprocessors, it still meets the chip computing power requirements of the air compressor algorithm, avoiding increased costs and chip supply risks. Simultaneously, it avoids problems such as decreased air compressor rotor angle calculation frequency, reduced rotor position resolution, and decreased control accuracy due to reduced control frequency, thus preventing increased air compressor power and reduced overall fuel cell system efficiency.

[0115] In one embodiment of the present invention, the first processor may further include a first coprocessor, and the second processor may further include a second coprocessor. The first coprocessor and the second coprocessor assist the first processor and the second processor in data processing and calculation, respectively. Through the parallel calculation of the processor CPU and the coprocessor CLA, the single-cycle execution time can be compressed. By allowing the processor CPU and the coprocessor CLA to process different tasks in parallel, including oversampling, slow signal sampling, modulation calculation, and sensorless algorithm calculation, the calculation efficiency can be effectively improved. In the present invention, the air compressor control algorithm, modulation calculation, and air compressor sensorless algorithm all adopt existing mature methods, which will not be described in detail here.

[0116] See Figure 2 The following describes the operation steps of the method of the present invention through different embodiments. Figure 2 In the diagram, the red letters A through L indicate the position of each step. Figure 2 In the algorithm, Task 3 processes the sampled data; Task 1 performs modulation calculations to determine the PWM signal comparison value; and Task 2 executes the sensorless algorithm for the air compressor to calculate the rotor position angle. Figure 2 In the illustration, since steps A to F of the first processor CPU1 and steps G to L of the second processor CPU2 are not within the same cycle, steps G to L appear to precede steps A to F in terms of timing. Figures 4 to 8 The same logic applies to the situation where steps G to L occur before steps A to F in the time sequence.

[0117] In Example 1: The first processor CPU1 executes the following steps:

[0118] Step A: The sampling clock triggers the first processor CPU1 to oversample the current and voltage;

[0119] Step B: When the air compressor drive clock signal reaches its highest point, the first processor CPU1 stops oversampling the voltage and current, and instead performs single-point sampling of the temperature.

[0120] Step C: Trigger the first coprocessor CLA1 to process the oversampling results, such as removing DC bias, linearity calibration, numerical conversion, and filtering.

[0121] Step D: After completing single-point temperature sampling, the first processor CPU1 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller;

[0122] Step E: After the first processor CPU1 calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor CLA1 to perform modulation calculation and calculate the PWM signal comparison value.

[0123] Step F: After completing the modulation calculation, based on the calculated PWM signal comparison value, the first coprocessor CLA1 is triggered to execute the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0124] After the calculation is completed, the first processor CPU1 will repeat steps A to F to start the next loop.

[0125] The second processor, CPU2, executes the following steps:

[0126] Step G: While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to oversample the current and voltage. In this embodiment, the misalignment time is half of the air compressor drive clock.

[0127] Step H: When the air compressor drive clock signal reaches its highest point, the second processor CPU2 stops rapidly sampling the voltage and current, and instead performs single-point sampling of the temperature.

[0128] Step I: Trigger the second coprocessor CLA2 to process the oversampling results, such as removing DC bias, linearity calibration, numerical conversion, and filtering.

[0129] Step J: After completing single-point temperature sampling, the second processor CPU2 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller;

[0130] Step K: After the second processor CPU2 calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor CLA2 to perform modulation calculation and calculate the PWM signal comparison value;

[0131] Step L: After completing the modulation calculation, based on the calculated PWM signal comparison value, the second coprocessor CLA2 is triggered to execute the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0132] After the calculation is completed, the second processor CPU2 will follow the steps of the first processor CPU1, repeating the process from step G to step L, and starting the next loop.

[0133] Figure 3 A comparison is given between the traditional air compressor architecture and the method proposed in Example 1 in terms of air compressor execution. Figure 3Each colored block represents the execution time of different functional software. During comparison, it is assumed that the execution time of the software code for each module is the same. During comparison, the moment the PWM signal comparison value is obtained (calculated in the modulation calculation module) is represented by an orange "○", and the moment the rotor position angle is obtained (calculated in the sensorless algorithm execution module) is represented by an orange "X". The minimum interval between obtaining both signals is considered the highest control frequency supported by the corresponding architecture.

[0134] Figure 3 In traditional serial architectures, all software is executed in the CPU, resulting in the longest air compressor drive clock signal. With the parallel architecture (A-F) proposed in this invention, the parallel computation of CPU1 and CLA1 compresses some of the software execution time, thus compressing the corresponding air compressor drive clock signal. Using a dual-core architecture, with the parallel architecture (A-L) proposed in this invention, although the length of the air compressor drive clock signal is the same as in the parallel architecture (A-F), two PWM comparison values ​​("○" signal) and rotor position angle ("X" signal) are obtained within the same time frame, effectively performing a frequency multiplication operation. Therefore, the control frequency is "doubled," resulting in optimal control performance.

[0135] Example 2:

[0136] See Figure 4 In Example 2, the solution is implemented according to the following steps:

[0137] The sampling clock triggers the first processor CPU1 to oversample the current, voltage, and temperature;

[0138] When the air compressor drive clock signal reaches its highest point, the first processor CPU1 stops oversampling;

[0139] The first coprocessor CLA1 is triggered to process the sampling results;

[0140] The first processor CPU1 executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller;

[0141] After the first processor CPU1 calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor CLA1 to perform modulation calculation and calculate the PWM signal comparison value.

[0142] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first coprocessor CLA1 is triggered to execute the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0143] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to oversample the current, voltage, and temperature.

[0144] When the air compressor drive clock signal reaches its highest point, the second processor CPU2 stops oversampling;

[0145] The second coprocessor CLA2 is triggered to process the sampling results;

[0146] The second processor, CPU2, executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller.

[0147] After the second processor CPU2 calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor CLA2 to perform modulation calculation and calculate the PWM signal comparison value.

[0148] After completing the modulation calculation, the second coprocessor CLA2 is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0149] The difference between Example 2 and Example 1 is that in Example 2, all sampling of current, voltage, and temperature is performed by oversampling.

[0150] Example 3:

[0151] See Figure 5 In Example 3, the solution is implemented according to the following steps:

[0152] The sampling clock triggers the first processor CPU1 to perform single-point sampling of current, voltage, and temperature;

[0153] After sampling is completed, the first processor CPU1 executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0154] After the first processor CPU1 calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor CLA1 to perform modulation calculation and calculate the PWM signal comparison value.

[0155] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first coprocessor CLA1 is triggered to execute the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0156] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to perform single-point sampling of current, voltage, and temperature.

[0157] After sampling is completed, the second processor CPU2 executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0158] After the second processor CPU2 calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor CLA2 to perform modulation calculation and calculate the PWM signal comparison value.

[0159] After completing the modulation calculation, the second coprocessor CLA2 is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0160] The difference between Example 3 and Example 1 is that in Example 3, single-point sampling is selected for all sampling of current, voltage, and temperature.

[0161] Example 4:

[0162] See Figure 6 In Example 4, the solution is implemented according to the following steps:

[0163] The sampling clock triggers the first processor CPU1 to oversample the current.

[0164] When the air compressor drive clock signal reaches its highest point, the first processor CPU1 stops oversampling the current and switches to single-point sampling of the voltage and temperature.

[0165] The first coprocessor CLA1 is triggered to process the oversampling results, such as removing DC bias, linearity calibration, numerical conversion, and filtering.

[0166] After completing single-point sampling of voltage and temperature, the first processor CPU1 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0167] After the first processor CPU1 calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor CLA1 to perform modulation calculation and calculate the PWM signal comparison value.

[0168] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first coprocessor CLA1 is triggered to execute the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0169] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to oversample the current.

[0170] When the air compressor drive clock signal reaches its highest point, the second processor CPU2 stops rapidly sampling the current and instead performs single-point sampling of the voltage and temperature.

[0171] The second coprocessor CLA2 is triggered to process the oversampling results, such as removing DC bias, linearity calibration, numerical conversion, and filtering.

[0172] After completing single-point sampling of voltage and temperature, the second processor CPU2 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0173] After the second processor CPU2 calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor CLA2 to perform modulation calculation and calculate the PWM signal comparison value.

[0174] After completing the modulation calculation, the second coprocessor CLA2 is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

[0175] The difference between Example 4 and Example 1 is that in Example 4, oversampling is performed for current sampling, while single-point sampling is performed for voltage and temperature sampling.

[0176] Example 5:

[0177] See Figure 7 In Example 5, the solution is implemented according to the following steps:

[0178] The sampling clock triggers the first processor CPU1 to oversample the current and voltage.

[0179] When the air compressor drive clock signal reaches its highest point, the first processor CPU1 stops oversampling the voltage and current, and instead performs single-point sampling of the temperature.

[0180] The first coprocessor CLA1 is triggered to process the oversampling results;

[0181] After completing single-point temperature sampling, the first processor CPU1 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0182] After the first processor CPU1 calculates the output voltage signal of the air compressor controller, the first processor CPU1 performs modulation calculation to calculate the PWM signal comparison value;

[0183] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor CPU1 executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0184] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to oversample the current and voltage.

[0185] When the air compressor drive clock signal reaches its highest point, the second processor CPU2 stops rapidly sampling the voltage and current, and instead performs single-point sampling of the temperature.

[0186] The second coprocessor CLA2 is triggered to process the oversampling results, such as removing DC bias, linearity calibration, numerical conversion, and filtering.

[0187] After completing single-point temperature sampling, the second processor CPU2 interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller.

[0188] After the second processor CPU2 calculates the output voltage signal of the air compressor controller, the second processor CPU2 performs modulation calculation to calculate the PWM signal comparison value;

[0189] After completing the modulation calculation, the second processor CPU2 executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0190] The difference between Example 5 and Example 1 is that in Example 5, the modulation calculation and sensorless control algorithm are performed by the first processor CPU1 and the second processor CPU2.

[0191] Example 6:

[0192] See Figure 8 In Example 6, the solution is implemented according to the following steps:

[0193] The first processor, CPU1, performs single-point sampling of current, voltage, and temperature;

[0194] After sampling is completed, the first processor CPU1 executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0195] After calculating the output voltage signal of the air compressor controller, the first processor CPU1 performs modulation calculation to calculate the PWM signal comparison value;

[0196] After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor CPU1 executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor.

[0197] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to perform single-point sampling of current, voltage, and temperature.

[0198] After sampling is completed, the second processor CPU2 executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller;

[0199] After calculating the output voltage signal of the air compressor controller, the second processor CPU2 performs modulation calculation to calculate the PWM signal comparison value;

[0200] After completing the modulation calculation, the second processor CPU2 executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

[0201] A control system for a high-speed air compressor for fuel cells, characterized in that it comprises: a first processor CPU1 and a second processor CPU2.

[0202] The first processor CPU1 samples the signals, including current, voltage, and temperature. The first processor CPU1 processes and calculates the data to obtain the rotor position angle of the air compressor.

[0203] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CPU2 is triggered to sample. The sampled signals include current, voltage, and temperature. The second processor CPU2 processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angle of the air compressor obtained by the first processor CPU1 and the second processor CPU2.

[0204] The difference between Example 6 and Example 1 is that in Example 6, the first processor CPU1 and the second processor CPU2 perform all data processing and calculations, including: processing the sampled data; executing the air compressor control algorithm to calculate the output voltage signal of the air compressor controller; executing modulation calculation to calculate the PWM signal comparison value; and executing the air compressor sensorless algorithm to calculate the rotor position angle of the air compressor.

[0205] Examples 1 to 6 illustrate that the method of the present invention is applicable to any combination of different sampling methods. Example 2 retains only oversampling, Example 3 retains only single-point sampling, and Example 4 adjusts the signals of oversampling and single-point sampling. Any combination of the above can be incorporated into the method of the present invention.

[0206] The method of this invention, which allocates tasks for data processing and computation, is only one example. In application, the tasks executed in the CPU and CLA can be arbitrarily changed and combined. For example, in embodiment 5, the CPU performs modulation calculations and sensorless control algorithms. In embodiment 6, without using the CLA, only CPU1 and CPU2 can be used to implement the simplified software architecture proposed in this invention. The method proposed in this invention is not targeted at a specific chip model and can also be implemented using multiple chips.

[0207] See Figure 1 In an embodiment of the present invention, a control system for a high-speed air compressor for fuel cells is also provided, comprising: a first processor CPU1 and a second processor CPU2.

[0208] The first processor CPU1 samples the signals, including current, voltage, and temperature. The first processor CPU1 processes and calculates the data to obtain the rotor position angle of the air compressor.

[0209] While the first processor CPU1 is running, when the set misalignment time is reached, the second processor CUP2 is triggered to sample. The sampled signals include current, voltage, and temperature. The second processor CUP2 processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angle of the air compressor obtained by the first processor CPU1 and the second processor CUP2.

[0210] In addition, in the embodiment, the first processor CPU1 and the second processor CPU2 also include a first coprocessor CLA1 and a second coprocessor CLA2, respectively, which assist the first processor CPU1 and the second processor CPU1 in data processing and calculation.

[0211] like Figure 1 As shown, CPU1 and CPU2 are the two main cores of a dual-core chip. CPU1 and CPU2 can communicate directly with each other. Each CPU is equipped with a coprocessor CLA. The first coprocessor CLA1 and the second coprocessor CLA2 can run independently with the first processor CPU1 and the second processor CPU2 to achieve parallel code computation. Each coprocessor CLA can only interact with its corresponding CPU. In this embodiment, the system may also include an external communication module, an AD sampling module, a PWM drive module, an external storage module, an IO drive module, and a DA output module. The functions of each module are as follows:

[0212] External communication module: responsible for external communication, compatible with all types on the market, such as CAN communication, serial communication, RS485 communication, I2C communication, SPI communication, USB communication, LAN communication and any other communication type;

[0213] AD sampling module: Samples the actual physical signals of the air compressor, converts them into digital quantities, and sends them to the chip. Commonly sampled signals include phase current, output voltage, bus voltage, bus current, motor temperature, and controller temperature. Any analog signal sent to the chip can be converted from analog to digital by this module.

[0214] PWM drive module: Outputs the air compressor drive signal, typically a three-phase synchronous PWM signal;

[0215] External storage module: Interfaces for commonly used storage devices such as EEPROM, Flash, and SD cards;

[0216] IO driver module: Common input and output signal interfaces, such as LED display, switch signal, emergency stop signal, hardware version number, etc., which are two-state signals with only high and low levels;

[0217] PWM sampling module: Samples the PWM signal, obtains its frequency and duty cycle signal results, and sends them into the chip.

[0218] DA output module: converts the digital signals inside the chip into continuous analog signals for output;

[0219] Figure 1 This abstracts only some commonly used modules; in practical applications, there may be other modules that can be incorporated. Figure 1 The ellipsis indicates the core. In practical applications, the internal code of the dual-core processor CPU1 and CPU2 is completely identical, but one CPU needs to be designated as the master core for external interaction. In this example, CPU1 is selected as the master CPU. The corresponding peripheral modules—external communication, external storage, IO driver, PWM sampling, and DA output module—are all connected to the CPU1 Bus, and their data signals can be sent to CPU1 via the CPU1 Bus. The AD sampling module and the "PWM driver" are connected to both the CPU1 Bus and the CPU2 Bus, so their corresponding signals can be sent to both the first processor CPU1 and the second processor CPU2 simultaneously.

[0220] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

[0221] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A control method for a high-speed air compressor for fuel cells, characterized in that, Includes the following steps: It includes at least a first processor and a second processor. The first processor performs sampling, and the sampled signals include current, voltage and temperature. The first processor performs data processing and calculation to obtain the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to sample. The sampled signals include current, voltage and temperature. The second processor processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angle of the air compressor obtained by the first processor and the second processor. The misalignment time is half of the air compressor drive clock cycle; Data processing and calculations include: Process the sampled data; The air compressor control algorithm is executed to calculate the output voltage signal of the air compressor controller; Perform modulation calculations to calculate the PWM signal comparison value; The sensorless algorithm for the air compressor is executed to calculate the rotor position angle of the air compressor.

2. The control method for a high-speed air compressor for fuel cells according to claim 1, characterized in that: The first processor and the second processor are different cores of a multi-core processor.

3. The control method for a high-speed air compressor for fuel cells according to claim 1, characterized in that: When sampling, sampling methods include oversampling and single-point sampling.

4. The control method for a high-speed air compressor for fuel cells according to claim 1, characterized in that: The first processor further includes a first coprocessor, and the second processor further includes a second coprocessor. The first coprocessor and the second coprocessor respectively assist the first processor and the second processor in data processing and calculation.

5. The control method for a high-speed air compressor for fuel cells according to claim 4, characterized in that: The plan will be implemented according to the following steps: The first processor is triggered by the sampling clock to oversample the current and voltage. When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the voltage and current and switches to single-point sampling of the temperature. The first coprocessor is triggered to process the oversampling results; After completing single-point temperature sampling, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current and voltage. When the air compressor drive clock signal reaches its highest point, the second processor stops oversampling the voltage and current and switches to single-point sampling of the temperature. The second coprocessor is triggered to process the oversampling results; After completing single-point temperature sampling, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

6. The control method for a high-speed air compressor for fuel cells according to claim 4, characterized in that: The plan will be implemented according to the following steps: The first processor is triggered by the sampling clock to oversample the current, voltage, and temperature; When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling; The first coprocessor is triggered to process the sampling results; The first processor executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller. After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current, voltage, and temperature. When the air compressor drive clock signal reaches its highest point, the second processor stops oversampling; The second coprocessor is triggered to process the sampling results; The second processor executes the air compressor control algorithm and calculates the output voltage signal of the air compressor controller. After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

7. The control method for a high-speed air compressor for a fuel cell according to claim 4, characterized in that: The plan will be implemented according to the following steps: The first processor is triggered by the sampling clock to perform single-point sampling of current, voltage, and temperature; After sampling is completed, the first processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller; After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to perform single-point sampling of current, voltage and temperature. After sampling is completed, the second processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller; After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

8. The control method for a high-speed air compressor for fuel cells according to claim 4, characterized in that: The plan will be implemented according to the following steps: The first processor is triggered by the sampling clock to oversample the current. When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the current and switches to single-point sampling of the voltage and temperature. The first coprocessor is triggered to process the oversampling results; After completing single-point sampling of voltage and temperature, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the first processor calculates the output voltage signal of the air compressor controller, it triggers the first coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the first coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current. When the air compressor drive clock signal reaches its highest point, the second processor stops oversampling the current and switches to single-point sampling of the voltage and temperature. The second coprocessor is triggered to process the oversampling results; After completing single-point sampling of voltage and temperature, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the second processor calculates the output voltage signal of the air compressor controller, it triggers the second coprocessor to perform modulation calculation and calculate the PWM signal comparison value. After completing the modulation calculation, the second coprocessor is triggered to execute the sensorless algorithm of the air compressor based on the calculated PWM signal comparison value, and the rotor position angle of the air compressor is calculated.

9. A control method for a high-speed air compressor for fuel cells according to claim 4, characterized in that: The plan will be implemented according to the following steps: The first processor is triggered by the sampling clock to oversample the current and voltage. When the air compressor drive clock signal reaches its highest point, the first processor stops oversampling the voltage and current and switches to single-point sampling of the temperature. The first coprocessor is triggered to process the oversampling results; After completing single-point temperature sampling, the first processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the first processor calculates the output voltage signal of the air compressor controller, the first processor performs modulation calculation to calculate the PWM signal comparison value; After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to oversample the current and voltage. When the air compressor drive clock signal reaches its highest point, the second processor stops oversampling the voltage and current and switches to single-point sampling of the temperature. The second coprocessor is triggered to process the oversampling results; After completing single-point temperature sampling, the second processor interrupts the sampling, executes the air compressor control algorithm, and calculates the output voltage signal of the air compressor controller. After the second processor calculates the output voltage signal of the air compressor controller, the second processor performs modulation calculation to calculate the PWM signal comparison value; After completing the modulation calculation, the second processor executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

10. A control method for a high-speed air compressor for a fuel cell according to claim 4, characterized in that, The plan will be implemented according to the following steps: The first processor performs single-point sampling of current, voltage, and temperature; After sampling is completed, the first processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller; After calculating the output voltage signal of the air compressor controller, the first processor performs modulation calculation to calculate the PWM signal comparison value; After completing the modulation calculation, based on the calculated PWM signal comparison value, the first processor executes the sensorless algorithm for the air compressor to calculate the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to perform single-point sampling of current, voltage and temperature. After sampling is completed, the second processor executes the air compressor control algorithm to calculate the output voltage signal of the air compressor controller; After calculating the output voltage signal of the air compressor controller, the second processor performs modulation calculation to calculate the PWM signal comparison value; After completing the modulation calculation, the second processor executes the sensorless algorithm for the air compressor based on the calculated PWM signal comparison value to calculate the rotor position angle of the air compressor.

11. A control system for a high-speed air compressor for fuel cells, characterized in that, include: It includes at least a first processor and a second processor. The first processor samples signals including current, voltage, and temperature. The first processor then processes and calculates the data to obtain the rotor position angle of the air compressor. While the first processor is running, when the set misalignment time is reached, the second processor is triggered to sample. The sampled signals include current, voltage and temperature. The second processor processes the sampled data and calculates the rotor position angle of the air compressor. The air compressor is controlled by combining the rotor position angle of the air compressor obtained by the first processor and the second processor. The misalignment time is half of the air compressor drive clock cycle; Data processing and calculations include: Process the sampled data; The air compressor control algorithm is executed to calculate the output voltage signal of the air compressor controller; Perform modulation calculations to calculate the PWM signal comparison value; The sensorless algorithm for the air compressor is executed to calculate the rotor position angle of the air compressor.

Images