Direct-current power supply module parallel operation current sharing control method

Abstract

The invention discloses a direct-current power supply module parallel operation current-sharing control method, which belongs to the technical field of power electronics and comprises the following steps of: acquiring a maximum virtual impedance coefficient of an LLC (Logical Link Control) resonant circuit for performing a droop control process; cAN communication is synchronously started to obtain the number of online devices, the total load size is calculated based on the number of the online devices, and initial voltage compensation is carried out on the devices based on the total load size; the actual current sum of the online equipment after initial voltage compensation is obtained, and the current sharing value of the online equipment is calculated based on the current sum and the number of the online equipment; selecting equipment with the minimum output current value for compensation based on the current sharing value, and stopping compensation until the current sharing error is within 3%; according to the CAN communication, slow gradient compensation is carried out on different loads through the compensation coefficient of droop control, the problem of poor current-sharing degree can be effectively solved, the accuracy of the output voltage can be corrected, and the current-sharing degree and the accuracy of the output voltage can be effectively ensured when the loads are suddenly changed.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronics, and in particular to a method for controlling parallel current sharing of direct current power supply modules. Background Art

[0002] DC power supply parallel current sharing technology is a technology that evenly distributes the output current of each module by connecting multiple power modules in parallel and adopting a specific control strategy in a DC power supply system. This technology is particularly important in high-power DC power supply systems because it can significantly improve the reliability, stability and scalability of the system.

[0003] With the rapid development of power electronics technology, DC power supplies are increasingly used in various fields. However, the output power of a single power module is often limited and difficult to meet the needs of high-power systems. Therefore, connecting multiple power modules in parallel and achieving current sharing has become an effective means to improve system output power and reliability.

[0004] The core of the DC power supply parallel current sharing technology is to make the output current of each parallel power module equal or distributed in a certain proportion through control strategy. This is usually achieved by detecting the output current of each module and adjusting the output voltage or current of each module accordingly.

[0005] There are many ways to implement the DC power supply parallel current balancing technology, mainly including the following: changing the output internal resistance method (droop control); master / slave control method (Master / Slave); average current automatic current balancing method; maximum current automatic current balancing method (democratic current balancing method); external controller method.

[0006] Droop control is also called virtual impedance method. It is a commonly used technology in parallel current sharing. It simulates the power-frequency droop characteristics in traditional power systems and introduces virtual impedance to adjust the output voltage and current of the power supply, thereby achieving balanced power distribution among multiple power supplies and stabilizing the grid voltage.

[0007] However, this method also has some disadvantages, mainly including the following aspects:

[0008] (1) Line impedance difference: In a DC power supply system, the line impedance between power modules may be different due to factors such as cable length and material. This difference will limit the current sharing accuracy of droop control in practical applications, especially in light load or large cable impedance differences, the output current deviation of each module may increase.

[0009] (2) Selection of droop curve coefficient: The selection of droop curve coefficient will also affect the current sharing accuracy. A larger droop curve coefficient can improve the shunt accuracy, but will increase the voltage difference; while a smaller droop curve coefficient can reduce the voltage difference, but the shunt accuracy will decrease. Therefore, when selecting the droop curve coefficient, a compromise needs to be made between the shunt accuracy and the voltage deviation.

[0010] (3) System stability: The droop control virtual impedance method may cause system stability to deteriorate in some cases. For example, when the system load changes rapidly, each power module needs to quickly adjust its output voltage and current to maintain current sharing, which may cause system oscillation or instability.

[0011] (4) Complex control strategy: The droop control virtual impedance method requires complex control algorithms and calculation processes to achieve accurate current sharing and stable control, which increases the complexity of the system and the difficulty of design.

[0012] (5) Application scenario limitations: The droop control virtual impedance method may not achieve the desired control effect in certain specific application scenarios. For example, this method may not meet the requirements when the load variation range is large, the output voltage accuracy requirement is high, or high-precision current sharing is required.

[0013] In summary, the virtual impedance method of DC droop control in parallel current sharing has certain advantages and application value, but it also has disadvantages such as limited current sharing accuracy, stability problems, complexity and application scenario restrictions, and limitations in practical applications. Summary of the invention

[0014] In order to solve the problems of the prior art, the present invention provides a DC power supply module parallel current sharing control method, the method comprising the following steps:

[0015] Step 1, obtaining the maximum virtual impedance coefficient of the LLC resonant circuit, obtaining the virtual impedance magnitude based on the product of the maximum virtual impedance coefficient and the LLC resonant circuit current, and then applying the virtual impedance magnitude to the voltage loop in the power module to implement the droop control process;

[0016] Step 2: Synchronously start CAN communication to obtain the number of online devices, calculate the total load size based on the number of online devices, and perform initial voltage compensation on the devices based on the total load size;

[0017] The actual total current of the online devices after initial voltage compensation is obtained, and the current average value of the online devices is calculated based on the total current and the number of the online devices;

[0018] Step 3: Based on the current averaging value, select the device with the smallest output current value for compensation, and stop compensation when the averaging error is within 3%-5%.

[0019] Furthermore, in step 1, the maximum virtual impedance coefficient is obtained by the virtual impedance method, wherein the maximum droop voltage obtained by multiplying the rated output voltage by 0.05% is 11V, the rated current of the equipment is 2A, and the maximum virtual impedance coefficient is calculated to be 5.5 by dividing the maximum droop voltage by the rated current. Based on the fact that the larger the droop coefficient, the greater the droop voltage when the parallel machines are connected for current sharing, the maximum virtual impedance coefficient is selected as 1.5 to reduce the voltage difference.

[0020] Furthermore, in step 2, CAN communication includes multiple devices. When starting, the first started device is the host, and the other devices are slaves, so that the overall setting of one master and multiple slaves is realized.

[0021] Furthermore, in step 2, the droop control process is continued to prevent uneven current flow caused by the failure of CAN communication to start or sudden disconnection.

[0022] Furthermore, in step 2, the process of obtaining the load size is to obtain the actual total current and the rated total current of the current online device according to the CAN communication, and obtain the load size by dividing the rated total current by the actual total current.

[0023] Furthermore, in order to increase the current-sharing effect of the parallel machines, the master-slave control method can be used to fine-tune the equipment;

[0024] The delt value is calculated through CAN communication. The delt value is calculated as the current value at the current moment - (high current value + low current value) × 10%. The delt value is the current of the host. The current value is obtained by subtracting the current of the slave from the current of the host, and then the current value is applied to the voltage ring of the power module.

[0025] Furthermore, the power supply in the DC power supply module adopts current and voltage dual-loop control, wherein the dual loops are a voltage loop and a current loop, wherein the voltage is an outer loop and the current is an inner loop, and the output voltage accuracy is compensated and the current balance is improved by increasing the target value of the voltage loop.

[0026] Furthermore, in step 2, when initially compensating each device, the voltage is increased by 0.005 each time until the initial compensation coefficient of each device is adjusted to the corresponding compensation coefficient.

[0027] Furthermore, in step 3, after stopping compensation, if it is detected that the load is suddenly unloaded to 30% of the total load, the compensation coefficients of all devices are simultaneously reduced to a constant 1.01, delayed for a period of time, and if there is any change, the cycle is repeated in sequence. When the device is powered off, the compensation coefficient is restored to the initial state.

[0028] Beneficial effects of the present invention:

[0029] The CAN communication of the present application uses the compensation coefficient of droop control to perform slow gradient compensation for different loads, which can effectively solve the problem of difference in average current and correct the output voltage accuracy. When the load changes suddenly, it can also effectively ensure the accuracy of average current and output voltage. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of a flow chart of a method for controlling parallel current sharing of DC power supply modules provided by the present invention;

[0031] Figure 2 This is a schematic diagram of the current sharing control principle provided by the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] The DC power module parallel current sharing control method of the present invention is applied to power supply current sharing when multiple power modules are connected in parallel. Each power module includes a power conversion part and a main control module. During the implementation of the present invention, an LLC resonant circuit is used as the power conversion module, and each power module is connected via CAN communication.

[0034] The power supply system adopts master-slave current sharing control, that is, one module is selected as the master module among several modules, and the others are slave modules. Each slave module performs current sharing adjustment according to the set current of the master module.

[0035] See also Figure 1-2 The present invention provides a method for controlling the current sharing of a DC power supply module in parallel, the method comprising the following steps:

[0036] Step 1: Start the LLC resonant circuit to obtain the maximum virtual impedance coefficient for the droop control process;

[0037] The virtual impedance size is obtained based on the product of the maximum virtual impedance coefficient and the LLC resonant circuit current, and then the virtual impedance size is applied to the voltage loop in the power module to realize the droop control process;

[0038] Among them, this circuit is an LLC resonant circuit, the output voltage is 400VDC-220VDC, the maximum virtual impedance coefficient is calculated according to the virtual impedance method, the maximum droop voltage is 0.05%, the maximum droop voltage is calculated to be 11V, the rated current of the machine is 2A, the maximum droop voltage is divided by the rated current to calculate the maximum virtual impedance coefficient is 5.5, the larger the droop coefficient, the greater the voltage difference, according to the experiment, the maximum virtual impedance coefficient of this machine is selected to be 1.5.

[0039] Because the output voltage accuracy must be met, including the current balance problem, this circuit uses CAN communication to parallel current balance based on droop control, and real-time adjustment of the current balance instability and output voltage accuracy. The output voltage accuracy is 1% accuracy, which is calculated to be 2V. If the output voltage reaches 220VDC within plus or minus 2V, the accuracy is up to standard.

[0040] Experimental demonstration found that: through the formula U=IR, it is concluded that voltage accuracy compensation will also affect the current sharing problem. When the load is constant, increasing the voltage will also increase the current. The compensation method is to compensate for the voltage. The power supply adopts current and voltage dual-loop control, which includes voltage loop and current loop. Among them, voltage is the outer loop and current is the inner loop. By increasing the target value of the voltage loop, the output voltage accuracy is compensated and the current sharing is improved.

[0041] Step 2: CAN communication current sharing process:

[0042] When the LLC resonant circuit is started, the CAN communication is started at the same time to calculate the delt value, and the current sharing process of the CAN communication is realized based on the delt value;

[0043] Among them, CAN communication includes multiple devices. At startup, the first started device is the host, and the other devices are slaves, realizing the overall setting of one master and multiple slaves.

[0044] When the delt value is calculated for CAN communication, the droop control process is carried out at the same time to prevent uneven current due to the failure of CAN communication to start or sudden disconnection. The delt value is the current value at the current moment - (high current value + low current value) × 10%. The delt value is the current of the host, which is obtained by subtracting the current of the slave from the current of the host. The current value is then applied to the target value of the voltage ring of the power module to achieve the effect of current sharing for CAN communication.

[0045] Obtaining the load size according to CAN communication, the process of obtaining the load size is to obtain the actual total current and the rated total current of the current online device according to CAN communication, obtain the load size by dividing the rated total current by the actual total current, and perform an initial compensation voltage on each device based on the load size;

[0046] Among them, the compensation voltage size is different for different load sizes, so as to improve the current balance and output voltage accuracy under different load conditions.

[0047] After the power-on output is stable, CAN communication performs voltage compensation. According to the calculated percentage of different loads in the total load, a gradient initial compensation voltage is performed. When performing compensation, each device increases the compensation by 0.005 each time until the initial compensation coefficient of each device is adjusted to the corresponding compensation coefficient. After CAN communication calculates the sum of all online devices, the average current value is obtained, and the smallest current device is selected based on the average current value for current sharing compensation. The compensation increases by 0.001 each time, and compensation is stopped when the current sharing error is less than 3%-5%.

[0048] The division of compensation coefficients is shown in the following table:

[0049]

[0050] The above table shows the gradient compensation coefficient. When the CAN communication detects that the load has been suddenly unloaded, which is 30% of the total load, the compensation coefficients of all devices are simultaneously reduced to 1.01 and kept constant, and a delay is given. If there is a change, it will be cycled in sequence. When the device is powered off, the compensation coefficient will be restored to the initial state.

[0051] It should be noted that the gradient compensation of the compensation coefficient adjusted by CAN communication can effectively solve the problems of average current difference and output voltage accuracy. Because the compensation coefficient compensates the target value of the voltage loop, U=IR, when the load remains unchanged and the voltage value is increased, the current value will also increase. The starting value of the segmented compensation increases by 0.005 each time, which effectively compensates for the output voltage accuracy problem. Compared with the average current, the smallest current is detected and compensated by 0.001 each time. When the average current difference is less than 3%-5%, the compensation is stopped, which can effectively solve the average current problem. Slow segmented compensation can solve the problem of system instability after one-time compensation.

[0052] See also Figure 2 , the host current minus the slave current gets the current value, and then the current value is applied to the voltage loop, the delt value acts on the current loop, and the compensation coefficient of the droop control is on the voltage loop;

[0053] The difference between the host current and the slave current (current value) is then sent to the voltage loop through PI regulation. Normally, it is a dual-loop control, namely a voltage loop and a current loop. Based on the original dual-loop control, this application obtains the current value by subtracting the slave current from the host current, and applies the current value to the voltage loop to form a third loop, thereby forming a three-loop control as a whole. The output of the voltage loop is given by the current loop, and then the output of the current loop generates a PWM wave through a ramp.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling the current sharing of parallel DC power supply modules, characterized in that: The method comprises the following steps: Step 1, obtaining the maximum virtual impedance coefficient of the LLC resonant circuit, obtaining the virtual impedance magnitude based on the product of the maximum virtual impedance coefficient and the LLC resonant circuit current, and then applying the virtual impedance magnitude to the voltage loop in the power module to implement the droop control process; Step 2: Synchronously start CAN communication to obtain the number of online devices, calculate the total load size based on the number of online devices, and perform initial voltage compensation on the devices based on the total load size; The actual total current of the online devices after initial voltage compensation is obtained, and the current average value of the online devices is calculated based on the total current and the number of the online devices; Step 3: Based on the current averaging value, select the device with the smallest output current value for compensation, and stop compensation when the averaging error is within 3%-5%.

2. The DC power module parallel current sharing control method according to claim 1, characterized in that: In step 1, the maximum virtual impedance coefficient is obtained by the virtual impedance method, wherein the maximum droop voltage is 11V obtained by multiplying the rated output voltage by 0.05%, the rated current of the equipment is 2A, and the maximum virtual impedance coefficient is calculated to be 5.5 by dividing the maximum droop voltage by the rated current. The larger the droop coefficient, the greater the droop voltage when the parallel machines are current-sharing. In order to reduce the voltage difference, the maximum virtual impedance coefficient is selected to be 1.

5.

3. The DC power module parallel current sharing control method according to claim 1, characterized in that: In step 2, CAN communication includes multiple devices. At startup, the first device started is the master, and the other devices are slaves, realizing a one-master-multiple-slave setting as a whole.

4. The DC power module parallel current sharing control method according to claim 3, characterized in that: In step 2 shown, in order to prevent uneven current due to failure of CAN communication to start or sudden disconnection, the droop control process is continued.

5. The method for controlling the current sharing of parallel DC power supply modules according to claim 1, characterized in that: In step 2, the process of obtaining the load size is to obtain the actual total current and the rated total current of the current online device according to the CAN communication, and obtain the load size by dividing the rated total current by the actual total current.

6. The DC power supply module parallel current sharing control method according to claim 3, characterized in that: In order to increase the current sharing effect of the parallel machines, the master-slave control method can be used to fine-tune the equipment; The delt value is calculated through CAN communication. The delt value is calculated as the current value at the current moment - (high current value + low current value) × 10%. The delt value is the current of the host. The current value is obtained by subtracting the current of the slave from the current of the host, and then the current value is applied to the voltage ring of the power module.

7. The method for controlling the current sharing of parallel DC power supply modules according to claim 6, characterized in that: The power supply in the DC power supply module adopts current and voltage dual-loop control, wherein the dual loops are a voltage loop and a current loop, wherein the voltage is the outer loop and the current is the inner loop. The output voltage accuracy is compensated and the current balance is improved by increasing the target value of the voltage loop.

8. The DC power supply module parallel current sharing control method according to claim 1, characterized in that: In step 2, when initially compensating each device, the voltage is increased by 0.005 each time until the initial compensation coefficient of each device is adjusted to the corresponding compensation coefficient.

9. The method for controlling the current sharing of parallel DC power supply modules according to claim 1, characterized in that: In step 3, after stopping compensation, if it is detected that the load is suddenly unloaded to 30% of the total load, the compensation coefficients of all devices are simultaneously reduced to a constant 1.01, delayed for a period of time, and if there is any change, the cycle is repeated in sequence. When the device is powered off, the compensation coefficient is restored to the initial state.

Images