Bidirectional DCDC circuit uninterrupted control method

Abstract

The invention relates to an uninterrupted control method for a bidirectional DCDC circuit, the bidirectional DCDC circuit adopts a conventional switching device to form a novel and unique topological structure, the uninterrupted control method is matched, a set of PID is charged (discharged), continuous charging and discharging with wider voltage are realized, the control is simple, and the reliability is high.

Description

Technical Field

[0001] This invention belongs to the field of power electronic equipment, and specifically relates to a method for uninterrupted control of a bidirectional DC-DC circuit. Background Art

[0002] The battery energy storage industry has developed rapidly in recent years. To pursue higher energy density and reduce system costs, the voltage level of battery compartments is becoming increasingly higher. A few years ago, the voltage range of battery compartments was generally 200V to 650V, but in recent years the voltage has risen to 1000V or even higher. The corresponding battery compartment charging and discharging equipment generally consists of a power storage converter (PCS) + bidirectional DC-DC circuit. The power storage converter (PCS) can convert the three-phase AC380V of the grid into a stable DC terminal voltage of DC650V, and can also invert the energy from the DC 650V terminal to the three-phase grid. When the DC terminal voltage is less than DC650V, energy flows from the grid to the DC terminal; when the DC terminal voltage is greater than DC650V, energy flows from the DC terminal to the three-phase AC grid. To stabilize the DC terminal voltage, energy automatically flows bidirectionally; this is the main function of the PCS. A bidirectional DC-DC converter is a DC-to-DC converter. When charging the battery, the DC-DC power supply converts the DC 650V output voltage of the PCS into an adjustable voltage of DC 200V to DC 1500V to charge the battery. When discharging, the DC-DC power supply converts the battery terminal voltage to 650V to discharge the DC terminal. When the DC terminal voltage is greater than 650V, the PCS automatically inverts the DC terminal energy to the three-phase power grid.

[0003] Common bidirectional DC-DC electrical topologies include Figure 1 As shown, the DC terminal (left end) is connected to the DC terminal of the PCS, with a voltage of DC650V; the battery terminal of the bidirectional DC-DC converter has a voltage range of DC200V to DC1500V, which can be set. Energy can flow from the DC terminal to the battery terminal, and vice versa.

[0004] The current flowing from left to right in inductor L1 is considered the positive direction (charging direction), and from right to left is the reverse direction (discharging direction). When operating in the positive direction, the right end of the bidirectional DC / DC unit can output a DC voltage that can be either high or low: Q21 is constant, Q1 is adjustable, and Q2 freewheeling enables a step-down output; Q1 is constant, Q22 is adjustable, and Q21 freewheeling enables a step-up output. When operating in the reverse direction, the left end of the bidirectional DC / DC unit can output a DC voltage that can be either high or low: Q1 is constant, Q21 is adjustable, and Q22 freewheeling enables a step-down output; Q21 is constant, Q2 is adjustable, and Q1 freewheeling enables a step-up output.

[0005] The disadvantages of existing bidirectional DC-DC circuits:

[0006] 1. High cost:

[0007] from Figure 1 It can be seen that, regardless of whether the output voltage is stepped down in the forward direction or stepped up to DC 650V in the reverse direction, the maximum voltage across each of the Q1 and Q2 switching transistors is 650V. Following the principle of selecting transistors based on approximately twice the operating voltage, IGBTs with a rated voltage of DC 1200V can be used for Q1 and Q2. However, the situation is different on the battery side. Whether charging in the forward direction or discharging in the reverse direction, the Q21 and Q22 switching transistors will withstand a voltage of DC 1500V. Following the principle of selecting transistors based on twice the operating voltage, IGBTs with a voltage rating of DC 3000V are required for Q21 and Q22. DC 3000V IGBTs are unconventional products. Following the principle of selecting the nearest available option, a DC 3300V IGBT must be used, which is very expensive—approximately 10 times the price of a DC 1700V or DC 1200V IGBT.

[0008] 2. The switching transistor experiences high switching stress and is prone to damage.

[0009] The voltage across transistors Q21 and Q22 is DC 1500V. During switching on and off, the voltage across the transistors changes from 0V to 1500V when off, and from DC 1500V to 0V when on. If the switching time is calculated as 2µs, Such a high dv / dt can easily damage the IGBT.

[0010] 3. Repeated switching occurs during buck-boost transitions, causing repeated fluctuations in charging and discharging current:

[0011] If the battery needs to be charged from 400V to 800V, the power supply initially charges the battery at a reduced voltage (Q21 constant, Q1 regulating, Q2 freewheeling), achieving charging below DC 650V. As charging progresses, the battery voltage increases. When the battery voltage reaches DC 650V, Q21 constant, Q1 regulating, and Q2 freewheeling are insufficient for charging. Q1 constant, Q22 regulating, and Q21 freewheeling are then required to achieve boost charging. After switching to boost charging, due to a momentary interruption in energy flow, the output voltage will again fall below DC 650V, causing the system to switch back to reduced voltage charging. When charging reaches DC 650V again, it switches back to boost mode. This cycle repeats, causing significant current fluctuations when charging near DC 650V. Summary of the Invention

[0012] To avoid the shortcomings of traditional bidirectional DC-DC converters, achieve a wider range of charging and discharging voltages, reduce costs, and improve power supply reliability.

[0013] A novel bidirectional DC-DC circuit main electrical topology is proposed, with a two-level bridge arm on the left (DC end) and a three-level bridge arm on the right (battery end). The DC end uses a two-level configuration, with the emitter (E) of switch Q1 and the collector (C) of switch Q2 connected to form the first bridge arm, and capacitor C1 bridging the two ends of the first bridge arm. The battery end uses a three-level configuration, with the collectors (C) and emitters (E) of switches Q21, Q22, Q23, and Q24 connected sequentially, and capacitors C2 and C3 forming the second bridge arm. The midpoint of the series capacitors is connected between the emitter (E) of Q22 and the collector (C) of Q23 via a wire. One end of inductor L1 is connected between the emitter (E) of Q1 and the collector (C) of Q2, and the other end is directly connected between the emitter (E) of Q21 and the collector (C) of Q22 via a current sensor Am1. The DC ground (DC-) is connected between the emitter (E) of Q23 and the collector (C) of Q24. Figure 2 shown.

[0014] In the new bidirectional DC-DC circuit, the left-end voltage is a constant 650V, connected to the DC bus of the power supply (PCS); the right-end voltage ranges from 200V to 1500V, connected to the battery. The bidirectional DC-DC circuit is divided into charging and discharging directions based on the energy flow direction. Charging is when energy flows from the left end of the power supply to the right end, and discharging is when energy flows from the right end to the left end. The principle of the conventional control method for bidirectional DC-DC circuits is as follows.

[0015] I. Control principle for charging direction:

[0016] When operating in the charging direction, if the charging current is set to 100A, the target charging voltage is 1000V, and the battery voltage starts at 500V, then Q21 and Q24 are constantly on, Q1 is regulated, and Q2 is used for freewheeling to charge the battery below 650V (the voltage at the left end). As the battery voltage gradually increases, when it reaches 650V, even with all Q1s on, the battery can no longer be charged. At this point, a boost circuit needs to be activated, which involves Q1 being constantly on, Q22 and Q23 being regulated, and Q21 and Q24 being used for freewheeling to achieve a boost output.

[0017] From the control principle above, it can be seen that when the battery voltage is less than 650V, the PID controller adjusts the duty cycle of Q1. When the battery voltage equals the voltage at the left end of the power supply, the current is zero because the voltages at both ends are equal, and charging cannot continue. When the voltage is greater than 650V, the PID controller adjusts the duty cycles of Q22 and Q23, which is the boost circuit in the charging direction. Therefore, the conventional control method uses the voltage at the left end of the power supply as the dividing line, designing two independent PID control algorithms. One PID controller forms a closed loop when the battery voltage is less than the left end voltage, continuously adjusting the duty cycle of the Q1 switch. The larger the duty cycle, the larger the charging current, until the adjusted duty cycle value equals the period, meaning that even with Q1 fully turned on, charging cannot continue. When the battery voltage is detected to be equal to or close to the left end voltage, the control center shuts down the first PID controller and starts the second. The target voltage and controlled object of the second PID controller are different from the first. The target voltage is the final charging voltage, and the controlled object is the boost switch on the right bridge arm. At this time, Q1 is fully turned on, and the duty cycles of Q22 and Q23 are adjusted. The charging direction has two independent PIDs, as follows: Figure 2-1 and Figure 2-2 shown.

[0018] II. Control principle for discharge direction:

[0019] When operating in the discharge direction, if the discharge current is set to 100A, the discharge will stop when the battery terminal voltage reaches 300V, and the battery voltage will start at 1000V. For this, Q1 needs to be constantly on, Q21 and Q24 need to be regulated, and Q22 and Q23 need to be freewheeling to achieve discharge when the battery terminal voltage is greater than 650V. As the battery voltage gradually decreases, when it drops to 650V, even if Q21 and Q24 are both on, the battery cannot continue discharging. At this point, a boost circuit needs to be activated, where Q21 and Q24 are constantly on, Q2 needs to be regulated, and Q1 needs to be freewheeling to achieve a boost output.

[0020] As can be seen from the above, when the battery voltage is greater than 650V, the PID controller adjusts the duty cycles of Q21 and Q24. When the battery voltage equals the voltage at the left end of the power supply, the current is zero, and discharge cannot continue. When the voltage is less than 650V, the PID controller adjusts the duty cycle of Q2. Therefore, the conventional control method uses the voltage at the left end of the power supply as a dividing line, designing two independent discharge PID control logics. The first PID controller is used when the battery voltage is greater than the left end voltage, adjusting the duty cycles of Q21 and Q24. When the battery voltage is detected to be equal to or close to the left end voltage of the power supply, the first PID controller is turned off, and the second PID controller is activated, adjusting the duty cycle of Q2. The principle is similar to that of the charging direction, and will not be elaborated further here.

[0021] The disadvantages of the above control method used in the bidirectional DC-DC circuit of this invention are:

[0022] 1. The method of designing two independent PID controllers with the left-side voltage of the power supply as the dividing line has the problem of discontinuous charging (discharging) current near the dividing line, which affects the battery or system. For example, during charging, when the battery voltage is charged to the same level as the left-side voltage of the power supply, the charging current is 0A. After reaching 0A, the second set of boost PID regulation is activated, and the current gradually increases to the set current value.

[0023] 2. The control is complex, requiring continuous monitoring of the boundary voltage, two sets of PID controllers, and two sets of parameters. Especially when the target voltage is equal to the voltage at the other end, there are situations where the switching is repeated or the system fails to work properly.

[0024] 3. Poor dynamic response: When switching between two sets of PID controllers, the parameters are not seamlessly integrated. Switching from the current PID controller to a steady-state PID controller requires finding suitable parameters and takes some time to reach a steady state. During the switching process, the system exhibits poor dynamic response.

[0025] 4. Repeated switching occurs during the buck-boost transition, causing the charging and discharging voltage and current to fluctuate near the boundary line, posing a safety hazard to the power supply itself and the system.

[0026] To address the aforementioned shortcomings, based on the topology of the bidirectional DC-DC circuit of this invention, a continuous control method for the bidirectional DC-DC circuit is proposed. This method includes setting the difference between the given output voltage Uref* and the actual output voltage Uout of the bidirectional DC-DC circuit and using PI (proportional-integral) regulation to output a given output current command Iref*. Then, the difference between the command Iref* and the actual output current is further adjusted by PI to output a duty cycle output value of 0 ≤ DuR < 2Period, where DuR is the duty cycle output value and Period is the period. In the PID calculation... When a new value is generated, the control system checks whether DuR is less than 1*Period. If so, it controls the first bridge arm switch Q2 to turn off, all switches in the second bridge arm to turn off, and the new duty cycle DuRN = DuR. Then, the value of DuRN is updated in the duty cycle register of switch Q1. Otherwise, it controls the first bridge arm switch Q1 to turn on, switch Q2 to turn off, and the new duty cycle DuRN = DuR - Period. Then, the value of DuRN is updated in the duty cycle registers of switches Q22 and Q23, and switches Q21 and Q24 are turned off.

[0027] Advantages of this invention:

[0028] 1. By adjusting the PID output value to twice the range and then classifying the output value using a logical comparison method, a new type of bidirectional DC-DC (two-level on the left and three-level on the right) uninterrupted buck-boost control is truly realized, providing a guarantee for the stepless adjustment of a wide voltage power supply.

[0029] 2. The charging (discharging) direction can be handled by a single PID controller, saving MCU resources and improving system reliability.

[0030] 3. The low-voltage side adopts a two-level structure, and the high-voltage side adopts a three-level structure, which realizes a wider voltage charging and discharging requirement, effectively solves the problem of excessive stress on the switching transistor, minimizes the power supply cost, and indirectly improves the reliability of the power supply.

[0031] 4. During charging and voltage boosting and voltage bucking during discharging, the switching frequency is indirectly increased, thereby reducing the ripple current of the power supply system.

[0032] In this invention, the DC terminal voltage of the bidirectional DC-DC circuit is less than 800V, and the maximum battery terminal voltage range is greater than 1000V. Based on this, as another improvement, the switching transistor is configured as an IGBT to achieve reliable switching. Further, the upper limit of the battery terminal output voltage of the bidirectional DC-DC circuit is configured as DC 1500V, and the switching transistor at the battery terminal of the bidirectional DC-DC circuit is a 1700V IGBT. Further, the DC terminal of the bidirectional DC-DC circuit is configured as DC 650V, and the switching transistor at the DC terminal of the bidirectional DC-DC circuit is a 1200V IGBT.

[0033] In this invention, the uninterrupted control method is configured by associating two independent control objects with a single PID control algorithm. The PID control algorithm consists of a voltage outer loop and a current inner loop, with logical judgments set at the PID output. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A common bidirectional DC-DC electrical topology diagram is shown;

[0035] Figure 2 The bidirectional DC-DC electrical topology diagram of the present invention is shown;

[0036] Figure 2-1 The control objects and control block diagram below the boundary line are shown;

[0037] Figure 2-2 The control objects and control block diagrams above the boundary line are shown;

[0038] Figure 3 A charging step-down electrical simulation diagram is shown;

[0039] Figure 4 The charging buck output voltage waveform is shown;

[0040] Figure 5 The voltage waveform on the charging buck switch is shown;

[0041] Figure 6 A simulation diagram of the charging boost electrical circuit is shown;

[0042] Figure 7 The charging boost output voltage waveform is shown;

[0043] Figure 8 The voltage waveform on the charging boost switch is shown;

[0044] Figure 9 The charging boost trigger pulse monitoring is shown;

[0045] Figure 10 The waveform of the switching frequency is shown;

[0046] Figure 11 A block diagram of the energy storage charging and discharging power supply is shown;

[0047] Figure 12 A logic block diagram of uninterrupted control is shown;

[0048] Figure 13 An electrical simulation model of the charging direction is shown;

[0049] Figure 14 The output voltage waveforms in the charging direction (0-1000V) are shown;

[0050] Figure 15 The switching transistor trigger pulse is shown when the charging direction is indicated. DETAILED DESCRIPTION

[0051] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0052] The main electrical topology of the bidirectional DC-DC circuit of this invention is as follows: Figure 2 As shown, the system consists of a pair of two-level bridge arms, a pair of three-level bridge arms, inductors, and capacitors. The DC terminal uses a two-level configuration, with the emitter (E) of switch Q1 and the collector (C) of switch Q2 connected to form the first bridge arm, and capacitor C1 bridging the two ends of the first bridge arm. The battery terminal uses a three-level configuration, with the collectors (C) and emitters (E) of switches Q21, Q22, Q23, and Q24 connected sequentially, and capacitors C2 and C3 forming the second bridge arm. The midpoint of the series capacitors is connected between the emitter (E) of Q22 and the collector (C) of Q23 via a wire. One end of inductor L1 is connected between the emitter (E) of Q1 and the collector (C) of Q2, and the other end is directly connected between the emitter (E) of Q21 and the collector (C) of Q22 via current sensor Am1. The DC ground (DC-) is connected between the emitter (E) of Q23 and the collector (C) of Q24. The closed-loop control current is derived from the converted value of the current sampled at the inductor position between the two bridge arms; no additional current sensors are needed at either end.

[0053] The current flowing from left to right in inductor L1 is considered the positive direction (charging direction), and from right to left is the reverse direction (discharging direction). When operating in the positive direction, the right end of the bidirectional DC / DC unit can output a DC voltage that can be either high or low: Q21 and Q24 are constantly on, Q1 is adjustable, and Q2 freewheels, achieving a step-down output; Q1 is constantly on, Q22 and Q23 are adjustable, and Q21 and Q24 freewheel, achieving a step-up output. When operating in the reverse direction, the left end of the bidirectional DC / DC unit can output a DC voltage that can be either high or low: Q1 is constantly on, Q21 and Q24 are adjustable, and Q22 and Q23 freewheel, achieving a step-down output; Q21 and Q24 are constantly on, Q2 is adjustable, and Q1 freewheel, achieving a step-up output. The following simulations illustrate the charging step-down and charging step-up methods.

[0054] Figure 3 The electrical simulation of the charging buck circuit topology of this invention is shown (output 500V). Figure 4 The charging buck output voltage waveform of the circuit topology of the present invention is shown. Figure 5 The voltage waveform across the charging buck switch is shown. From... Figure 5 As you can see from the output, the CE voltage of all the switching transistors is not high.

[0055] Figure 6 The electrical simulation of the circuit topology of this invention during charging boost (output 1500V) is shown. Figure 7 The charging boost output voltage waveform is shown. Figure 8 The voltage waveform across the charging boost switch is shown. From... Figure 8 As can be seen from the output, the collector-emitter voltage (CE) of all the switching transistors is not high. The CE voltage of the battery-side switching transistor is half of the output voltage (DC 1500V), i.e., 750V. It is evident that this topology significantly reduces the switching stress on the transistors, improving system reliability. With the voltage reduction, a 1200V IGBT can be used on the DC side, and a 1700V IGBT can be used on the battery side. Both of these IGBTs are standard models, effectively reducing power supply costs.

[0056] To address the issue of repeated switching between buck and boost modes, regardless of whether it's in the forward (charging) or reverse (discharging) direction, when operating in either buck or boost mode, if the invention detects that the battery terminal voltage and DC terminal voltage are close (e.g., the difference is within a certain range), it controls the DC terminal (PCS DC terminal) voltage to decrease or increase by a certain value, and then switches to the other mode. For example, during buck charging, when the battery voltage rises to near DC 650V, the control system controls the PCS DC terminal voltage to decrease slightly before switching to boost charging mode. This prevents repeated switching.

[0057] Figure 9An electrical simulation of charging boost trigger pulse monitoring is shown. Figure 10 The waveform of the switching frequency is shown, from... Figure 10 It can be seen that the frequency of the voltage at the output end of the three-level bridge arm (on inductor L1) is twice the switching frequency of the switching transistor. Therefore, by adopting this topology, the switching frequency is indirectly increased and the output ripple of the power supply is reduced.

[0058] like Figure 11 The energy storage charging power supply of this invention consists of a PCS energy storage converter and a bidirectional DC-DC circuit. The DC power of the PCS energy storage converter and the DC terminal of the bidirectional DC-DC circuit are connected together. This allows for both converting grid energy into DC to charge the battery and inverting battery energy back to the grid, enabling bidirectional energy flow. Furthermore, the bidirectional DC-DC circuit offers a wide voltage output, capable of charging and discharging batteries of various voltage levels. The advantages of the main electrical topology of the energy storage charging power supply of this invention are:

[0059] 1. The two-level bridge arm and the three-level bridge arm are used together to achieve a wider range of charging and discharging voltages, effectively solving the problem of excessive stress on the switching transistor;

[0060] 2. Low cost: all switching devices can use conventional switching devices, effectively reducing costs;

[0061] 3. Improved reliability; by reducing switching stress, the reliability of the power supply is indirectly improved.

[0062] 4. Before the buck-boost switching, the given voltage of the PCS was appropriately changed to prevent repeated false switching of the system.

[0063] 5. During charging and voltage boosting and voltage bucking during discharging, the switching frequency is indirectly increased, thereby reducing the ripple current of the power supply system.

[0064] Due to the differences between the given object and the controlled object, conventional control methods involve setting two different PID controllers in the charging (discharging) direction, each forming an independent closed loop. This results in discontinuous buck-boost switching time, preventing continuous output and even causing fluctuations at the switching point. To solve this problem, based on the aforementioned bidirectional DC-DC circuit topology, control logic is set as follows: Figure 12 As shown, it includes:

[0065] The difference between the given output voltage Uref* and the actual output voltage Uout of the bidirectional DC-DC circuit is set and PI (proportional-integral) is used to adjust the output current command Iref*. Then, the difference between the command Iref* and the actual output current is set and PI is used to adjust the output duty cycle value of 0≤DuR<2Period, where DuR is the duty cycle output value and Period is the period.

[0066] When the PID calculates a new value, the control system checks whether DuR is less than 1*Period. If so, it controls the first bridge arm switch Q2 to turn off, all switches in the second bridge arm to turn off, and the new duty cycle DuRN = DuR. Then, the value of DuRN is updated in the duty cycle register of switch Q1. Otherwise, it controls the first bridge arm switch Q1 to turn on, switch Q2 to turn off, and the new duty cycle DuRN = DuR - Period. Then, the value of DuRN is updated in the duty cycle registers of switches Q22 and Q23, and switches Q21 and Q24 are turned off.

[0067] Figure 12 In this configuration, a PID controller is set up for the entire charging (discharging) direction to prepare for continuous and uninterrupted control. At this time, the current loop PID output is no longer a duty cycle of 0 to Period, but rather an output value of 0 to 2 * Period. For example, if the PID adjusts to a value of 1.5 * Period, it means that a duty cycle of 1.5 times the period is needed in the charging (discharging) direction to meet the current adjustment requirements. However, duty cycles greater than or equal to the period cannot be expressed within the same controlled object. Therefore, the 1.5 of the 1.5 times period is split into a period of 1 + 0.5. The first 1 represents the activation of Q1 in the charging (discharging) direction, and the second 0.5 represents the PWM adjustment of the switching transistors Q22 and Q23 in the charging (discharging) direction. Therefore, when the PID calculates a new value, the MCU needs to determine the magnitude of the output value. When it detects that the output value DuR ≥ Period, it indicates that the charging (discharging) adjustment has switched to the right bridge arm. When the right bridge arm Q1 is turned on and Q2 is turned off, the new duty cycle DuRN = DuR - Period is set. Then, the value of DuRN is updated in the duty cycle registers of Q22 and Q23, and Q23 and Q24 are turned off.

[0068] The boost principle of charging (discharging) is that the inductor first stores energy and then releases it. In this process, the more energy the inductor stores, the more energy is released. Therefore, to obtain a higher voltage, the larger the DuR value adjusted by the PID controller, the longer the on-time within the cycle, resulting in more energy stored and released by the inductor. Here, Uout is the output voltage, Uin is the input voltage, L is the inductance, and di / dt is the change in inductor current. Because energy needs to be stored and released, and due to integral considerations, the PID controller's output value cannot reach 2*Period. In this embodiment, a minimum of 0.1*Period is allowed for energy release.

[0069] Taking the charging direction as an example, an electrical simulation model is established. The voltage at the left end is DC650V, and the target output voltage at the right end is set to DC1000V, with a current limit of 100A, a 5Ω resistor, and a frequency of 5kHz. Figure 13 shown.

[0070] from Figure 13 The trigger waveforms of Q1, Q22, and Q23 show that the switching is crisp and clean, without repeated switching, which ensures stable voltage and current output.

[0071] The control during discharge is similar to that during charging, using a PID controller to adjust the output value from 0 to 2*Period. This value is then seamlessly and automatically switched to the corresponding bridge arm via a comparator. Whether boosting or bucking, the same PID parameters are used, ensuring truly seamless switching.

[0072] Figure 14 The output voltage waveform in the charging direction (0-1000V) is shown. It can be seen from the voltage waveform in the charging direction that when the output voltage equals the input voltage (650V), the voltage is continuous without any pauses.

[0073] Figure 15 The switching transistor trigger pulse is shown when the charging direction is indicated. From Figure 15 The trigger waveforms of Q1, Q22, and Q23 show that the switching is crisp and clean, without repeated switching, which ensures stable voltage and current output.

[0074] The control during discharge is similar to that during charging, using a PID controller to adjust the output value from 0 to 2*Period. This value is then seamlessly and automatically switched to the corresponding bridge arm via a comparator. Whether boosting or bucking, the same PID parameters are used, ensuring truly seamless switching.

[0075] The advantages of the bidirectional DC-DC circuit uninterrupted control method of the present invention are:

[0076] 1. By adjusting the PID output value to twice the range and then classifying the output value using a logical comparison method, a new type of bidirectional DC-DC (two-level on the left and three-level on the right) uninterrupted buck-boost control is truly realized, providing a guarantee for the stepless adjustment of a wide voltage power supply.

[0077] 2. The charging (discharging) direction can be handled by a single PID controller, saving MCU resources and improving system reliability.

[0078] 3. Seamless switching occurs when the battery voltage equals the DC terminal voltage, resulting in no abrupt changes in the voltage waveform.

[0079] In this invention, the control system uses a DSP28335 with an operating frequency of 150MHz; the control system architecture adopts a 1*DSP+1*FPGA approach, with the two chips communicating with each other via parallel ports and directly sharing registers; the PID algorithm is implemented in the DSP, and the calculated value is sent to the FPGA for judgment and comparison before being output to the corresponding IO port, and then controlled by the drive circuit to control the IGBT, with a switching frequency of 5kHz, or 1.5kHz to 6kHz.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A bidirectional DCDC circuit uninterrupted control method, characterized in that: The bidirectional DCDC circuit comprises a switch tube Q1, a switch tube Q2, a switch tube Q21, a switch tube Q22, a switch tube Q23, a switch tube Q24, a capacitor C1, a capacitor C2, a capacitor C3 and an inductor L1; the DC end of the bidirectional DCDC circuit adopts a two-level mode, and the E pole of the switch tube Q1 and the C pole of the switch tube Q2 are connected to form a first bridge arm, and the capacitor C1 is connected across the two ends of the first bridge arm; the battery end of the bidirectional DCDC circuit adopts a three-level mode, and ... Capacitors C2 and C3 form a second bridge arm, and the C poles and E poles of the switch tubes Q21, Q22, Q23, and Q24 are connected in sequence, and the midpoint of the series connection of capacitors C2 and C3 is connected between the E pole of the switch tube Q22 and the C pole of the switch tube Q23; one end of the inductor L1 is connected between the E pole of the switch tube Q1 and the C pole of the switch tube Q2, and the other end is connected between the E pole of the switch tube Q21 and the C pole of the switch tube Q22, and the ground of the DC end of the bidirectional DCDC circuit is connected between the E pole of the switch tube Q23 and the C pole of the switch tube Q24; The uninterrupted control method comprises setting the difference between the given output voltage Uref* and the actual output voltage Uout of the bidirectional DCDC circuit to perform PI regulation to output a given output current command Iref*, and then setting the difference between the command Iref* and the actual output current to output a duty cycle output value of 0≤DuR<2Period through PI regulation, wherein DuR is the duty cycle output value and Period is the period; When PID calculates a new value, it checks whether DuR is less than 1*Period. If so, the switch tube Q2 of the first bridge arm is controlled to be turned off, and all the switch tubes of the second bridge arm are turned off. The new duty cycle DuRN=DuR, and then the value of DuRN is used to update the duty cycle register of the switch tube Q1; otherwise, the switch tube Q1 of the first bridge arm is controlled to be turned on, and the switch tube Q2 is turned off. The new duty cycle DuRN=DuR-Period, and then the value of DuRN is used to update the duty cycle registers of the switch tubes Q22 and Q23, and the switch tubes Q21 and Q24 are turned off.

2. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: The duty cycle output value DuRN is configured to be between 0 and 1.9*Period.

3. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: When the bidirectional DCDC circuit works in the buck charging mode, the switch tubes Q21 and Q24 are controlled to be constantly on, the switch tube Q1 is regulated, and the switch tube Q2 is controlled to be freewheeling; When the bidirectional DCDC circuit works in the boost charging mode, the switch tube Q1 is controlled to be constantly on, the switch tubes Q22 and Q23 are controlled to be regulated, and the switch tubes Q21 and Q24 are controlled to be freewheeling; When the bidirectional DCDC circuit operates in the step-down discharge mode, the switch tube Q1 is controlled to be constantly on, the switch tubes Q21 and Q24 are regulated, and the switch tubes Q22 and Q23 are freewheeling; When the bidirectional DCDC circuit operates in the boost discharge mode, the switch tubes Q21 and Q24 are controlled to be constantly on, the switch tube Q2 is regulated, and the switch tube Q1 is freewheeling.

4. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: When operating in either the buck mode or the boost mode, if it is detected that the difference between the battery terminal voltage and the DC terminal voltage of the bidirectional DCDC circuit is within a certain range, the DC terminal given voltage is controlled to be reduced or boosted by a certain value and then switched to the other mode.

5. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: One end of the inductor L1 is connected between the E pole of the switch tube Q1 and the C pole of the switch tube Q2 , and the other end is connected in series with the current sensor Am1 and then connected between the E pole of the switch tube Q21 and the C pole of the switch tube Q22 .

6. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: The control system architecture adopts 1*DSP+1*FPGA. The two chips communicate with each other via parallel port and share registers. The PID algorithm is implemented in DSP, and the calculated value is sent to FPGA for judgment and comparison to control the corresponding switch tube.

7. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: The DC terminal voltage of the bidirectional DCDC circuit is less than 800V, and the maximum voltage range of the battery terminal is greater than 1000V.

8. The uninterrupted control method of a bidirectional DCDC circuit according to claim 7, characterized in that: The switch tube is configured as an IGBT.

9. The uninterrupted control method of a bidirectional DCDC circuit according to claim 8, characterized in that: The upper limit of the output voltage of the battery end of the bidirectional DCDC circuit is configured to DC1500V, and the switch tube of the battery end of the bidirectional DCDC circuit adopts 1700V IGBT.

10. The uninterrupted control method of a bidirectional DCDC circuit according to claim 8, characterized in that: The DC end of the bidirectional DCDC circuit is configured as DC650V, and the switching tube at the DC end of the bidirectional DCDC circuit adopts a 1200V IGBT.

11. The uninterrupted control method of a bidirectional DCDC circuit according to claim 1, characterized in that: The uninterrupted control method is configured such that two sets of independent control objects are associated with one set of PID control algorithms.

12. The uninterrupted control method of a bidirectional DCDC circuit according to claim 11, characterized in that: The PID control algorithm is composed of a voltage outer loop and a current inner loop, and a logical judgment is set at the PID output end.

Images