Double-rectification single-inversion redundant uninterruptible power supply system and control protection method thereof

Abstract

The invention relates to the technical field of power electronics, and particularly provides a double-rectification single-inversion redundant uninterruptible power supply system and a control protection method thereof, and the system comprises a power distribution system, a variable current power module and a control protection system. The power distribution system has double AC inputs, a common DC bus and AC outputs. The converter module comprises three converters, two of which serve as rectifiers and are connected in parallel with the direct current bus, and the other one serves as inverter output. The control method comprises the following steps: in a double-path load balancing mode, realizing power equalization through composite voltage control; in a single-path operation mode, fault seamless switching is realized based on load feedforward and adaptive slope control; in a power distribution mode, triggering according to capacity margin and cooperatively adjusting a droop coefficient based on an optimization target to realize power redistribution; and complete real-time fault protection logic is realized. According to the invention, uninterruptible power supply with high reliability and high electric energy quality is realized, and active regulation and control of power are realized.

Description

Technical Field

[0001] This invention relates to the field of power electronics technology, specifically to a dual-rectifier single-inverter redundant uninterruptible power supply system and its control and protection method. Background Technology

[0002] In practical applications of mainstream traditional power supply backup equipment for critical events, numerous unavoidable pain points have gradually emerged, severely restricting the safety redundancy and emergency response capabilities of power backup for these events. Traditional Automatic Transfer Switches (ATS), as the core device for power switching, rely on the physical action of mechanical contacts to achieve power switching. However, in scenarios involving long-term continuous operation of critical events, mechanical contacts are susceptible to factors such as arc burning, wear and aging, and changes in environmental humidity, frequently resulting in contact adhesion, poor contact, or switching delays, directly leading to power switching failures and the risk of power outages. While UPS (Uninterruptible Power Supply) with a single rectifier module structure can provide temporary power during mains power outages, its "single-module dependence" design flaw creates a significant single point of failure vulnerability. Once the rectifier module fails, the entire UPS will completely lose its power supply capability, unable to provide continuous power support to the load, and failing to meet the core requirement of power redundancy for critical events. Furthermore, while static transfer switches (SSTS) offer fast switching speeds, their complex structure, difficult maintenance of core components, and high spare parts costs place a heavy operational burden on service providers due to the high maintenance costs over the long term. Moreover, all of these traditional devices lack power flow control capabilities. During critical events, with the concentrated deployment of various high-precision equipment and high-power loads, upstream power supplies often face issues such as load imbalance and sudden power surges. Traditional equipment can only passively respond to power demands and cannot actively adjust power distribution or smooth load fluctuations, easily leading to severe overload operation of the upstream power supply. This further exacerbates the instability of the power supply system and poses significant safety hazards to power supply security. Summary of the Invention

[0003] To address the aforementioned issues, this invention provides a dual-rectifier single-inverter redundant uninterruptible power supply system and its control and protection method, achieving high reliability, high power quality, and active power regulation.

[0004] In a first aspect, the technical solution of the present invention provides a dual-rectifier single-inverter redundant uninterruptible power supply system, including a power distribution system, a converter power module and a control and protection system. The power distribution system includes a first AC input branch, a second AC input branch, a common DC bus, and an AC output branch; The power converter module includes a first bidirectional converter, a second bidirectional converter, and a third bidirectional converter. The AC sides of the first and second bidirectional converters are respectively connected to the first AC input branch and the second AC input branch, and their DC sides are both connected to a common DC bus to convert AC input power into DC power. The DC side of the third bidirectional converter is connected to the common DC bus, and its AC side is connected to the AC output branch to convert DC power into AC output power. The control and protection system is communicatively connected to the first bidirectional converter, the second bidirectional converter, and the third bidirectional converter, and is used to coordinate the working mode of the converter power module and execute fault protection logic.

[0005] Secondly, the technical solution of the present invention provides a control and protection method, applied to the system described in any of the above claims, including an operation mode control process and a fault protection process; The operation mode control process includes: In response to the normal operation of the dual input power supply, the first bidirectional converter and the second bidirectional converter are controlled to operate in constant voltage mode, and the output power difference between the two is maintained within a preset threshold through coordinated control, thus entering the dual load balancing mode. In response to a power supply failure in one of the input circuits, the pulse of the corresponding bidirectional converter is blocked, and the other normal bidirectional converter is controlled to increase its output power to handle the entire load, switching to single-circuit operation mode. In response to insufficient input power capacity on one of the circuits, the droop coefficients of the first and second bidirectional converters are adjusted to redistribute the load power of the two circuits and enter the power distribution mode. The fault protection process includes: Real-time monitoring of the system's DC voltage, DC current, AC voltage, AC current, frequency, and temperature parameters; When any monitored parameter exceeds the corresponding safety threshold, the corresponding protection action is executed. The protection action includes: immediately or after a delay blocking the drive pulse of the relevant power device, tripping the AC / DC side switch, recording fault information and triggering an alarm.

[0006] As can be seen from the above technical solutions, this application has the following advantages: By using redundant parallel connection of dual converters and a common DC bus design, the potential for single-point faults is eliminated; by utilizing all-solid-state electronic switching and control logic, seamless, zero-interruption switching during power supply failures is achieved, overcoming the problems of mechanical switching failures in traditional ATS and single-point faults in single-rectifier UPS; through coordinated control, automatic load balancing between the two power supplies is achieved to extend equipment lifespan, and power distribution is adjusted when the capacity of a single power supply is insufficient, realizing proactive power flow regulation and preventing overload of the upstream power supply, solving the problems of passive function and lack of control in traditional equipment; furthermore, based on the ANPC three-level converter topology and stable common DC bus structure, and through closed-loop control, harmonics are effectively suppressed and grid disturbances are isolated, providing high-quality power supplies with stable voltage, frequency, and waveform for critical loads. Attached Figure Description

[0007] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a schematic block diagram of a dual-rectifier single-inverter redundant uninterruptible power supply system provided in an embodiment of the present invention.

[0009] Figure 2 This is a schematic diagram of the operation mode control process provided in an embodiment of the present invention.

[0010] Figure 3 This is a flowchart illustrating the fault protection process. Detailed Implementation

[0011] To make the purpose, features, and advantages of this application more apparent and understandable, specific embodiments and accompanying drawings will be used to clearly and completely describe the technical solution protected by this application. Obviously, the embodiments described below are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this application and in the specification of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0013] Figure 1A schematic block diagram of a dual-rectifier single-inverter redundant uninterruptible power supply system provided in an embodiment of the present invention is shown below. Figure 1 As shown, the power system includes a power distribution system, a converter power module, and a control and protection system.

[0014] The power distribution system is responsible for the input, distribution, convergence and output of electrical energy, including the first AC input branch, the second AC input branch, the common DC bus and the AC output branch.

[0015] First AC input branch: Used to connect to the first independent AC power source, such as AC mains A or generator A. This branch typically has an input circuit breaker (QF1), a filter inductor, voltage / current sampling sensors, and other components connected in series to achieve power switching, protection, and status monitoring.

[0016] The second AC input branch is used to connect to a second independent AC power source, such as AC mains circuit B or generator B. Its configuration is symmetrical to the first AC input branch, including an input circuit breaker (QF2) and corresponding filtering and sampling units. The two input power sources are electrically isolated from each other, ensuring the independence of the power supply from the source.

[0017] Common DC Bus: This is a single, shared DC bus. The DC power generated by the rectification of the first and second AC input branches converges here. The common DC bus provides a unified and stable DC voltage platform for the downstream inverters and potential DC loads. The bus is equipped with supporting capacitor banks (C1, C2), voltage sensors, and pre-charge / discharge circuitry.

[0018] AC output branch: Used to supply power to critical AC loads. This branch is equipped with an output circuit breaker (QF3), a filter circuit, and an output sampling unit to ensure the quality and safety of the output power.

[0019] The power converter module includes a first bidirectional converter, a second bidirectional converter, and a third bidirectional converter, all three employing a unified topology. The AC sides of the first and second bidirectional converters are connected to the first and second AC input branches, respectively, while their DC sides are connected to a common DC bus, used to convert AC input power into DC power. The DC side of the third bidirectional converter is connected to the common DC bus, and its AC side is connected to the AC output branch, used to convert DC power into AC output power.

[0020] Specifically, the first bidirectional converter has its AC side connected to the first AC input branch via an AC bus or cable; its DC side positive and negative output terminals are directly connected in parallel to the common DC bus. During normal system operation, its main function is to efficiently convert the first AC input power into DC power and inject it into the common DC bus.

[0021] The second bidirectional converter has its AC side connected to the second AC input branch; its DC side is also connected in parallel to the common DC bus. Its function is symmetrical to the first bidirectional converter, responsible for converting the second AC input power into DC power. The two converters together maintain the voltage stability of the common DC bus.

[0022] The third bidirectional converter connects its positive and negative input terminals on the DC side to the common DC bus, obtaining DC power from the bus; its AC side is connected to the AC output branch via an AC bus or cable. Its main function is to invert the DC power on the common DC bus into stable AC power to supply critical loads downstream.

[0023] The first, second, and third bidirectional converters all adopt an active neutral-point clamped three-level topology and use power switching devices such as IGBTs or SiC.

[0024] The control and protection system is communicatively connected to the first bidirectional converter, the second bidirectional converter, and the third bidirectional converter, and is used to coordinate the working mode of the converter power module and execute fault protection logic.

[0025] Specifically, the control and protection system communicates with the control units of the first, second, and third bidirectional converters via a high-speed communication bus. Simultaneously, the control and protection system receives signals from various sampling sensors in the power distribution system via hardwiring or communication methods, and controls the opening and closing of each circuit breaker and contactor.

[0026] In this embodiment, the power distribution system further includes a first bypass switch and a second bypass switch; the first bypass switch is connected in parallel between the first AC input branch and the AC output branch; the second bypass switch is connected in parallel between the second AC input branch and the AC output branch; when the converter power module is out of operation, the control and protection system controls the closing of the corresponding first bypass switch or the second bypass switch to directly feed the power of the corresponding AC input branch to the AC output branch.

[0027] Specifically, one end of the first bypass switch is connected after the input circuit breaker (QF1) of the first AC input branch and before the AC input terminal of the first bidirectional converter; the other end is directly connected before the output circuit breaker (QF3) of the AC output branch. When KBP1 is closed, the first AC input power supply will directly power the downstream load without passing through any converter.

[0028] The second bypass switch is connected in a completely symmetrical manner to the first bypass switch, and is connected in parallel between the second AC input branch and the AC output branch to realize the direct supply function of the second input power.

[0029] The bypass switch is preferably a static switch, which has an extremely fast switching speed, typically in the millisecond or even microsecond range. It can achieve "on-then-off" or "synchronous switching" of the power supply path, ensuring absolute continuity of power supply to the load and completely avoiding the risk of momentary interruption caused by mechanical switch operation. The closing and opening of the bypass switch is uniformly managed by the control and protection system. The control and protection system sends a trigger signal to the gate (or control terminal) of the bypass switch through the drive circuit.

[0030] For example, when planned maintenance, upgrades, or replacements are required for the power converter module or the control system itself, maintenance personnel can issue a "bypass activation" command through the human-machine interface of the control and protection system or the remote monitoring system. The control and protection system will execute the following logical sequence: First, it issues a command to block all drive pulses of the first, second, and third bidirectional converters, causing the power converter module to stop working and enter an electrically isolated state; subsequently, the control and protection system controls the closing of the bypass switch to be activated. Due to the use of a static switch, this process can achieve a smooth transition at the current zero-crossing point, and for the load, there is no perceptible interruption or disturbance in the amplitude, phase, and frequency of the supply voltage. After the bypass switch is closed, the corresponding AC input branch power supply (such as the first mains power) will directly supply power to the load. At this time, the power converter module can be safely de-energized, isolated, and maintained.

[0031] This embodiment provides a control and protection method applied to the dual-rectifier single-inverter redundant uninterruptible power supply system described in the above embodiment, including an operation mode control process and a fault protection process.

[0032] Figure 2 This is a schematic diagram of the operation mode control process provided in an embodiment of the present invention, such as... Figure 2 As shown, the procedure includes the following steps S101 to S103.

[0033] S101, in response to the normal operation of the dual input power supply, controls the first bidirectional converter and the second bidirectional converter to operate in constant voltage mode, and maintains the difference in output power between the two within a preset threshold through coordinated control, thus entering the dual-path load balancing mode.

[0034] When the system detects that both input power supplies are normal, it automatically enters this basic high-efficiency operating mode. In this mode, both the first and second bidirectional converters operate in constant voltage mode and are adjusted in real time through a collaborative control algorithm to keep the output power of the two converters balanced, i.e., the difference does not exceed a preset threshold. This maximizes the utilization of the dual power supply capacity, distributes equipment losses, and provides the highest redundancy for the system. This is the normal operating mode for scenarios such as power supply protection during major events.

[0035] S102, in response to a fault in one of the input power supplies, blocks the pulse of the corresponding bidirectional converter and controls the other normal bidirectional converter to increase its output power to handle the entire load, switching to single-channel operation mode.

[0036] When a fault is detected in one of the input power supplies (such as a power outage or a severe voltage drop), the system immediately triggers the fault ride-through mechanism. The control and protection system blocks the pulse of the fault-side converter within milliseconds and simultaneously controls the normal-side converter to rapidly increase its output power, seamlessly taking over the entire load. This switching process is based on a feedforward and adaptive control strategy, ensuring almost no disturbance to the DC bus voltage and the final AC output, achieving zero-interruption power supply switching.

[0037] S103, in response to insufficient input power capacity of one of the input power sources, adjusts the droop coefficients of the first and second bidirectional converters, redistributes the load power of the two sources, and enters power distribution mode.

[0038] When the system detects that the load rate of a certain input power source is too high and its capacity is close to insufficient, it automatically enters this mode to prevent overload and ensure the normal operation of other loads on that power source. By dynamically and collaboratively adjusting the droop coefficients of the two converters, the load power borne by the two power sources is redistributed, automatically transferring part of the load from the heavily loaded power source to the lightly loaded power source. This mode represents a leap from "passive power supply" to "active intelligent control," effectively optimizing the operating status of the upstream power grid and improving the stability and economy of the overall power supply system.

[0039] As a refinement and extension of the specific implementation of the above embodiments, in order to fully explain the specific implementation process of this embodiment, the following will provide possible embodiments to describe the specific implementation of the above steps in a non-limiting manner.

[0040] In some optional implementations, in the dual-path load balancing mode, in order to achieve power balancing between the two converters, a dual-closed-loop collaborative control strategy based on virtual impedance and dynamic voltage compensation is adopted. This strategy dynamically fine-tunes the output voltage command of each converter through virtual droop characteristics and real-time power deviation feedback, so that the system can automatically and smoothly achieve power balancing while ensuring the stability of the DC bus voltage.

[0041] Specifically, assume there are N=2 bidirectional converters operating in parallel in the system. For the i-th converter (i=1,2), its outer control loop uses DC voltage control, and its voltage reference value is... Based on its output power Dynamic adjustment, represented as:

[0042] in, This is the global setting value for the DC bus voltage. The virtual droop coefficient preset for the i-th converter. Let be the real-time output power of the i-th converter. Let i be the dynamic voltage compensation amount of the i-th converter, where i = 1 and 2, representing the first bidirectional converter and the second bidirectional converter, respectively.

[0043] The DC bus voltage represents the global target value that the entire system expects to maintain. It is the common anchor point or ultimate goal for the voltage control of all converters and determines the operating voltage level of the system's DC side. This value is a fixed setpoint, providing a stable and unified voltage reference for the entire control system.

[0044] This is a static droop term, based on a droop controller, which makes the converter's voltage reference value correlated with its own output active power. They are inversely proportional. That is, when the power output of the converter... When the voltage increases, its reference value It will decrease by one amount accordingly. On a single DC bus supported by multiple converters connected in parallel, if a converter tends to output more current due to internal or external reasons, the decrease in its voltage reference value will suppress the growth trend of its output current through closed-loop control. Conversely, a converter with lower output current has a relatively higher voltage reference value, which will encourage it to increase its output. This inherent negative feedback characteristic provides the parallel system with a power distribution tendency and inherent small-signal stability, avoiding power circulation and oscillation.

[0045] This is a dynamic compensation term, a dynamically generated voltage compensation amount calculated based on the real-time power deviation between the two converters. The collaborative control device continuously compares the output power of the two converters. and If a power deviation is detected Exceeding the preset allowable dead zone threshold If the result is zero, then the system is considered not to have reached an ideal equilibrium state. In this case, non-zero values ​​are generated. For converters with higher actual power, their It is calculated as a negative value or a reduced positive value; for converters with lower actual power, its It is calculated as a positive value or an increasing positive value. Substituting into the overall formula is equivalent to fine-tuning the voltage reference values ​​of the two converters based on the original droop characteristics. The higher-power converter experiences an additional reduction in its total voltage reference value, thus actively yielding some load; the lower-power converter experiences an additional increase in its total voltage reference value, thus actively absorbing more load. This dynamic fine-tuning process continues until the power deviation between the two converters is driven to the dead zone. Within that time, the system reaches a high-precision equilibrium state.

[0046] visible, This forms the basic operating framework of the system, providing stability and initial power allocation capabilities. As a dynamic compensation measure, it detects and eliminates power distribution errors that cannot be avoided by a static frame in real time. Each converter, with its own... Closed-loop (voltage loop, current loop) control is performed on the target device to adjust its PWM output. Its output power... The changes, in turn, affect The calculation. Through this closed loop, It is automatically adjusted to a suitable value, so that the final and By getting infinitely close, the preset power balance target can be achieved.

[0047] Based on the above strategy, in step S101, the output power difference between the two is maintained within a preset threshold through coordinated control, specifically including the following steps S101.1 to S101.2.

[0048] S101.1, Obtain the real-time output power of the first bidirectional converter and the second bidirectional converter respectively. and Calculate the average power and the deviation between their respective real-time output power and average power. .

[0049] S101.2, for each bidirectional converter, the following control is executed: a) Determine the power deviation value Is the absolute value greater than the preset power deviation dead zone threshold? If so, the power deviation compensation regulator will be activated to generate dynamic voltage compensation. If not, then keep the current dynamic voltage compensation amount unchanged. b) Real-time output power based on the converter Preset virtual droop coefficient DC bus voltage global setting value and dynamic voltage compensation Generate an independent DC voltage reference value for the converter. , is represented as:

[0050] c) The converter uses its independent DC voltage reference value as the control target and controls its output through closed-loop regulation via its internal voltage and current loops.

[0051] Through the cyclic execution of steps (a) to (c) above, the dynamic voltage compensation is automatically adjusted so that the power deviation value is converged and limited within the range of the power deviation dead zone threshold, thereby achieving dual-path power balance.

[0052] Power deviation value Input a proportional-integral regulator with dead time to generate a dynamic voltage compensation term. The calculation process is as follows: Update the integral term, represented as follows:

[0053] in, The integral coefficient is... To control the cycle duration, This is the kth cycle.

[0054] For the integral term Limit the amplitude to meet the requirements. , This is the preset maximum compensation amplitude.

[0055] Calculate the dynamic voltage compensation term , This is the proportionality coefficient.

[0056] The preset virtual droop coefficient should be minimized while ensuring the stability of the system's large signal, in order to reduce the inherent impact on power distribution accuracy and allow the subsequent dynamic compensation mechanism to work more efficiently. Specifically, the preset virtual droop coefficient ensures that DC bus voltage fluctuations remain within allowable limits during load step changes or single-unit failures, preventing system oscillations and providing a basic power distribution tendency between the two units, avoiding severe power conflicts during startup or transient processes. Simultaneously, the droop effect should not be excessive to avoid overly "masking" power deviations and affecting the sensitivity and adjustment speed of the dynamic compensation loop. Based on design requirements, the preset virtual droop coefficient is determined based on the converter's rated power and the maximum allowable DC voltage deviation range, expressed as:

[0057] in, This refers to the maximum static adjustment of the voltage reference value of a single converter allowed by the system from no-load to rated load. This refers to the rated output power of a single converter.

[0058] When a converter carries the full rated load At that time, its voltage reference value will decrease. This value is within the range where the controller can respond smoothly, thus ensuring that the system does not become unstable or oscillate in the event of a single-machine failure or a sharp increase in load. The formula determines... The value is the minimum value under the condition of satisfying the stability requirement. A larger deviation will result in a larger inherent voltage difference for the same power, increasing the burden on the dynamic compensation loop and reducing the balancing speed and accuracy; if If the value is smaller, the aforementioned stability boundary cannot be satisfied. Therefore, this value minimizes the adverse effects of droop on power averaging accuracy while ensuring stability.

[0059] In some alternative implementations, in single-path operation mode, in order to control another normal bidirectional converter to increase its output power to take on the entire load, load current feedforward and adaptive voltage-current slope control are used to enable the normal side converter to actively and on demand increase its power the instant a fault occurs, reducing the bus voltage fluctuation during the switching process to almost zero, achieving zero-perception switching.

[0060] The system is configured to operate in dual-path load balancing mode, with converter one and converter two each handling approximately 50% of the load. The total load current referred to the DC side is... The DC bus voltage is set to... .

[0061] During normal operation, the collaborative controller calculates and dynamically updates an expected single-machine full load factor. ,in This represents the rated output current of a single converter. The expected single-unit full-load factor. This indicates the load factor if the current load is entirely borne by a single converter.

[0062] When an input power failure is confirmed in converter one, the drive pulse on the faulty side (converter one) is immediately blocked. Simultaneously, a feedforward-based instantaneous power boost command is sent to the normal side (converter two). This command is based on the instantaneous total load current before the fault. and expected single-unit full load factor The calculations show that the instantaneous total load current before the fault is... Used for feedforward processing The goal is to make converter two instantly bear the entire load current, and the corresponding power feedforward command is: Expected single-unit full load factor Used to calculate voltage-current compensation slope The voltage reference value received by the inner current loop It includes slope-based The dynamic adjustment item.

[0063] Upon receiving the fault signal and feedforward command, converter two immediately switches its voltage outer loop control mode. It switches from the original constant voltage mode to an adaptive slope boost mode, with the new control law as follows:

[0064] in, Here is the voltage reference value of the converter at time t. This is the output current of converter two at the moment of the fault.

[0065] This is the voltage-current compensation slope, which is adjusted based on the expected load rate that the normal-side converter needs to handle before the fault occurs. The higher the expected load rate, the smaller the value of this slope. Defined as... , The preset constant adjustment coefficient is a small normal number, such as 0.01 V / A. When When the load is large (i.e., heavy), Taking a smaller value means that the voltage reference value increases more gradually with the current, avoiding overshoot and oscillation caused by pursuing rapid compensation; when When the load is relatively small (lighter), Taking a larger value allows the voltage reference value to rise more quickly to establish a new balance, enabling adaptive adjustment of the controller stiffness according to the load.

[0066] Receive reference voltage from the outer voltage loop and current command from feedforward The combined effect of the current setpoint for:

[0067] in It is the voltage loop proportional coefficient, and the min function ensures that the rated current is not exceeded. The value of the DC bus voltage obtained by actual sampling at time t is t.

[0068] Through the high-speed response of the inner current loop (response time is typically on the order of hundreds of microseconds), the output current of converter two... It begins to rise rapidly and smoothly within 1-2 switching cycles, tracking... Due to feedforward instructions Provides an accurate power demand reference and adaptive slope control. It provides an optimized dynamic response path, enabling DC bus voltage... The decline is greatly suppressed; for example, the fluctuation is less than ±0.5%. When Stable at Nearby, and bus voltage Stable at After remaining within ±0.5% of the setpoint for several power frequency cycles, the control mode of converter two automatically switches from adaptive slope boost mode back to constant voltage mode. However, at this point, the voltage setpoint has stabilized at slightly higher than the setpoint. A new equilibrium point is reached to offset the line voltage drop, and the system enters a stable single-path operation mode.

[0069] Based on the above strategy, step S102 involves controlling another normal bidirectional converter to increase its output power to handle the entire load. Specifically, this includes: responding to a fault confirmation signal, synchronously executing a drive pulse to block the fault-side converter, and initiating a seamless power transfer control sequence for the normal-side converter; the seamless power transfer control sequence includes: injecting a current into the normal-side converter based on the instantaneous total load current before the fault. The feedforward current command is simultaneously switched from the constant voltage mode to the boost compensation mode for the outer voltage loop; in boost compensation mode, the DC voltage reference value of the normal side converter is... Based on its output current Dynamically generated, represented as:

[0070] in, This is the global setting value for the DC bus voltage. This represents the output current at the moment a fault occurs on the normal side. This represents the voltage-current compensation slope. N indicates the normal side.

[0071] The feedforward power boost command is based on the total system load current at the instant before the fault occurs. The target value is to generate the output current command of the normal-side bidirectional converter. rapidly approaching .

[0072] This strategy breaks through the bottleneck of traditional feedback control lag and achieves "predictive" compensation for power demand. It dynamically adjusts the response characteristics of the outer loop controller according to the real-time load rate, realizing optimized control that prevents overshoot under high load and provides fast response under low load, thus improving the switching quality across the entire load range. By binding fault blocking with power feedforward and control mode switching, the system response chain is greatly shortened. Through the above combination, the bus voltage fluctuation and inverter output disturbance caused by a power supply failure are minimized, achieving zero-aware switching of the back-end load and meeting the extreme requirements for power supply continuity, stability, and high reliability in power supply protection scenarios for major events.

[0073] In some optional implementations, adjusting the droop coefficient in the power allocation mode to redistribute load power demand is achieved through intelligent droop coefficient collaborative adjustment based on dynamic load capacity assessment and predictive regulation, in order to achieve global optimization of power allocation under dynamic constraints. This strategy monitors the current power, simultaneously assesses the power supply's dynamic response capability and available capacity margin, and predicts short-term power demand trends based on historical load change data, fine-tuning the droop coefficient in advance to achieve pre-regulation and avoid acting only when the capacity limit is reached. Moreover, the adjustment process is not a one-sided voltage drop adjustment, but a two-sided collaborative optimization, seeking a new equilibrium point with minimum system losses or highest power utilization while satisfying constraints.

[0074] The system is set to run in dual-path mode. Power supply one (corresponding to flowmeter one) is about to run out of capacity, while power supply two has sufficient capacity.

[0075] Define the safety capacity limit for each power supply as follows: The current output is Real-time capacity margin The controller maintains a short-term predicted trend of load power changes. The collaborative controller obtains data in real time. And calculate the margin When detected ( When a warning threshold is reached (e.g., 10% of rated capacity), the intelligent power reallocation process is triggered immediately, rather than waiting until... It only activates when overloaded.

[0076] Find a new power distribution point Under the constraints and total system power requirements Under the premise of optimizing a system metric, for example: Option A (Minimize system losses): Objective function ,in This is the equivalent line and converter loss coefficient; Option B (maximizing capacity equilibrium margin): Objective function This means balancing the margins of the two power supplies as much as possible to improve the overall system's ability to withstand disturbances.

[0077] By solving this constrained optimization problem, the optimal power allocation objective is obtained. Online, fast solution algorithms, such as the Lagrange multiplier method, can be used.

[0078] Given the current droop coefficient is The corresponding initial voltage reference value is In order to transfer power from Guide to the target A new droop coefficient needs to be calculated. This strategy sets a new steady-state operating point voltage. To ensure bus voltage stability, the system still operates at a uniform DC bus voltage. Under the new steady state, the following should be satisfied:

[0079]

[0080] because It is unknown but will eventually be achieved; a virtual common voltage target can be set. For example, let ,in It is a very small positive number to ensure that the system has a clear direction of voltage regulation. Then, the inverse solution is obtained:

[0081]

[0082] The droop factor is inversely proportional to the power target allocated. The larger the power target allocated to a power source, the smaller its droop factor will be, thus making its equivalent stiffness in the parallel system higher and naturally able to bear more power.

[0083] The adjustment process uses a first-order inertial element or a ramp function for a smooth transition to prevent oscillations caused by abrupt changes in coefficients, expressed as:

[0084] in, To adjust the rate coefficient.

[0085] While adjusting the droop coefficient, based on the predicted trend Perform feedforward compensation. If the predicted load will continue to grow, calculate... At that time, Based on total power demand, proactive adjustments can be made. If a decrease in load is predicted, the adjustment range can be appropriately widened or the adjustment can be delayed to avoid unnecessary control actions.

[0086] Assuming insufficient power capacity on one side of the converter, after adjusting the coefficient, continuous monitoring is performed. And bus voltage fluctuations. If the actual power distribution does not converge as expected, a slow integral correction stage is initiated to fine-tune the process. Or fine-tune directly until satisfied .once Once the system recovers to and stabilizes above the safety threshold, it can automatically or manually switch back to dual-path load balancing mode.

[0087] Based on this strategy, the droop coefficients of the first and second bidirectional converters are adjusted to redistribute the load power of the two circuits. Specifically, this includes: triggering a power redistribution process when the capacity margin of one of the input power supplies is detected to be lower than a preset warning threshold; the power redistribution process includes: a) Based on the detected power capacity status and system operating parameters, set an optimization objective for power reallocation; based on the optimization objective, the power supply's safety capacity constraints, and the total load power demand, solve for an optimal power allocation target value. ,in These are the target output powers of the first bidirectional converter and the second bidirectional converter under the new steady state, respectively. b) Based on the preset control law and the target value of optimal power allocation DC bus voltage global setting value and the preset virtual voltage target value Calculate the new droop coefficients for the first and second bidirectional converters respectively. The control law is as follows: when the converter operates with the new droop factor and reaches the optimal power distribution target value, its corresponding DC voltage reference value is equal to the virtual voltage target value. ; c) Adjust the droop coefficients of the first and second bidirectional converters from their current values ​​to a new droop coefficient using a preset smooth transition method. Adjustment; during and after the droop coefficient adjustment process, the output power of each converter is allocated to the target value of optimal power distribution through its own closed-loop control system. Convergence continues until the insufficient power supply is resolved.

[0088] This strategy sets optimization targets to make power allocation results clear and reliable. Combined with load forecasting, it achieves predictive adjustment, significantly improving the smoothness and preventative nature of control. By setting a unified virtual voltage target and calculating the new droop coefficients on both sides in reverse collaboration, it ensures the inherent stability and non-circulating current characteristics of the adjusted system. The entire adjustment process is smooth, observable, and correctable, avoiding power oscillations and possessing adaptive capabilities to cope with model errors.

[0089] In some optional implementations, the operating mode control process further includes: in response to a system maintenance command, blocking all pulses of the bidirectional converter and controlling the bypass switch to close, switching the load to bypass power supply, and entering the bypass maintenance mode.

[0090] Specifically, when the system requires planned maintenance, software upgrades, or internal checks, maintenance personnel can issue a "system maintenance" command through the human-machine interface or remote monitoring system. Upon responding to this command, the control and protection system will execute the following standard operating sequence: Immediately send pulse blocking signals to the first, second and third bidirectional converters to turn off all their power switching devices, stop the converter power modules from working and disconnect them from the power grid; After confirming that the converter has been reliably locked, the control and protection system drives the preset bypass switch (e.g., the first bypass switch) to close. This bypass switch directly connects the corresponding AC input power supply (e.g., the first mains power) to the AC output branch; The power supply to the load is handled by a bypass switch, and the power is directly supplied to the load without flowing through any converter. At this time, the power converter module and control and protection system can be completely de-energized for safe inspection, maintenance, or replacement operations, while the power supply to the critical load remains uninterrupted throughout the entire process. After maintenance is completed, the system performs a reverse operation: first, the bypass switch is disconnected, then the converter power module is restarted and self-tested and synchronized. After the operation is stable, the bypass switch is locked and the system is switched back to converter power supply mode.

[0091] Figure 3 This is a flowchart illustrating the fault protection process, such as... Figure 3 As shown, the fault protection process includes the following steps S201 and S202.

[0092] S201 monitors the DC voltage, DC current, AC voltage, AC current, frequency, and temperature parameters of the system in real time.

[0093] S202, when any monitored parameter exceeds the corresponding safety threshold, the corresponding protection action is executed. The protection action includes: immediately or after a delay blocking the drive pulse of the relevant power device, tripping the AC / DC side switch, recording fault information and triggering an alarm.

[0094] Specifically, the system presets corresponding safety thresholds for each monitoring parameter, such as overvoltage, undervoltage, overcurrent, and overtemperature values. Once a real-time monitored value exceeds the corresponding safety threshold, the system will immediately trigger the preset protection logic. Protection actions are executed differently based on the severity and type of the fault, mainly including: Drive pulse blocking: Immediately or after a short delay, the drive pulses of power switching devices (such as IGBTs) in the fault-related converter are blocked, cutting off the abnormal energy flow at the source; Switch tripping: Controls the corresponding AC input circuit breaker, DC contactor, or output circuit breaker to trip, achieving electrical isolation; Information recording and alarms: Detailed fault information (including fault type, parameters, timestamp, etc.) is recorded on the local human-machine interface, fault indicator lights are lit, and alarm signals are sent to the remote monitoring center through the communication interface so that maintenance personnel can quickly locate and handle the fault.

[0095] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dual rectifier single inverter redundant uninterruptible power supply system, characterized by, The system comprises a power distribution system, a variable current power module, and a control and protection system. The power distribution system comprises a first AC input branch, a second AC input branch, a common DC bus, and an AC output branch. The variable current power module comprises a first bidirectional converter, a second bidirectional converter, and a third bidirectional converter; the AC sides of the first and second bidirectional converters are connected to the first and second AC input branches respectively, and the DC sides thereof are connected to the common DC bus, for converting AC input power into DC power; the DC side of the third bidirectional converter is connected to the common DC bus, and the AC side thereof is connected to the AC output branch, for converting DC power into AC output power. The control and protection system is communicatively connected to the first, second, and third bidirectional converters, for coordinating the working mode of the variable current power module and executing fault protection logic.

2. The dual rectifier single inverter redundant uninterruptible power supply system of claim 1, wherein, The first, second, and third bidirectional converters all adopt an active neutral point clamped three-level topology structure.

3. The dual rectifier single inverter redundant uninterruptible power supply system of claim 1, wherein, The power distribution system further comprises a first bypass switch and a second bypass switch. The first bypass switch is connected in parallel between the first AC input branch and the AC output branch. The second bypass switch is connected in parallel between the second AC input branch and the AC output branch. When the variable current power module exits operation, the control and protection system controls the corresponding first or second bypass switch to be closed, so as to directly feed the power of the corresponding AC input branch to the AC output branch.

4. A control protection method characterized by, The method is applied to the system of any one of claims 1 to 3, and comprises a working mode control process and a fault protection process. The working mode control process comprises: in response to normal double-path input power, controlling the first and second bidirectional converters to work in a constant voltage mode, and through coordinated control, maintaining the output power difference of the two within a preset threshold, and entering a double-path load balancing mode; in response to failure of one of the input power sources, locking the pulse of the corresponding bidirectional converter, and controlling the other normal bidirectional converter to increase the output power to bear the entire load, and switching to a single-path operation mode; in response to insufficient capacity of one of the input power sources, adjusting the droop coefficients of the first and second bidirectional converters, and re-distributing the load power of the two, and entering a power distribution mode; The fault protection process comprises: real-time monitoring of DC voltage, DC current, AC voltage, AC current, frequency, and temperature parameters of the system; when any monitored parameter exceeds the corresponding safety threshold, performing a corresponding protection action, which comprises: immediately or with a delay locking the driving pulse of the related power device, tripping the AC / DC side switch, recording fault information, and triggering an alarm.

5. The control protection method of claim 4, wherein, The maintaining of the output power difference of the two within a preset threshold through coordinated control specifically comprises: respectively acquiring the real-time output power of the first and second bidirectional converters, calculating the average power, and the deviation of the respective real-time output power from the average power; for each bidirectional converter, the following control is performed: a) judging whether the absolute value of the power deviation value is greater than a preset power deviation dead zone threshold value; if yes, starting a power deviation compensation regulator to generate a dynamic voltage compensation amount; if no, keeping the current dynamic voltage compensation amount unchanged; b) generating an independent DC voltage reference value of the converter based on real-time output power of the converter, a preset virtual droop coefficient, a DC bus voltage global set value and a dynamic voltage compensation amount , which is expressed as: wherein, is a DC bus voltage global setpoint, is a virtual droop coefficient preset for the i-th converter, is a real-time output power of the i-th converter, is a dynamic voltage compensation amount of the i-th converter, i = 1, 2, respectively representing the first bidirectional converter and the second bidirectional converter; c) the converter taking its independent DC voltage reference value as a control target, performing closed-loop regulation through its internal voltage loop and current loop, and controlling its output.

6. The control protection method of claim 5, wherein, The preset virtual droop coefficient is determined based on the rated power of the converter and the allowed maximum deviation range of the DC voltage, and is expressed as: wherein, is the maximum static adjustment of the voltage reference value of a single converter allowed by the system from no load to rated load, is the rated output power of a single converter.

7. The control protection method of claim 5, wherein, The control of the other normal bidirectional converter to increase the output power to bear the entire load specifically includes: in response to the fault confirmation signal, synchronously executing a drive pulse blocking of the fault side converter, and starting a seamless power transfer control sequence of the normal side converter; the seamless power transfer control sequence comprises: injecting a feedforward current command based on the instantaneous total load current before the fault into the normal side converter, while switching the control mode of the voltage outer loop of the normal side converter from the constant voltage mode to the boost compensation mode; In the boost compensation mode, the DC voltage reference value of the normal side converter According to its output current Dynamically generated, denoted as: wherein, Vdc_set is a DC bus voltage global set value, Iout is an output current at the time of normal side fault occurrence, K is a voltage-current compensation slope, which is adjusted according to an expected load rate required to be taken by the normal side converter before the fault occurs, the higher the expected load rate, the smaller the value of the slope.

8. The control protection method of claim 7, wherein, Voltage-current compensation slope Determined by the following steps: According to the instantaneous total load current before the fault , rated output current of the single converter , calculate the expected single-machine full-load factor ; Computing a voltage-current compensation slope , is a preset constant adjustment factor.

9. The control protection method of claim 5, wherein, Adjusting the droop coefficients of the first bidirectional converter and the second bidirectional converter to re-distribute the load power of the two paths, specifically including: When it is detected that the capacity margin of one of the input power sources is lower than a preset early warning threshold value, triggering a power redistribution process; The power redistribution process includes: a) setting an optimization target for power re-allocation according to the detected power supply capacity state and system operating parameters; solving an optimal power allocation target value based on the optimization target, the safety capacity constraint of the power supply and the total load power demand wherein are the target output powers of the first bidirectional converter and the second bidirectional converter respectively at the new steady state b) Based on the preset control law and the target value of optimal power allocation DC bus voltage global setting value and the preset virtual voltage target value Calculate the new droop coefficients for the first and second bidirectional converters respectively. The control law is as follows: when the converter operates with the new droop factor and reaches the optimal power distribution target value, its corresponding DC voltage reference value is equal to the virtual voltage target value. ; c) adjusting the droop coefficients of the first and second bi-directional converters from current values to new droop coefficients in a preset smooth transition manner adjustment; during and after the droop coefficient adjustment, the output power of each converter is adjusted by its own closed-loop control system to the optimal power distribution target value converge until the state of insufficient power supply is resolved.

10. The control protection method of claim 4, wherein, The operation mode control process further includes: in response to a system maintenance instruction, locking the pulses of all bidirectional converters, and controlling the bypass switch to be closed to switch the load to bypass power supply and enter a bypass maintenance mode.

Images