A high-voltage UPS power supply system for high-voltage lithium battery energy storage

Abstract

The present invention belongs to the technical field of intelligent power management. The present invention discloses a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage, including: obtaining a synchronous reference signal of the UPS inverter switching ripple, synchronously collecting the terminal voltage ripple of each series module and the current ripple of each parallel branch, obtaining the ripple voltage phasor and ripple current phasor of each module through synchronous correlation extraction, and calculating the dynamic impedance of the module containing resistive components and reactive components through operation; calculating the equivalent output impedance of the energy storage system based on the dynamic impedance of each module and the UPS load current information, and predicting the DC bus voltage drop amount during the power failure of the mains; when detecting an abnormal warning of the mains, pre-lifting the DC bus voltage control set value by a corresponding amplitude, so that the bus voltage still remains within the normal working range of the UPS inverter after the power failure of the mains is switched, thereby avoiding the under-voltage protection action during the switching instant and ensuring the continuity and reliability of the power supply to critical loads.

Description

Technical Field , , , , ,

[0005] , ,

[0004] ,

[0001] The present invention relates to the technical field of intelligent power management. More specifically, the present invention relates to a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage. Background Art

[0002] With the continuous improvement of the power supply continuity requirements in fields such as data centers, communication base stations, and industrial control, UPS (uninterruptible power supply) systems are widely deployed in power supply guarantee scenarios for critical loads. In recent years, high-voltage lithium battery energy storage systems have gradually replaced traditional lead-acid batteries as the mainstream energy storage solution for UPS due to their advantages such as high energy density, long cycle life, and fast response speed. In a high-voltage lithium battery energy storage system, a large number of battery modules are combined in series and parallel to meet the high-voltage and large-capacity requirements of the UPS DC bus.

[0003] However, the existing power supply matching scheme for UPS and high-voltage lithium battery energy storage has obvious deficiencies in the power-off switching link of the mains power. At the moment of mains power failure, the power supply source of the UPS DC bus switches from mains rectification to lithium battery energy storage discharge. During the switching process, the load current is entirely borne by the energy storage system. When this current flows through the internal resistance of each series module, a non-negligible voltage drop will occur. Since the internal resistance of each module is affected by factors such as the aging degree of the battery cells, state of charge (SOC), and temperature, there are individual differences and continuous dynamic changes. The existing scheme lacks an online sensing method for the real-time internal resistance state of the energy storage system and cannot accurately predict the voltage drop amount of the bus before power failure, resulting in the voltage drop amplitude of the DC bus exceeding the allowable range of the UPS inverter during the switching moment, triggering the under-voltage protection action of the UPS and even power supply interruption, seriously affecting the power supply continuity and reliability of critical loads. Summary of the Invention

[0004] In order to overcome the above problems of the prior art, the present invention proposes a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage to solve the above problems.

[0005] The present invention provides the following technical solutions: A UPS high-voltage power supply system supporting high-voltage lithium battery energy storage, comprising: A reference signal acquisition module, configured to acquire a synchronous reference signal of the switching ripple generated during the operation of the UPS inverter as a ripple reference signal; A synchronous sampling and phasor extraction module, configured to synchronously collect the terminal voltage ripple of each series module in the high-voltage lithium battery energy storage and the current ripple of each parallel branch, and respectively perform synchronous correlation extraction on the terminal voltage ripple and the current ripple based on the ripple reference signal to obtain the module ripple voltage phasor sum of each module and the module ripple current phasor corresponding to each module; The dynamic impedance calculation module is used to calculate the dynamic impedance of each module based on the module ripple voltage phasor and the module ripple current phasor, and obtain the module dynamic impedance. The voltage drop prediction module is used to predict the DC bus voltage drop during a mains power outage based on the module's dynamic impedance and the UPS's load current information, and to obtain the predicted voltage drop amount. The power supply control module is used to raise the DC bus voltage setpoint in advance according to the predicted voltage drop when an abnormal mains power warning is detected. After the mains power is cut off, the high-voltage lithium battery energy storage will continue to supply power to the UPS based on the raised DC bus voltage setpoint.

[0006] Preferably, the method for obtaining the ripple reference signal includes: obtaining the carrier synchronization pulse of the UPS controller, performing phase-locked tracking on the carrier synchronization pulse, and generating an orthogonal reference signal with the same frequency as the switch ripple, which is used as the ripple reference signal.

[0007] Preferably, the synchronous acquisition of the terminal voltage ripple of each series module and the current ripple of each parallel branch in the high-voltage lithium battery energy storage includes: The voltage across each series module is sampled using an isolated differential sampling circuit to obtain the module sampling voltage. The sampling current of each branch is obtained by independently sampling the current of each branch using current sensors that are set in each parallel branch. By using a synchronous trigger signal to control the module sampling voltage and the sampling current of each branch to start sampling at the same time, the synchronously sampled terminal voltage ripple and current ripple are obtained.

[0008] Preferably, the isolation differential sampling circuit uses a capacitive isolation front end to suppress common-mode voltage across the module and uses an oversampling analog-to-digital converter to digitize the voltage to obtain the module sampling voltage.

[0009] Preferably, the method for obtaining the module ripple voltage phasor and the corresponding module ripple current phasor of each module includes: Bandpass filtering is performed on the terminal voltage ripple and the current ripple to obtain filtered voltage ripple and filtered current ripple. The filtered voltage ripple and the filtered current ripple are multiplied by the in-phase component and quadrature component of the ripple reference signal, respectively, to obtain the multiplication result; The multiplication results are averaged using a sliding window to obtain the module ripple voltage phasor and the corresponding module ripple current phasor for each module.

[0010] Preferably, the method for obtaining the dynamic impedance of the module includes: The module dynamic impedance is obtained by performing complex division on the module ripple voltage phasor and module ripple current phasor. The module dynamic impedance is a complex number, the real part of which is the equivalent series resistive component of the module at the switching frequency, and the imaginary part of which is the equivalent reactive component of the module at the switching frequency.

[0011] Preferably, the method for obtaining the predicted voltage drop includes: The total complex impedance of the branch is obtained by summing the dynamic impedances of the series modules in the same parallel branch. The real part of the total complex impedance of the branch is the total equivalent series resistance of the branch, and the imaginary part is the total equivalent reactance of the branch. The equivalent output complex impedance of the energy storage system is calculated based on the total complex impedance of each parallel branch according to the parallel equivalence principle, and the equivalent output resistance and equivalent output reactance are obtained. Based on the UPS load current information and the equivalent output resistance, a resistive predicted voltage drop is obtained; Based on the UPS load current information and the equivalent output reactance, an additional predictive voltage drop with reactance is obtained; The predicted voltage drop is obtained by superimposing the resistive predicted voltage drop with the reactive additional predicted voltage drop.

[0012] Preferably, the resistive predicted voltage drop based on the load current information of the UPS and the equivalent output resistance includes: multiplying the steady-state load current in the load current information by the equivalent output resistance and then multiplying by a margin factor to obtain the resistive predicted voltage drop; The method of obtaining the reactive additional predicted voltage drop based on the load current information of the UPS and the equivalent output reactance includes: multiplying the estimated instantaneous current change in the load current information with the equivalent output reactance and then multiplying by a margin coefficient to obtain the reactive additional predicted voltage drop. The margin coefficient is greater than 1, and adaptive correction is performed based on the ratio of measured voltage drop to predicted voltage drop in historical power outage events.

[0013] Preferably, the step of raising the DC bus voltage setpoint in advance based on the predicted voltage drop when a mains power anomaly warning is detected includes: When the mains voltage amplitude drops to a preset percentage of the rated value, it is determined as a mains power abnormality warning; The predicted voltage drop is increased by increasing the current DC bus voltage setting value by a preset ramp rate to obtain the raised DC bus voltage setting value. The preset climb rate is determined based on the dynamic response bandwidth of the UPS inverter.

[0014] Preferably, the step of continuously supplying power to the UPS by high-voltage lithium battery energy storage based on the raised DC bus voltage setpoint after a mains power outage includes: Calculate the current distribution coefficient according to the total equivalent series resistance of each parallel branch, pre-distribute the discharge current commands of each parallel branch according to the current distribution coefficient, and use the lifted DC bus voltage setting value to close-loop control the DC bus voltage; and as the discharge process continues, periodically update the module dynamic impedance of each module. When it is detected that the equivalent series resistive component of any module increases by more than a preset threshold relative to the initial value, recalculate the predicted voltage drop amount and dynamically correct the DC bus voltage setting value.

[0015] The present invention provides a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage, which has the following beneficial effects: During the normal operation of the UPS inverter, the present invention uses the periodic ripple continuously injected into the DC bus by its switching action as a natural excitation signal, and combines the synchronous correlation extraction method to online and real-time measure the complex dynamic impedance of each series module of the high-voltage lithium battery energy storage system. Without interrupting the power supply or applying an external test excitation, it overcomes the deficiency of the existing scheme lacking a real-time sensing means for the internal resistance state of the energy storage system. On this basis, based on the real-time dynamic impedance of each module and the UPS load current information, quantitatively predict the possible voltage drop amplitude of the DC bus at the moment of switching when the mains power fails, and when detecting an abnormal warning of the mains power, pre-lift the DC bus voltage control setting value by a corresponding amplitude, so that the bus voltage after switching can still be maintained above the minimum voltage required for the normal operation of the UPS inverter even if it drops by the predicted amplitude. Fundamentally, it avoids the under-voltage protection action caused by the over-limit of the bus voltage drop at the moment of switching, and ensures the continuity and reliability of the power supply for critical loads. Brief Description of the Drawings

[0016] Figure 1 It is a module schematic diagram of a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage according to the present invention. Detailed Embodiments

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. Embodiment

[0018] [[ID=2']]Please refer to Figure 1 In this embodiment, a UPS high-voltage power supply system supporting high-voltage lithium battery energy storage includes: A reference signal acquisition module, configured to acquire a synchronous reference signal of the switching ripple generated during the operation of the UPS inverter as a ripple reference signal; The method for obtaining the ripple reference signal includes: obtaining the carrier synchronization pulse of the UPS controller, performing phase-locked tracking on the carrier synchronization pulse, and generating an orthogonal reference signal with the same frequency as the switch ripple, which is used as the ripple reference signal.

[0019] In this embodiment, during normal operation, the UPS inverter's internal pulse width modulation controller periodically drives the power switching transistors to switch on and off at a fixed carrier frequency (typically 10kHz to 20kHz). This switching action continuously injects periodic current ripples at the same frequency as the carrier frequency into the DC bus.

[0020] A carrier synchronization pulse, a square wave signal strictly synchronized with the switching action, is extracted from the digital signal processor or pulse width modulation module of the UPS controller. This pulse is tracked by a phase-locked loop (PLL). The PLL outputs a sine and cosine signal with the same frequency as the switching ripple, serving as the in-phase and quadrature components of the ripple reference signal, respectively. These two signals together form a quadrature reference signal pair used for synchronization correlation extraction in subsequent steps. Since the carrier synchronization pulse is a square wave containing abundant odd harmonics, directly using it for correlation extraction would introduce harmonic interference. Therefore, using the pure sine / cosine signal output by the PLL ensures that the correlation extraction is only sensitive to the fundamental frequency switching ripple, effectively suppressing the interference of harmonic components on subsequent impedance calculations.

[0021] The synchronous sampling and phasor extraction module is used to synchronously collect the terminal voltage ripple of each series module and the current ripple of each parallel branch in the high-voltage lithium battery energy storage, and to synchronously extract the terminal voltage ripple and current ripple based on the ripple reference signal to obtain the module ripple voltage phasor and the module ripple current phasor corresponding to each module. In this embodiment, the high-voltage lithium battery energy storage system consists of several parallel branches, with several battery modules connected in series within each branch. Taking a typical configuration as an example: the system has a total of 3 parallel branches, with 20 battery modules connected in series in each branch. Therefore, it is necessary to synchronously collect the terminal voltage ripple of 60 modules and the current ripple of the 3 branches.

[0022] The synchronous acquisition of the terminal voltage ripple of each series module and the current ripple of each parallel branch in the high-voltage lithium battery energy storage includes: The voltage across each series module is sampled using an isolated differential sampling circuit to obtain the module sampling voltage. The sampling current of each branch is obtained by independently sampling the current of each branch using current sensors that are set in each parallel branch. By using a synchronous trigger signal to control the module sampling voltage and the sampling current of each branch to start sampling at the same time, the synchronously sampled terminal voltage ripple and current ripple are obtained.

[0023] The isolated differential sampling circuit uses a capacitive isolation front end to suppress common-mode voltage across the module and uses an oversampling analog-to-digital converter to digitize the voltage to obtain the module sampling voltage.

[0024] The methods for obtaining the module ripple voltage phasor of each module and the corresponding module ripple current phasor of each module include: Bandpass filtering is performed on the terminal voltage ripple and the current ripple to obtain filtered voltage ripple and filtered current ripple. The filtered voltage ripple and the filtered current ripple are multiplied by the in-phase component and quadrature component of the ripple reference signal, respectively, to obtain the multiplication result; The multiplication results are averaged using a sliding window to obtain the module ripple voltage phasor and the corresponding module ripple current phasor for each module.

[0025] In this embodiment, it should be noted that the core challenge of terminal voltage sampling lies in the high-voltage series floating ground problem: the reference ground potentials of each module are different, and the higher the module is in series, the higher its voltage relative to the common-mode voltage of the system ground, which ordinary differential sampling circuits cannot handle. This embodiment uses a capacitive isolation front end to couple the differential-mode voltage across the module to the low-voltage sampling side through an isolation capacitor. The high-voltage common-mode component is blocked by the isolation capacitor, effectively solving the safety and accuracy problems of floating ground sampling. The oversampling analog-to-digital converter, by sampling at a rate much higher than the Nyquist frequency and then performing digital decimation filtering, can significantly improve the effective resolution without increasing the ADC bit depth, which helps to extract effective signals from small-amplitude terminal voltage ripple.

[0026] The synchronous trigger signal can be derived from the zero-crossing point of the quadrature reference signal output by the phase-locked loop, ensuring that all sampling channels start simultaneously at a fixed phase point in the switching ripple cycle, eliminating phase errors introduced by inconsistent sampling times in each channel. If there is a time deviation between the sampling of the terminal voltage of each module and the sampling of the branch current, the dynamic impedance phase angle calculated in subsequent steps will have a systematic deviation, affecting the accuracy of the imaginary part (reactive component). Therefore, synchronous sampling is a necessary prerequisite for ensuring the accuracy of impedance calculation.

[0027] In this embodiment, bandpass filtering with the switching frequency as the center frequency is applied to the synchronously acquired terminal voltage ripple of each module and the branch current ripple to filter out DC components, low-frequency slowly varying components (such as terminal voltage drift caused by SOC changes) and high-order harmonic components, and only retain the ripple components near the switching frequency to obtain filtered voltage ripple and filtered current ripple.

[0028] Subsequently, synchronous correlation extraction is performed on the filtered signal. Specifically, the filtered voltage ripple is multiplied by the corresponding in-phase component and then by the quadrature component, resulting in two multiplication results. The same operation is performed on the filtered current ripple, resulting in four multiplication results. A sliding window average is then applied to each of the four multiplication results. The average window length is an integer multiple of the switching ripple period to ensure that the averaging operation eliminates the alternating component within the ripple period, retaining only the DC component. After averaging, the two voltage results correspond to the real and imaginary parts of the module ripple voltage phasor, respectively, and the two current results correspond to the real and imaginary parts of the module ripple current phasor, respectively. This yields the module ripple voltage phasor and the corresponding module ripple current phasor, expressed in complex form.

[0029] It should be noted that only the signal components that are in phase and frequency with the reference frequency are retained in the mean after multiplication; the mean of other frequency components approaches zero. This allows the switching ripple to be reliably extracted from the noise background even when its amplitude at the module terminal voltage is extremely small, providing a high signal-to-noise ratio input for subsequent impedance calculations.

[0030] The dynamic impedance calculation module is used to calculate the dynamic impedance of each module based on the module ripple voltage phasor and the module ripple current phasor, and obtain the module dynamic impedance. The method for obtaining the dynamic impedance of the module includes: The module dynamic impedance is obtained by performing complex division on the module ripple voltage phasor and module ripple current phasor. The module dynamic impedance is a complex number, the real part of which is the equivalent series resistive component of the module at the switching frequency, and the imaginary part of which is the equivalent reactive component of the module at the switching frequency.

[0031] In this embodiment, a complex division is performed on the ripple voltage phasor and the corresponding ripple current phasor of each module, i.e., the voltage phasor is divided by the current phasor, to obtain the complex dynamic impedance of the module at the switching frequency. The real part of the complex dynamic impedance reflects the equivalent series resistance (ESR) of the module at the switching frequency, which is mainly composed of the ohmic internal resistance of the cell body, the contact resistance of the electrode tabs, and the resistance of the connecting copper busbars; the imaginary part reflects the equivalent reactance of the module at the switching frequency, which is mainly inductive at typical switching frequencies (10kHz to 20kHz), and mainly comes from the parasitic inductance of the internal connecting conductors of the module.

[0032] It should be noted that the dynamic impedance of the module changes with the module's SOC, temperature, and aging degree. As the battery's depth of discharge increases (SOC decreases) or the cell ages, the equivalent series resistive component usually shows an upward trend; when the temperature decreases, the equivalent series resistive component also increases significantly. By updating the dynamic impedance of each module in real time, subsequent voltage drop predictions are always based on the module's current true impedance state.

[0033] The voltage drop prediction module is used to predict the DC bus voltage drop during a mains power outage based on the module's dynamic impedance and the UPS's load current information, and to obtain the predicted voltage drop amount. The methods for obtaining the predicted voltage drop include: The total complex impedance of the branch is obtained by summing the dynamic impedances of the series modules in the same parallel branch. The real part of the total complex impedance of the branch is the total equivalent series resistance of the branch, and the imaginary part is the total equivalent reactance of the branch. The equivalent output complex impedance of the energy storage system is calculated based on the total complex impedance of each parallel branch according to the parallel equivalence principle, and the equivalent output resistance and equivalent output reactance are obtained. Based on the UPS load current information and the equivalent output resistance, a resistive predicted voltage drop is obtained; Based on the UPS load current information and the equivalent output reactance, an additional predictive voltage drop with reactance is obtained; The predicted voltage drop is obtained by superimposing the resistive predicted voltage drop with the reactive additional predicted voltage drop.

[0034] The process of obtaining the resistive predicted voltage drop based on the load current information of the UPS and the equivalent output resistance includes: multiplying the steady-state load current in the load current information by the equivalent output resistance and then multiplying by a margin factor to obtain the resistive predicted voltage drop. The method of obtaining the reactive additional predicted voltage drop based on the load current information of the UPS and the equivalent output reactance includes: multiplying the estimated instantaneous current change in the load current information with the equivalent output reactance and then multiplying by a margin coefficient to obtain the reactive additional predicted voltage drop. The margin coefficient is greater than 1, and adaptive correction is performed based on the ratio of measured voltage drop to predicted voltage drop in historical power outage events.

[0035] In this embodiment, the specific calculation process for predicting voltage drop is illustrated using the typical configuration described above (3 parallel branches, 20 series modules in each branch).

[0036] First, the complex dynamic impedances of each series module within each parallel branch are summed to obtain the total complex impedance of that branch. Taking the i-th branch as an example, its total equivalent series resistance is the sum of the equivalent series resistive components of the 20 modules within that branch, and its total equivalent reactance is the sum of the equivalent reactant components of the 20 modules. Since the aging degree and temperature distribution of each module are different, the dynamic impedances of different branches and even different modules within the same branch vary. By summing the impedances of each module individually, rather than simply using nominal values, the current true impedance distribution of the system can be accurately reflected.

[0037] Subsequently, the equivalent output complex impedance of the energy storage system is calculated based on the parallel equivalence principle using the total complex impedance of the three parallel branches. Specifically, the parallel equivalence calculation can be implemented by "first calculating the admittance and then taking the reciprocal": take the reciprocal of the total complex impedance of each branch to obtain the corresponding branch admittance, sum them up, and then take the reciprocal of the admittance sum to obtain the equivalent output complex impedance of the energy storage system; then take the real and imaginary parts of this equivalent output complex impedance to obtain the equivalent output resistance and equivalent output reactance. The physical significance of parallel equivalence is that the equivalent impedance presented by the system after the branches are connected in parallel is less than the impedance of any single branch, and the lower the branch impedance, the greater its current sharing ratio. The parallel equivalence calculation process can reflect this current distribution trend.

[0038] Regarding the acquisition of load current information, in this embodiment, the UPS collects the load current at its output terminal in real time. The steady-state load current can be converted from AC side quantity to DC side equivalent current using existing technology. For example, the output active power can be calculated from the UPS output voltage and output current, and the DC side steady-state current can be estimated by combining the DC bus voltage before power failure (an efficiency coefficient can be introduced for correction if necessary). Alternatively, the UPS controller can directly provide an estimated DC side current value and average it over a short time window as the steady-state discharge current that the energy storage system needs to continuously bear after power failure. Multiplying the estimated current surge at the moment of power failure with the equivalent output reactance yields the reactive additional predicted voltage drop; the sum of these two values ​​is the predicted voltage sag. The resistive predicted voltage drop corresponds to the continuous voltage drop across the module's internal resistance during steady-state discharge after power failure. The reactive additional predicted voltage drop is an engineering approximation used to conservatively cover the transient additional voltage drop caused by the module's parasitic inductance during the rapid change of current at the moment of switching. This estimation uses the equivalent reactance measured at the switching frequency as a characteristic parameter of the parasitic inductance, multiplied by the estimated current mutation to obtain an approximate upper bound for the additional voltage drop. The actual transient voltage drop may be smaller than this estimated value, and the difference is further covered by a margin coefficient. The two are superimposed to ensure the completeness and conservatism of the prediction.

[0039] Multiplying the steady-state load current by the equivalent output resistance yields the resistive predicted voltage drop, while multiplying the estimated current surge at the moment of power failure by the equivalent output reactance yields the reactive additional predicted voltage drop. The sum of these two values ​​gives the predicted voltage sag. The resistive predicted voltage drop corresponds to the continuous voltage drop across the module's internal resistance during steady-state discharge after power failure, while the reactive additional predicted voltage drop corresponds to the transient inductive voltage drop caused by the current surge at the moment of switching across the module's parasitic inductance. Both occur in the initial stage of power failure switching and therefore need to be considered together to ensure the completeness of the prediction.

[0040] In this embodiment, the margin coefficient is introduced to compensate for the following three types of errors: First, there is a frequency extrapolation deviation between the dynamic impedance measured at the switching frequency and the actual internal resistance under high-current DC discharge conditions. The switching frequency is usually above 10kHz, while power-off discharge is a DC or low-frequency condition, and the battery impedance values ​​differ at different frequencies. Second, the dynamic impedance measurement itself has measurement uncertainty due to signal-to-noise ratio limitations. Third, the actual current overshoot at the moment of power-off may exceed the expected value. The margin coefficient can be initially set to 1.1, that is, multiplying the calculated resistive predicted voltage drop and the reactive additional predicted voltage drop by 1.1 each before summing them. As historical power outage events accumulate during system operation, the measured bus voltage drop at each power outage is compared with the corresponding predicted voltage drop. If the measured value is consistently greater than the predicted value, the margin coefficient is appropriately increased; if the measured value is consistently less than the predicted value, the margin coefficient is appropriately decreased, so that the margin coefficient gradually converges to a level that matches the actual operating conditions of the system. At the same time, the margin coefficient should always be greater than 1 to ensure that the predicted value conservatively covers the actual voltage drop.

[0041] The power supply control module is used to raise the DC bus voltage setpoint in advance according to the predicted voltage drop when an abnormal mains power warning is detected. After the mains power is cut off, the high-voltage lithium battery energy storage will continue to supply power to the UPS based on the raised DC bus voltage setpoint.

[0042] The step of raising the DC bus voltage setpoint in advance based on the predicted voltage drop when a mains power anomaly warning is detected includes: When the mains voltage amplitude drops to a preset percentage of the rated value, it is determined as a mains power abnormality warning; The predicted voltage drop is increased by increasing the current DC bus voltage setting value by a preset ramp rate to obtain the raised DC bus voltage setting value. The preset climb rate is determined based on the dynamic response bandwidth of the UPS inverter.

[0043] In this embodiment, a mains voltage drop to 90% of the rated value is used as the threshold for determining a mains power anomaly warning. This threshold can be adjusted based on the actual power grid quality and UPS product specifications. Upon detecting a mains power anomaly warning, a pre-rise process for the DC bus voltage setpoint is immediately initiated: the current DC bus voltage setpoint is gradually increased by the predicted voltage drop at a preset rise rate using a slope-limited method to obtain the risen DC bus voltage setpoint. The rise rate is determined based on the dynamic response bandwidth of the UPS inverter's DC bus voltage closed-loop control. An excessively fast rise rate will exceed the tracking capability of the control loop, leading to overshoot; an excessively slow rise rate may fail to complete the rise before the mains power is completely cut off. Typically, if the UPS DC bus voltage control bandwidth is tens of hertz, the rise time constant can be set to 50ms to 200ms. In scenarios of slow undervoltage in the power grid (where a drop from the rated value to 90% typically takes hundreds of milliseconds to several seconds), this ensures that the rise is completed before the power outage.

[0044] It's important to note that this adjustment refers to the control setpoint of the DC bus voltage, not the direct manipulation of the bus voltage itself. The UPS DC bus voltage is tracked in a closed loop by the energy storage system's charging and discharging control loop, with the setpoint as the target. After raising the setpoint, the control loop drives the energy storage system to appropriately increase the output voltage until the bus voltage follows the new setpoint. This is because, at the moment of mains power failure, the entire load current is transferred to the energy storage system. The voltage drop caused by the current flowing through the internal resistance of each series module will cause the DC bus voltage to drop from its current actual value. The magnitude of this drop is closely related to the dynamic impedance and actual current carried by each module, and according to the aforementioned calculations, it may reach the magnitude of the predicted voltage drop. If the actual bus voltage value before the switch has been tracked to the raised setpoint by the control loop, even if the bus voltage drops by the aforementioned magnitude after the switch, its actual value after the drop will still remain above the minimum DC bus voltage required for normal operation of the UPS inverter, thereby avoiding undervoltage protection activation and ensuring power supply continuity.

[0045] It should be noted that the raised DC bus voltage setting should have an upper limit constraint, and must not exceed the product of the maximum allowable charging voltage set by the high-voltage lithium battery energy storage system BMS and the number of series modules, to prevent the energy storage system's overvoltage protection from being triggered during the pre-raise process. The provision that the high-voltage lithium battery energy storage system continuously supplies power to the UPS based on the raised DC bus voltage setting after a mains power outage includes: The current distribution coefficient is calculated based on the total equivalent series resistance of each parallel branch. The discharge current command of each parallel branch is pre-distributed according to the current distribution coefficient. The DC bus voltage is controlled in a closed loop using the raised DC bus voltage setting value. As the discharge process continues, the module dynamic impedance of each module is periodically updated. When the equivalent series resistive component of any module is detected to increase more than the preset threshold relative to the initial value, the predicted voltage drop is recalculated and the DC bus voltage setting value is dynamically corrected.

[0046] In this embodiment, after the mains power fails, the energy storage system enters an independent discharge mode. Specifically, the current distribution coefficient of each branch is inversely proportional to its total equivalent series resistance. That is, the branch with lower impedance bears a larger proportion of the discharge current, and the branch with higher impedance bears a smaller proportion. The discharge current command for each branch is pre-distributed according to this ratio, and the DC bus voltage is simultaneously controlled in a closed loop with the raised DC bus voltage setpoint as the target. The pre-distributed current command ensures that each branch bears the load in proportion to impedance matching during the initial switching phase, avoiding overload of a branch due to impedance differences and triggering BMS protection.

[0047] As the discharge process continues, online impedance measurement operates continuously and periodically updates the dynamic impedance of each module. When the equivalent series resistive component of any module is detected to have increased by more than a preset threshold (e.g., 10%) relative to its initial value at the time of power failure, a recalculation is triggered: the updated dynamic impedance of each module is used to recalculate the new predicted voltage drop, and the DC bus voltage setpoint is dynamically corrected accordingly, ensuring that the bus voltage setpoint is always higher than the minimum operating voltage required after the actual voltage drop. This dynamic correction mechanism effectively compensates for the gradual increase in internal resistance caused by the decrease in SOC during continuous discharge, ensuring the continuous reliability of UPS power supply in long-term discharge scenarios.

[0048] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0049] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0050] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-voltage lithium battery energy storage complete UPS high-voltage power supply system, characterized in that, include: The reference signal acquisition module is used to acquire the synchronous reference signal of the switching ripple generated during the operation of the UPS inverter, and use it as the ripple reference signal. The synchronous sampling and phasor extraction module is used to synchronously collect the terminal voltage ripple of each series module and the current ripple of each parallel branch in the high-voltage lithium battery energy storage, and to synchronously extract the terminal voltage ripple and current ripple based on the ripple reference signal to obtain the module ripple voltage phasor and the module ripple current phasor corresponding to each module. The dynamic impedance calculation module is used to calculate the dynamic impedance of each module based on the module ripple voltage phasor and the module ripple current phasor, and obtain the module dynamic impedance. The voltage drop prediction module is used to predict the DC bus voltage drop during a mains power outage based on the module's dynamic impedance and the UPS's load current information, and to obtain the predicted voltage drop amount. The power supply control module is used to raise the DC bus voltage setpoint in advance according to the predicted voltage drop when an abnormal mains power warning is detected. After the mains power is cut off, the high-voltage lithium battery energy storage will continue to supply power to the UPS based on the raised DC bus voltage setpoint.

2. The UPS high-voltage power supply system for high-voltage lithium battery energy storage according to claim 1, characterized in that, The method for obtaining the ripple reference signal includes: obtaining the carrier synchronization pulse of the UPS controller, performing phase-locked tracking on the carrier synchronization pulse, and generating an orthogonal reference signal with the same frequency as the switch ripple, which is used as the ripple reference signal.

3. The high-voltage lithium battery energy storage complete UPS high-voltage power supply system according to claim 1, characterized in that, The synchronous acquisition of the terminal voltage ripple of each series module and the current ripple of each parallel branch in the high-voltage lithium battery energy storage includes: The voltage across each series module is sampled using an isolated differential sampling circuit to obtain the module sampling voltage. The sampling current of each branch is obtained by independently sampling the current of each branch using current sensors that are set in each parallel branch. By using a synchronous trigger signal to control the module sampling voltage and the sampling current of each branch to start sampling at the same time, the synchronously sampled terminal voltage ripple and current ripple are obtained.

4. The high-voltage lithium battery energy storage complete UPS high-voltage power supply system according to claim 3, characterized in that, The isolated differential sampling circuit uses a capacitive isolation front end to suppress common-mode voltage across the module and uses an oversampling analog-to-digital converter to digitize the voltage to obtain the module sampling voltage.

5. The high-voltage lithium battery energy storage complete UPS power supply system according to claim 1, characterized in that, The methods for obtaining the module ripple voltage phasor of each module and the corresponding module ripple current phasor of each module include: Bandpass filtering is performed on the terminal voltage ripple and the current ripple to obtain filtered voltage ripple and filtered current ripple. The filtered voltage ripple and the filtered current ripple are multiplied by the in-phase component and quadrature component of the ripple reference signal, respectively, to obtain the multiplication result; The multiplication results are averaged using a sliding window to obtain the module ripple voltage phasor and the corresponding module ripple current phasor for each module.

6. The high-voltage lithium battery energy storage complete UPS high-voltage power supply system according to claim 5, characterized in that, The method for obtaining the dynamic impedance of the module includes: The module dynamic impedance is obtained by performing complex division on the module ripple voltage phasor and module ripple current phasor. The module dynamic impedance is a complex number, the real part of which is the equivalent series resistive component of the module at the switching frequency, and the imaginary part of which is the equivalent reactive component of the module at the switching frequency.

7. The high-voltage lithium battery energy storage complete UPS power supply system according to claim 6, characterized in that, The method for obtaining the predicted voltage drop includes: The total complex impedance of the branch is obtained by summing the dynamic impedances of the series modules in the same parallel branch. The real part of the total complex impedance of the branch is the total equivalent series resistance of the branch, and the imaginary part is the total equivalent reactance of the branch. The equivalent output complex impedance of the energy storage system is calculated based on the total complex impedance of each parallel branch according to the parallel equivalence principle, and the equivalent output resistance and equivalent output reactance are obtained. Based on the UPS load current information and the equivalent output resistance, a resistive predicted voltage drop is obtained; Based on the UPS load current information and the equivalent output reactance, an additional predictive voltage drop with reactance is obtained; The predicted voltage drop is obtained by superimposing the resistive predicted voltage drop with the reactive additional predicted voltage drop.

8. The high-voltage lithium battery energy storage complete UPS power supply system according to claim 7, characterized in that, The process of obtaining the resistive predicted voltage drop based on the load current information of the UPS and the equivalent output resistance includes: multiplying the steady-state load current in the load current information by the equivalent output resistance and then multiplying by a margin factor to obtain the resistive predicted voltage drop. The method of obtaining the reactive additional predicted voltage drop based on the load current information of the UPS and the equivalent output reactance includes: multiplying the estimated instantaneous current change in the load current information with the equivalent output reactance and then multiplying by a margin coefficient to obtain the reactive additional predicted voltage drop. The margin coefficient is greater than 1, and adaptive correction is performed based on the ratio of measured voltage drop to predicted voltage drop in historical power outage events.

9. The high-voltage lithium battery energy storage complete UPS power supply system according to claim 8, characterized in that, The step of raising the DC bus voltage setpoint in advance based on the predicted voltage drop when a mains power anomaly warning is detected includes: When the mains voltage amplitude drops to a preset percentage of the rated value, it is determined as a mains power abnormality warning; The predicted voltage drop is increased by increasing the current DC bus voltage setting value by a preset ramp rate to obtain the raised DC bus voltage setting value. The preset climb rate is determined based on the dynamic response bandwidth of the UPS inverter.

10. A UPS high-voltage power supply system for high-voltage lithium battery energy storage according to claim 9, characterized in that, The provision that, after a mains power outage, the high-voltage lithium battery energy storage system continuously supplies power to the UPS based on the elevated DC bus voltage setpoint includes: The current distribution coefficient is calculated based on the total equivalent series resistance of each parallel branch. The discharge current command of each parallel branch is pre-distributed according to the current distribution coefficient. The DC bus voltage is controlled in a closed loop using the raised DC bus voltage setting value. As the discharge process continues, the module dynamic impedance of each module is periodically updated. When the equivalent series resistive component of any module is detected to increase more than the preset threshold relative to the initial value, the predicted voltage drop is recalculated and the DC bus voltage setting value is dynamically corrected.

Images