Uninterruptible power supply control method and system based on twelve-pulse rectification

Abstract

This application relates to the field of power supply control and discloses a control method and system for an uninterruptible power supply (UPS) based on a twelve-pulse rectification. The method includes: acquiring real-time load demand parameters and real-time mains input parameters of the UPS; the real-time load demand parameters include the load active power requirement and the load current harmonic content requirement; the real-time mains input parameters include the effective value of the mains voltage and the mains frequency; constructing a rectifier-filter coordinated control model and determining the bi-objective optimization function and constraints; solving the bi-objective optimization function using a dynamic programming algorithm to obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter; and controlling the firing circuit parameters of the twelve-pulse rectifier bridge and the switching action of the 11th filter based on the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. This application can achieve both high power quality and low operating losses.

Description

Technical Field

[0001] This application relates to the field of power control technology, and more specifically, to a method and system for controlling an uninterruptible power supply based on twelve-pulse rectification. Background Technology

[0002] In demanding applications such as data centers and medical equipment, existing uninterruptible power supplies (UPS) face a core contradiction between harmonic suppression and operational efficiency. To achieve high-quality power output with low total harmonic distortion (TDH) of the input current, UPS systems require complex filtering schemes (such as 11th order filters) and precise firing angle control (such as twelve-pulse rectifier bridges). This increases equipment losses and reduces overall operating efficiency. Conversely, pursuing high-efficiency operation (such as reducing filter switching or relaxing firing angle control) leads to an increase in input current harmonic content, failing to meet the stringent power quality requirements of critical loads.

[0003] The existing uninterruptible power supply (UPS) uses a 12-pulse rectifier bridge and an 11th-order filter operating independently, resulting in unstable harmonic suppression and an inability to balance high power quality and low operating losses. Relying solely on the 12-pulse rectification results in a total harmonic distortion (THD) of approximately 10% in the input current, which is insufficient for demanding scenarios. While operating the 11th-order filter alone can reduce the THD, the copper and iron losses of the filter increase the UPS's operating costs, and the filter's harmonic suppression performance fluctuates when the rectifier bridge output current changes. Summary of the Invention

[0004] The purpose of this application is to provide an uninterruptible power supply control method and system based on twelve-pulse rectification, which solves the technical problem of being unable to simultaneously achieve high power quality and low operating loss, and achieves the technical effect of balancing high power quality and low operating loss.

[0005] This application provides a control method for an uninterruptible power supply (UPS) based on a twelve-pulse rectifier. The method includes: acquiring real-time load demand parameters and real-time mains input parameters of the UPS; the real-time load demand parameters include the load active power requirement and the load current harmonic content requirement; the real-time mains input parameters include the effective value of the mains voltage and the mains frequency; constructing a rectifier-filter coordinated control model and determining a bi-objective optimization function and constraints; wherein, the rectifier-filter coordinated control model includes the output current characteristics of the twelve-pulse rectifier bridge, the impedance characteristics of the 11th-order filter, and the interaction characteristics; the interaction characteristics characterize the interaction between the output current characteristics of the twelve-pulse rectifier bridge and the impedance characteristics of the 11th-order filter; the first objective of the bi-objective optimization function is that the total harmonic distortion (THD) of the UPS input current is less than 4. The second objective of the bi-objective optimization function is to minimize the operating loss of the 11th filter, which includes copper and iron losses. Constraints include that the maximum output current of the twelve-pulse rectifier bridge is less than its rated current, and the switching state of the 11th filter is less than its design capacity. The bi-objective optimization function is solved using dynamic programming to obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. Based on these parameters, the triggering circuit parameters of the twelve-pulse rectifier bridge and the switching action of the 11th filter are controlled. The optimal firing angle controls the harmonic content of the rectifier bridge output current, and the optimal switching state includes operation or shutdown to control the filter's operating loss.

[0006] In one possible implementation, the method further includes: obtaining the real-time total harmonic distortion (THD) of the uninterruptible power supply (UPS) input current and the loss value of the 11th filter; when the THD of the UPS input current is greater than or equal to 4.5%, reconstructing the collaborative control model of the twelve-pulse rectifier bridge and the 11th filter and solving it again; when the operating loss of the 11th filter is greater than or equal to the preset loss, redetermining the optimization objective weight of the bi-objective optimization function; wherein the preset loss is 1.2 times the rated loss of the 11th filter.

[0007] In another possible implementation, when the operating loss of the 11th-order filter is greater than or equal to a preset threshold, the optimization objective weights of the bi-objective optimization function are redefined, including: when the operating loss of the 11th-order filter is greater than or equal to the preset threshold, determining the operating loss range to which the operating loss of the 11th-order filter belongs; wherein, the first operating loss range is 1.2 to 1.5 times the rated loss of the 11th-order filter, the second operating loss range is 1.5 to 2 times the rated loss of the 11th-order filter, and the third operating loss range is the 11th-order filter's rated loss. The operating loss is more than twice the rated loss of the waveguide; when the operating loss range is within the first operating loss range, the weight of the first target is determined to be 0.5~0.6 and the weight of the second target is determined to be 0.4~0.5; when the operating loss range is within the second operating loss range, the weight of the first target is determined to be 0.3~0.4 and the weight of the second target is determined to be 0.6~0.7; when the operating loss range is within the third operating loss range, the weight of the first target is determined to be 0.2~0.3 and the weight of the second target is determined to be 0.7~0.8.

[0008] In another possible implementation, the method further includes: obtaining the real-time load current parameters of the uninterruptible power supply; calculating the load current change rate corresponding to the real-time load current parameters using the differential method; when the load current change rate is greater than or equal to a preset load current change rate, setting the solution frequency of the dynamic programming algorithm to a first solution frequency; when the load current change rate is less than the preset load current change rate, setting the solution frequency of the dynamic programming algorithm to a second solution frequency; wherein the second solution frequency is less than the first solution frequency.

[0009] In another possible implementation, the method further includes: determining the upper limit constraint of the output current change rate of the twelve-pulse rectifier bridge based on the rated current and thermal stability parameters of the twelve-pulse rectifier bridge; when the current change rate corresponding to the firing angle of the twelve-pulse rectifier bridge obtained by the dynamic programming algorithm exceeds or equals the upper limit constraint of the output current change rate, increasing the firing angle adjustment step size to reduce the output current change rate of the twelve-pulse rectifier bridge.

[0010] In another possible implementation, the method further includes: acquiring historical operating data of the uninterruptible power supply under different combinations of operating conditions within the data update period; wherein, different combinations of operating conditions correspond to different load active power levels, load current harmonic content ranges, mains voltage RMS ranges, and mains frequency fluctuation ranges; the historical operating data includes the firing angle, output current, and rated current operating percentage of the twelve-pulse rectifier bridge, the switching status of the 11th filter, copper loss data, iron loss data, and the measured value of the total harmonic distortion rate of the input current; and determining that the measured value of the total harmonic distortion rate of the input current in the historical operating data is greater than or equal to 4.5%. The target time period is defined as when the operating loss of the 11th filter is greater than or equal to a preset loss. Data before fluctuation is acquired within a preset first time period before the target time period, and data after fluctuation is acquired within a preset second time period after the target time period. Both the pre-fluctuation and post-fluctuation data include load parameters, mains parameters, harmonic data, and loss data. Based on the pre-fluctuation, target time period, and post-fluctuation data, the variation curves of load active power, mains voltage RMS value, input current total harmonic distortion rate, and 11th filter operating loss are determined. The peak value and fluctuation amplitude of the variation curves under different combined operating conditions are then analyzed. The study determined the influence of each operating condition combination on harmonic suppression and loss control based on the degree and duration of compliance. Based on this influence, dynamic optimization parameters for each operating condition combination were determined. The target operating condition combination and its corresponding dynamic optimization parameters were determined based on the load active power level and mains voltage fluctuation range. These dynamic optimization parameters were then updated to a bi-objective optimization function. The updated bi-objective optimization function was solved using a dynamic programming algorithm to determine the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. The dynamic optimization parameters included the dynamic total harmonic distortion rate threshold and loss optimization weights.

[0011] In another possible implementation, the method further includes: determining the first duration of the first time period, the second duration of the second time period, and the data update duration based on the target duration of the target time period; wherein the data update duration is a preset multiple of the sum of the first duration, the target duration, and the second duration.

[0012] In another possible implementation, the influence of each operating condition combination on harmonic suppression and loss control is determined based on the peak value, fluctuation amplitude, and time to reach the target value of the variation curves under different operating conditions. This includes: determining the curve characteristics corresponding to each operating condition combination based on the harmonic distortion rate curve and loss variation curve corresponding to each operating condition combination; wherein the curve characteristics include the curve peak value, curve fluctuation amplitude, and curve time to reach the target value; determining the quantitative correlation between the operating conditions combination and the curve characteristics through a correlation mining algorithm; and determining the dynamic optimization parameters corresponding to each operating condition combination based on the influence law. This includes: when the peak value of the harmonic distortion rate curve is greater than or equal to the preset peak value, and the time for the loss change curve to meet the standard is less than the preset time for the curve to meet the standard, multiplying the total harmonic distortion rate threshold of the input current by a first proportional value; when the fluctuation amplitude of the curve corresponding to the loss change curve is less than the preset fluctuation amplitude, and the time for the curve to meet the standard is greater than or equal to the preset time for the curve to meet the standard, multiplying the weight of the first target by a second proportional value, and multiplying the weight of the second target by a third proportional value; wherein, the first proportional value is less than 1, the second proportional value is greater than 1, and the third proportional value is less than 1.

[0013] In another possible implementation, the method further includes: determining the product of the data update duration and the proportion of the target duration as the preset curve target duration; wherein, the proportion of the target duration is determined by the curve peak value and the curve fluctuation amplitude.

[0014] This application also provides an uninterruptible power supply control system based on twelve-pulse rectification, including a unit for implementing the above-described uninterruptible power supply control method based on twelve-pulse rectification.

[0015] The beneficial effects of the embodiments in this application compared with the prior art are:

[0016] This application provides a control method for an uninterruptible power supply (UPS) based on a twelve-pulse rectifier. The method includes: acquiring real-time load demand parameters and real-time mains input parameters of the UPS; the real-time load demand parameters include the load active power requirement and the load current harmonic content requirement; the real-time mains input parameters include the effective value of the mains voltage and the mains frequency; constructing a rectifier-filter collaborative control model and determining the bi-objective optimization function and constraints; solving the bi-objective optimization function using a dynamic programming algorithm to obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter; and controlling the firing circuit parameters of the twelve-pulse rectifier bridge and the switching action of the 11th filter based on the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. The method in this application, by establishing a collaborative control model of the twelve-pulse rectifier bridge and the 11th filter, comprehensively considers the output current characteristics of the rectifier bridge, the impedance characteristics of the filter, and their interaction, avoiding fluctuations in the filter's harmonic suppression effect caused by changes in the rectifier bridge's output current when the two operate independently, thus improving the stability of the harmonic suppression effect. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0018] Figure 1 A flowchart illustrating the first uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0019] Figure 2 A schematic diagram illustrating the workflow of the first uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0020] Figure 3 A flowchart illustrating a second uninterruptible power supply control method based on twelve-pulse rectification provided in an embodiment of this application;

[0021] Figure 4 A flowchart illustrating the third uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0022] Figure 5 A schematic diagram illustrating the workflow of the third uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0023] Figure 6A flowchart illustrating the fourth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0024] Figure 7 A schematic diagram illustrating the workflow of the fourth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0025] Figure 8 A flowchart illustrating the fifth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0026] Figure 9 A schematic diagram illustrating the workflow of the fifth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0027] Figure 10 A flowchart illustrating the sixth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment;

[0028] Figure 11 This is a schematic diagram of the logic structure of an uninterruptible power supply control system based on twelve-pulse rectification, provided as an embodiment of this application. Detailed Implementation

[0029] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0030] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0031] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0032] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0033] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0034] The existing uninterruptible power supply's twelve-pulse rectifier bridge and eleventh-order filter operate independently, resulting in unstable harmonic suppression and an inability to balance high power quality and low operating losses.

[0035] Based on the above reasons, this application provides a control method for an uninterruptible power supply (UPS) based on a twelve-pulse rectifier. The method includes: acquiring real-time load demand parameters and real-time mains input parameters of the UPS; the real-time load demand parameters include the load active power requirement and the load current harmonic content requirement; the real-time mains input parameters include the effective value of the mains voltage and the mains frequency; constructing a rectifier-filter collaborative control model and determining the bi-objective optimization function and constraints; solving the bi-objective optimization function using a dynamic programming algorithm to obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter; and controlling the trigger circuit parameters of the twelve-pulse rectifier bridge and the switching action of the 11th filter based on the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. The method in this application, by establishing a collaborative control model for the twelve-pulse rectifier bridge and the 11th filter, comprehensively considers the output current characteristics of the rectifier bridge, the impedance characteristics of the filter, and their interaction, avoiding fluctuations in the filter harmonic suppression effect caused by changes in the rectifier bridge output current when the two operate independently, thus improving the stability of the harmonic suppression effect.

[0036] In some scenarios, the uninterruptible power supply control method based on twelve-pulse rectification according to the embodiments of this application can be applied to the control of uninterruptible power supplies, which can improve the power supply quality and enhance the control effect of uninterruptible power supplies in high-requirement scenarios such as data centers and medical equipment.

[0037] The following describes in detail, with specific examples, an uninterruptible power supply control method based on twelve-pulse rectification provided in the embodiments of this application.

[0038] Figure 1 A flowchart illustrating the first uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 1As shown, the uninterruptible power supply control method based on twelve-pulse rectification includes S110 to S130, and S110 to S130 will be explained in detail below.

[0039] S110. Obtain the real-time load demand parameters and real-time mains input parameters of the uninterruptible power supply. The real-time load demand parameters include the load active power requirement and the load current harmonic content requirement. The real-time mains input parameters include the mains voltage RMS value and the mains frequency.

[0040] Figure 2 A schematic diagram illustrating the workflow of the first uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 2 As shown, two types of key parameters of the uninterruptible power supply can be continuously collected. One type is real-time data reflecting load demand, including the load's active power demand and current harmonic content requirements; the other type is real-time status data of the mains input, including the effective voltage value and frequency. These data provide the basis for subsequent control strategy formulation.

[0041] For example, the active power requirement for a data center server load may be 500kW at full load, and the current harmonic content may not exceed 5% to prevent the server from crashing; another type is real-time mains input parameters, which include the effective value of the mains voltage reflecting voltage stability and the frequency reflecting the power supply synchronization status.

[0042] For example, in industrial scenarios, the effective value of the mains voltage is detected as 380V and the frequency is 50Hz, while in civilian scenarios it is 220V and 50Hz. These parameters are the basis for subsequent control strategies to ensure that the strategies are adapted to the current load and mains power status.

[0043] For example, real-time load demand parameters can be obtained in combination with specific scenarios. For instance, when a server cluster in a data center is used as a load, the load active power requirement in its real-time load demand parameters can be reflected as the active power required for the server to run at full load, and the load current harmonic content requirement can be set as the allowable value to avoid harmonic interference with the server operation.

[0044] S120. Construct a rectifier-filter coordinated control model and determine the dual-objective optimization function and constraints. The rectifier-filter coordinated control model includes the output current characteristics of the twelve-pulse rectifier bridge, the impedance characteristics of the 11th filter, and the interaction characteristics. The interaction characteristics characterize the interaction between the output current characteristics of the twelve-pulse rectifier bridge and the impedance characteristics of the 11th filter. The first objective of the dual-objective optimization function is to ensure that the total harmonic distortion rate of the uninterruptible power supply's input current is less than 4.5%. The second objective of the dual-objective optimization function is to minimize the operating losses of the 11th filter. Operating losses include copper losses and iron losses. The constraints include that the maximum output current of the twelve-pulse rectifier bridge is less than the rated current of the twelve-pulse rectifier bridge, and the switching state of the 11th filter is less than the design capacity of the 11th filter.

[0045] This implementation integrates the operating characteristics of a twelve-pulse rectifier bridge and an eleventh-order filter to build a coordinated rectifier-filter control model, while clearly defining the dual-objective optimization direction and equipment safety constraints. The coordinated control model is used to correlate the interaction between the rectifier bridge and the filter, the dual-objective function balances harmonic suppression and filter loss, and the constraints ensure that the equipment operates within its rated range.

[0046] In this implementation, the rectifier-filter coordinated control model encompasses three key characteristics: the output current characteristics of the twelve-pulse rectifier bridge (referring to the harmonic distribution and amplitude variation of the output current, for example, the amplitude of the 11th harmonic is 8% of the fundamental frequency when the firing angle is 10°), the impedance characteristics of the 11th filter (low impedance to the 11th harmonic (e.g., 0.5Ω) and high impedance to the fundamental frequency (e.g., 100Ω)), and the interaction characteristics (referring to the cyclic relationship in which changes in rectifier bridge harmonics affect filter impedance, which in turn affects rectifier bridge harmonics). For example, an increase in the amplitude of the 11th harmonic of the rectifier bridge will increase the filter shunt, raise the temperature, and slightly increase the impedance, which in turn will cause the amplitude of the rectifier bridge harmonics to rise again. This interaction is the interaction characteristic.

[0047] For example, the definition of the interaction characteristic can be illustrated by a specific scenario. For instance, when the amplitude of the 11th harmonic in the output current of the twelve-pulse rectifier bridge increases, the 11th filter, due to its low impedance to the 11th harmonic, will shunt more harmonic current, causing a slight change in the filter's own impedance characteristics under high current, which in turn affects the harmonic distribution of the rectifier bridge output current. This interaction is the interaction characteristic.

[0048] In this implementation, the dual-objective optimization function sets two core objectives: First, to control the total harmonic distortion (THD-i) of the UPS input current within 4.5%. The THD-i of the UPS input current can be calculated by collecting the input current with a power quality analyzer and decomposing the fundamental and harmonic frequencies using FFT. For example, when the fundamental frequency is 100A and the 11th harmonic is 4A, THD-i ≈ 4.12%. Second, to minimize the operating losses (including copper and iron losses) of the 11th harmonic filter to ensure system energy efficiency.

[0049] In this implementation, the operating loss of the 11th filter consists of copper loss and iron loss. Copper loss can be calculated by measuring the coil resistance and current. For example, when the coil resistance is 0.1Ω and the current is 10A, the copper loss is 10W. Iron loss can be measured by an iron loss tester to obtain an empirical value. For example, if the iron core has an iron loss of 5W, the total loss is 15W.

[0050] In this implementation, the constraints ensure equipment safety: the maximum output current of the twelve-pulse rectifier bridge must not exceed the rated current (for example, the rated current of the rectifier bridge adapted to a 100kVA UPS is 150A); the switching state of the 11th filter must not exceed the design capacity (for example, when the design capacity is 20A, the harmonic current of 15A can be safely put into operation).

[0051] For example, the rated current of the twelve-pulse rectifier bridge can be determined according to the equipment specifications. For instance, a certain model of twelve-pulse rectifier bridge is adapted to a 50kVA UPS system, and its rated current can be set to a specific value to ensure long-term stable operation under this current.

[0052] It should be noted that in this UPS control method based on twelve-pulse rectification, the current path is as follows: AC power supply, phase-shifting transformer, twelve-pulse rectifier bridge, (part of the harmonic current is diverted to) the 11th order filter, DC bus or load. The 11th order filter is not a necessary path for the current, but rather provides a low-impedance bypass path for the harmonic current, thereby achieving the purpose of harmonic mitigation.

[0053] It should be noted that because the UPS adopts a controllable rectification topology on the input side and actively changes its own input circuit characteristics through a dual-objective optimization algorithm and coordinated control of rectification and filtering, it can reverse control the current waveform flowing from the grid into the UPS, thereby achieving the target of current harmonic distortion rate (THD-i) of less than 4.5%.

[0054] For example, the design capacity of the 11th harmonic filter can be set according to system requirements. For instance, for a twelve-pulse rectifier bridge with a rated current of a specific value, the design capacity of the 11th harmonic filter can be set to withstand the 11th harmonic current generated by the rectifier bridge under maximum load, ensuring that there is no overload during switching.

[0055] S130. Solve the bi-objective optimization function using a dynamic programming algorithm to obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. Based on the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter, control the firing circuit parameters of the twelve-pulse rectifier bridge and control the switching action of the 11th filter. The optimal firing angle is used to control the harmonic content of the rectifier bridge output current, and the optimal switching state includes operation or shutdown to control the operating loss of the filter.

[0056] In this implementation, a dynamic programming algorithm can be used to solve the bi-objective optimization function to obtain two controllable key parameters: the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. Then, based on these two parameters, the firing circuit of the rectifier bridge and the switching of the filter can be adjusted to control the harmonic content and loss.

[0057] For example, the total harmonic distortion rate of the input current of an uninterruptible power supply can be detected by a power quality monitoring module.

[0058] For example, the detection method of filter operating loss can be achieved through parameter calculation. For instance, for copper loss, the resistance value of the filter coil and the effective value of the current passing through it can be detected and calculated using the copper loss calculation formula. Iron loss includes hysteresis loss and eddy current loss, which are related to magnetic flux density and frequency. It is a constant loss that does not change with the load, so it can be measured and stored in advance.

[0059] For example, the optimal firing angle can be determined in combination with the operating scenario. For instance, when the load is under medium load, the optimal firing angle of the twelve-pulse rectifier bridge can be obtained by solving the dynamic programming algorithm. At this time, the 11th harmonic content of the rectifier bridge output current can be controlled at a low level.

[0060] For example, the optimal switching state can be determined based on the harmonic content. For instance, when the total harmonic distortion rate of the UPS input current is detected to be higher than 3%, the optimal switching state is to put it into operation, and the 11th harmonic filter is put into operation to divert harmonics; when the total harmonic distortion rate is lower than 1%, the optimal switching state is to shut it down to reduce unnecessary losses.

[0061] In this implementation, when using a dynamic programming algorithm to solve the dual objective function, different combinations of firing angles and switching states can be traversed to find the optimal solution. For example, when the load is 2kW and the mains voltage is 220V, the optimal firing angle is 12°. At this time, the amplitude of the 11th harmonic of the rectifier bridge is 3% of the fundamental frequency, which meets the THD-i requirement.

[0062] For example, the dynamic programming algorithm can be an ε-constrained dynamic programming algorithm or a weighted sum dynamic programming (WDP) algorithm, which can determine the optimal solution.

[0063] In this implementation, the parameters of the rectifier bridge trigger circuit can be adjusted according to the optimal firing angle to change the output current waveform; the filter switching action (operation or shutdown) can be controlled according to the optimal switching state. For example, the filter can be operated to shunt harmonics when THD-i is 4%, and shut down when THD-i is 1.5% to reduce losses.

[0064] This implementation establishes a collaborative control model for the twelve-pulse rectifier bridge and the 11th-order filter. It comprehensively considers the output current characteristics of the rectifier bridge, the impedance characteristics of the filter, and the interaction between the two, thus avoiding fluctuations in the harmonic suppression effect of the filter caused by changes in the output current of the rectifier bridge when the two operate independently, and improving the stability of the harmonic suppression effect.

[0065] This implementation method obtains real-time load demand parameters and real-time mains input parameters. Considering the interactive influence characteristics of the twelve-pulse rectifier bridge and the 11th filter, the optimal trigger angle and switching state are solved by dynamic programming algorithm. Based on this, the rectifier bridge trigger circuit parameters and filter switching are controlled to respond to real-time parameter changes. The constraints ensure equipment safety and improve control adaptability and reliability.

[0066] Figure 3 A flowchart illustrating the second uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 3 As shown, in some implementations, the above method also includes S140 to S150, which will be described in detail below.

[0067] S140: Obtain the real-time input current total harmonic distortion rate and 11th filter loss value of the uninterruptible power supply.

[0068] In this implementation, two key operating indicators can be collected in real time through a power parameter monitoring system: one is the total harmonic distortion rate (THD-i) of the input current; the other is the operating loss of the 11th filter. These data are used to evaluate the effectiveness of the current control strategy in real time and provide a basis for subsequent adjustments.

[0069] S150. When the total harmonic distortion rate of the uninterruptible power supply input current is greater than or equal to 4.5%, reconstruct the collaborative control model of the twelve-pulse rectifier bridge and the 11th filter and solve it again. When the operating loss of the 11th filter is greater than or equal to the preset loss, redetermine the optimization objective weights of the bi-objective optimization function. The preset loss is 1.2 times the rated loss of the 11th filter.

[0070] In this implementation, when the total harmonic distortion rate of the input current reaches or exceeds 4.5%, it indicates that the current model can no longer meet the harmonic suppression requirements, and the collaborative control model needs to be reconstructed. During reconstruction, the output current characteristics of the twelve-pulse rectifier bridge need to be updated (e.g., changes in the rectifier bridge current harmonic distribution caused by load or mains power variations), the impedance characteristics of the 11th-order filter need to be updated (e.g., increased coil resistance due to temperature rise), and the interaction between the two needs to be requantified (e.g., the reverse effect of increased rectifier bridge harmonic current on the filter's shunting effect).

[0071] For example, when the mains voltage drops from 220V to 210V and the load power increases from 50% to 100%, the 11th harmonic of the rectifier bridge output current increases from 5% to 8%, and the 11th harmonic impedance of the filter increases from 0.5Ω to 0.55Ω due to the temperature rise. At this time, it is necessary to re-collect the current harmonic distribution data of the rectifier bridge and the current impedance data of the filter, reconstruct the model by combining the new interaction relationship, and then use the dynamic programming algorithm to solve for the new optimal firing angle and switching state.

[0072] In this implementation, the preset loss is set to 1.2 times the rated loss of the 11th filter, which is used to indicate that the loss is close to the critical value. When the filter loss reaches or exceeds this value, the weight of the bi-objective optimization function can be adjusted to balance the priority of harmonic suppression and loss control.

[0073] When redefining weights, the weighting of loss targets can be increased while the weighting of harmonic suppression targets can be decreased, making the optimization more focused on reducing losses. For example, a filter has a rated loss of 100W and a preset loss of 120W; when the filter's total loss rises to 130W due to an increase in shunt current, the original weighting ratio (harmonic suppression: loss control = 6:4) can be adjusted to 5:5, allowing the optimization algorithm to prioritize control parameters that reduce losses, while ensuring that THD-i remains less than 4.5%.

[0074] This implementation method monitors the THD of the uninterruptible power supply input current and the loss value of the 11th filter in real time. When the THD exceeds 4.5%, it returns to re-establish the collaborative control model for solution. When the filter loss exceeds the preset loss, it returns to re-optimize the target weight to form a closed-loop control. This ensures that high power quality and low operating loss can still be maintained when the load demand or mains input parameters change, thus adapting to the needs of dynamic scenarios.

[0075] Figure 4 A flowchart illustrating the third uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 4 As shown, in S150 above, when the operating loss of the 11th filter is greater than or equal to the preset threshold, the optimization target weights of the dual-objective optimization function are re-determined, including S151 to S152. S151 to S152 will be explained in detail below.

[0076] S151. When the operating loss of the 11th filter is greater than or equal to a preset threshold, determine the operating loss range to which the operating loss of the 11th filter belongs. Specifically, the first operating loss range is 1.2 to 1.5 times the rated loss of the 11th filter, the second operating loss range is 1.5 to 2 times the rated loss of the 11th filter, and the third operating loss range is more than 2 times the rated loss of the 11th filter.

[0077] Figure 5 The flowchart of the third uninterruptible power supply control method based on twelve-pulse rectification provided in the embodiments of this application is shown in the figure. In this implementation, when the operating loss of the 11th filter reaches or exceeds the preset threshold (i.e., 1.2 times the rated loss), the current loss value is first identified to determine the operating loss range to which it belongs.

[0078] It should be noted that the three ranges here correspond to 1.2-1.5 times, 1.5-2 times, and more than 2 times the rated loss, respectively. By dividing the ranges into quantitative intervals, the degree of loss overload can be accurately determined, providing a clear basis for the subsequent adjustment of the target weights and ensuring that the weight adjustment matches the loss status.

[0079] For example, if the rated loss of an 11th-order filter is 10W and the preset threshold is 12W, when the detected operating loss is 13W, it falls within the first operating loss range; if the loss is 16W, it falls within the second range; and if the loss reaches 21W, it enters the third range.

[0080] S152. When the operating loss range is within the first operating loss range, the weight of the first objective is determined to be 0.5~0.6, and the weight of the second objective is determined to be 0.4~0.5. When the operating loss range is within the second operating loss range, the weight of the first objective is determined to be 0.3~0.4, and the weight of the second objective is determined to be 0.6~0.7. When the operating loss range is within the third operating loss range, the weight of the first objective is determined to be 0.2~0.3, and the weight of the second objective is determined to be 0.7~0.8.

[0081] In this implementation, the weights of the two objectives in the dual-objective optimization function can be adjusted according to different operating loss ranges. When the loss is in the first range, the weight of the first objective (total harmonic distortion of input current less than 4.5%) remains at 0.5-0.6, and the weight of the second objective (minimizing filter loss) is 0.4-0.5, thus balancing power quality and loss control. When the loss enters the second range, the weight of the first objective is reduced to 0.3-0.4, and the weight of the second objective is increased to 0.6-0.7, focusing more on loss suppression. When the loss reaches the third range, the weight of the first objective is further reduced to 0.2-0.3, and the weight of the second objective is increased to 0.7-0.8, prioritizing ensuring that the filter is not overloaded.

[0082] For example, if the filter's rated loss is 10W, when the loss is 13W (first range), the first target weight is 0.55 and the second target weight is 0.45, which ensures the THD requirement while moderately reducing the loss; if the loss is 16W (second range), the first target weight is adjusted to 0.35 and the second target weight is 0.65, focusing more on reducing the loss; if the loss is 21W (third range), the first target weight is 0.25 and the second target weight is 0.75, prioritizing loss control to avoid damage.

[0083] With this implementation, when the operating loss of the eleventh-order filter is greater than or equal to a preset threshold, the operating loss range to which the loss belongs is first obtained, and then the weights of the first and second objectives are allocated according to different ranges, so that the weight adjustment is adapted to the loss situation, improving the pertinence and rationality of the dual-objective optimization; the loss range is determined first and then the weights are accurately assigned, and the rectifier bridge and filter are controlled after being solved by the dynamic programming algorithm, so that the loss control of the eleventh-order filter is more accurate and optimization deviation is avoided.

[0084] This implementation method solves the overload problem by dynamically adjusting the target priority under the hard constraint of the total harmonic distortion rate of the input current, avoiding the shortened life or damage of the filter due to long-term overload, without sacrificing the power quality of the power grid.

[0085] In addition to the control methods mentioned above, the solution frequency of the dynamic programming algorithm can be adjusted according to the dynamic changes in the load current to adapt to different load changes.

[0086] Figure 6 A flowchart illustrating the fourth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 6 As shown, in some implementations, the above method also includes S210 to S220, which will be described in detail below.

[0087] S210. Obtain the real-time load current parameters of the uninterruptible power supply. Calculate the load current change rate corresponding to the real-time load current parameters using the differential method.

[0088] Figure 7 A schematic diagram illustrating the workflow of the fourth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 7 As shown, the real-time load current parameters of the uninterruptible power supply can be obtained. These parameters are acquired by the current sensor inside the UPS and directly reflect the changes in the load's current demand.

[0089] In this implementation, the load current change rate can be calculated using the differential method. The differential method calculates the ratio of the difference in load current between adjacent moments to the time interval to obtain the change in current per unit time, which intuitively reflects the rate of change of load current.

[0090] It should be noted that the process of calculating the load current change rate using the differential method can be as follows: Select the load current values ​​I1 and I2 at two consecutive times t1 and t2, and calculate the time interval. Then use the formula The rate of change of load current, r, is calculated. For example, the current is 18A at time t1 and 23A at time t2. If the time is 0.2 seconds, then the rate of change .

[0091] S220. When the load current change rate is greater than or equal to the preset load current change rate, the solution frequency of the dynamic programming algorithm is set to the first solution frequency. When the load current change rate is less than the preset load current change rate, the solution frequency of the dynamic programming algorithm is set to the second solution frequency. The second solution frequency is less than the first solution frequency.

[0092] In this implementation, the solution frequency of the dynamic programming algorithm can be set according to the magnitude of the load current change rate. When the change rate is greater than or equal to a preset value, it is set to the first solution frequency to ensure a fast response to load changes. When the change rate is less than the preset value, it is set to the second solution frequency to reduce computational consumption, and the second solution frequency is less than the first solution frequency.

[0093] For example, the preset load current change rate can be set according to the application scenario of the UPS. For example, for motor loads in industrial scenarios, the preset value can be set to 30A / s; for server loads in data centers, the preset value can be set to 15A / s. When the change rate exceeds the preset value, it indicates that the load is in a state of rapid change.

[0094] For example, the first solution frequency can be set to 10Hz (10 solutions per second), and the second solution frequency can be set to 1Hz (1 solution per second). For instance, when the load current change rate is 35A / s (greater than the preset 30A / s), the dynamic programming algorithm solves at a frequency of 10Hz; when the change rate is 12A / s (less than 30A / s), it solves at a frequency of 1Hz.

[0095] This implementation obtains the real-time load current parameters of the uninterruptible power supply (UPS), calculates the load current change rate using the differential method, and sets different solution frequencies for the dynamic programming algorithm based on the magnitude of the change rate. A first solution frequency is used when the change rate is large, and a second solution frequency with a lower frequency is used when the change rate is small, adapting to load changes. Timely solution during large changes ensures precise control, while reducing the frequency during small changes decreases power consumption, improving control adaptability and efficiency. During sudden load changes, high-frequency solution can quickly obtain the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter, allowing for timely parameter adjustments. This ensures that even during sudden load changes, the total harmonic distortion (THD) of the input current can still be quickly controlled to meet the standard, guaranteeing power supply stability.

[0096] This implementation method obtains real-time load current parameters and calculates the rate of change using the differential method. The dynamic programming algorithm solution frequency is set according to the rate of change, with low rates of change corresponding to low frequencies. Solving at low frequencies reduces the computational load of the algorithm and reduces operational losses. While meeting control requirements, this reduces the running costs of the dynamic programming algorithm and improves the overall operating efficiency of the uninterruptible power supply.

[0097] In some implementations, the above method also includes S230 to S240, which will be described in detail below.

[0098] S230. Based on the rated current and thermal stability parameters of the twelve-pulse rectifier bridge, determine the upper limit constraint of the output current change rate of the twelve-pulse rectifier bridge.

[0099] In this implementation, the upper limit constraint of the output current change rate can be determined by combining the rated current and thermal stability parameters of the twelve-pulse rectifier bridge. The rated current is the maximum current value of the rectifier bridge for long-term stable operation. The thermal stability parameters reflect the heat dissipation capacity and temperature tolerance of the rectifier bridge when the current changes. The combination of the two can ensure that the upper limit constraint does not exceed the current carrying capacity of the rectifier bridge, nor does it affect the life of the device due to excessively high temperature caused by excessively rapid current changes.

[0100] It should be noted that the upper limit constraint of the output current change rate can be determined by an empirical value table. The empirical value table records the upper limit value corresponding to different combinations of rated current and thermal stability parameters in advance. For example, when the rated current of the rectifier bridge is 100A and the thermal stability parameter is at a certain level, the upper limit constraint corresponding to the empirical value table is 5A / ms.

[0101] For example, the upper limit constraint of the output current change rate can be set according to the specific equipment parameters. For instance, a certain model of twelve-pulse rectifier bridge has a rated current of 200A and a maximum allowable temperature rise rate of 1℃ / s for thermal stability parameters. Its upper limit constraint of the output current change rate can be set to 8A / ms.

[0102] For example, thermal stability parameters may include inherent parameters of the device such as rated junction temperature, thermal resistance, and thermal time constant.

[0103] S240. When the rate of change of current corresponding to the firing angle of the twelve-pulse rectifier bridge obtained by the dynamic programming algorithm exceeds the upper limit constraint of the rate of change of output current, increase the firing angle adjustment step size to reduce the rate of change of output current of the twelve-pulse rectifier bridge.

[0104] In this implementation, after the firing angle of the twelve-pulse rectifier bridge is obtained by the dynamic programming algorithm, it can be checked whether the current change rate corresponding to the firing angle exceeds the upper limit constraint of the output current change rate. If it exceeds or equals the upper limit constraint of the output current change rate, the adjustment step size of the firing angle can be increased to slow down the change rate of the firing angle, thereby reducing the change rate of the rectifier bridge output current and avoiding damage to the rectifier bridge caused by excessively rapid current change.

[0105] For example, if the original trigger angle adjustment step size is 0.1° / time, when the current change rate exceeds the upper limit constraint, the adjustment step size can be increased to 0.2° / time. This can reduce the current change rate from 10A / ms to below the upper limit constraint of 8A / ms more quickly, until the constraint requirement is met.

[0106] In this implementation, after the dynamic programming algorithm solves for the firing angle, it verifies the corresponding current change rate. When the change rate is greater than or equal to the upper limit constraint, the firing angle adjustment step size is increased to reduce the output current change rate. This improves the constraint system of the control model, avoids excessive current change rate leading to rectifier bridge overcurrent or overheating, and effectively protects the device. Based on the solution frequency mechanism of load current change rate, the increase of the adjustment step size can quickly bring the current change rate to the standard, avoid adjustment lag, and improve the accuracy and response efficiency of parameter adjustment.

[0107] Figure 8 A flowchart illustrating the fifth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 8 As shown, in some implementations, the above method also includes S310 to S340, which will be described in detail below.

[0108] S310. Obtain historical operating data of the uninterruptible power supply under different combinations of operating conditions within the data update period. These different combinations correspond to different load active power levels, load current harmonic content ranges, mains voltage RMS ranges, and mains frequency fluctuation ranges. Historical operating data includes the firing angle, output current, and rated current operating percentage of the twelve-pulse rectifier bridge, the switching status of the 11th order filter, copper loss data, iron loss data, and the measured value of the total harmonic distortion rate of the input current.

[0109] Figure 9A schematic diagram illustrating the workflow of the fifth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 9 As shown, in this implementation, historical operating data of the uninterruptible power supply can be collected within a set update period, covering multiple scenarios. These scenarios are composed of different levels of load active power (such as light load, medium load, heavy load), different ranges of load current harmonic content (such as low harmonic, medium harmonic, high harmonic), different ranges of mains voltage effective value (such as normal, low voltage, high voltage), and different fluctuation ranges of mains frequency (such as small fluctuation, large fluctuation), ensuring coverage of various operating conditions in actual operation.

[0110] It should be noted that the historical data should include the rectifier bridge's firing angle, output current and rated current ratio, filter switching status, copper loss and iron loss, and the measured value of the total harmonic distortion rate of the input current, so as to provide a basis for subsequent analysis.

[0111] It should be noted that different operating conditions can be classified according to the scenario, such as "light load (10%-30% of rated power), low harmonics (<2%), normal voltage (210V-230V), small frequency fluctuation (49.5Hz-50.5Hz)" and "heavy load (70%-100%), high harmonics (>5%), high voltage (230V-250V), large frequency fluctuation (49Hz-49.5Hz or 50.5Hz-51Hz)", which correspond to different historical operating data.

[0112] For example, under the operating conditions of a data center UPS with "medium load (50% of rated power), medium harmonics (2%-5%), normal voltage (220V), and small frequency fluctuations (50Hz±0.2Hz)," historical operating data may include: the trigger angle of the twelve-pulse rectifier bridge is 15°, the output current is 80A, and the rated current operation rate is 60%; the 11th filter is in operation, with copper loss of 150W and iron loss of 80W; and the measured value of the total harmonic distortion rate of the input current is 3.2%.

[0113] S320. Determine the target time period in historical operating data where the measured total harmonic distortion rate of the input current is greater than or equal to 4.5% or the operating loss of the 11th filter is greater than or equal to a preset loss. Obtain the pre-fluctuation data within a preset first time period before the target time period and the post-fluctuation data within a preset second time period after the target time period. Both the pre-fluctuation and post-fluctuation data include load parameters, mains parameters, harmonic data, and loss data.

[0114] In this implementation, time periods that meet the abnormal conditions can be selected from historical data. During these time periods, the total harmonic distortion rate of the input current is ≥4.5%, or the operating loss of the 11th filter is ≥ the preset loss (such as 1.2 times the rated loss), and these time periods are used as the target time periods.

[0115] In this implementation, pre-fluctuation data within a first time period (e.g., 10 minutes) before the target time period and post-fluctuation data within a second time period (e.g., 10 minutes) after the target time period can be obtained. Both pre-fluctuation and post-fluctuation data must include load parameters (active power, harmonic content), mains parameters (voltage, frequency), harmonic data (total harmonic distortion), and loss data (filter copper loss, iron loss) for analyzing parameter changes before and after the anomaly.

[0116] It should be noted that the target time period is a continuous or discontinuous period in historical data that meets the abnormal conditions. For example, if the rated loss of the filter is 100W and the preset loss is 120W, the target time period is defined as the loss exceeding 120W for 3 consecutive minutes or the total harmonic distortion exceeding 4.5% for 2 consecutive minutes.

[0117] For example, if a UPS has a total harmonic distortion (THD) of input current that is consistently 4.8% (≥4.5%) between 14:00 and 14:05 on May 10, 2024, then this period is the target period. The data before the fluctuation are: load active power 50%, mains voltage 220V, THD 3.0%, and filter loss 110W between 13:50 and 14:00; and the data after the fluctuation are: load active power 45%, mains voltage 218V, THD 3.5%, and filter loss 105W between 14:05 and 14:15.

[0118] S330. Based on the data before the fluctuation, the target time period data, and the data after the fluctuation, determine the variation curves of load active power, RMS mains voltage, total harmonic distortion rate of input current, and operating loss of the 11th filter. Based on the peak value, fluctuation amplitude, and time to reach the target under different operating conditions, determine the influence of each operating condition combination on harmonic suppression and loss control. Based on the influence patterns, determine the dynamic optimization parameters corresponding to each operating condition combination. Based on the load active power level and the mains voltage fluctuation range, determine the target operating condition combination and the corresponding dynamic optimization parameters.

[0119] In this implementation, data from before the fluctuation, the target time period, and after the fluctuation can be integrated to plot the change curves of load active power, mains voltage, total harmonic distortion (THD), and filter loss, visually displaying the changing trends of the parameters. Furthermore, the peak values ​​of the curves (such as the maximum value of THD), fluctuation amplitudes (such as the range of voltage changes), and compliance times (the time from abnormality to compliance) under different operating conditions can be analyzed to determine the influence of operating conditions on harmonic suppression (whether the THD meets the standard) and loss control (whether the filter loss exceeds the standard).

[0120] It should be noted that the influence pattern is the correlation between operating conditions and effects. For example, under heavy load conditions, the peak value of total harmonic distortion (THD) increases from 3.5% to 5.0%, and the time to meet the standard increases from 5 minutes to 10 minutes, indicating that heavy load will reduce the harmonic suppression effect; under high voltage conditions, the peak value of filter loss increases from 100W to 130W, indicating that high voltage will increase loss.

[0121] For example, under the operating conditions of "heavy load (90% rated power), high harmonics (5%), high voltage (240V), and large frequency fluctuation (50.6Hz)," the change curve shows that the total harmonic distortion rate increases from 3.0% to 5.2%, the filter loss increases from 100W to 140W, and the time to reach the standard is 12 minutes. From this, we can conclude that this combination of operating conditions will significantly reduce the harmonic suppression effect and at the same time greatly increase the filter loss.

[0122] In this implementation, dynamic optimization parameters can be set for each combined operating condition based on the influence law. The dynamic optimization parameters include the threshold for adjusting harmonic suppression (corresponding to the dynamic total harmonic distortion rate threshold) and the priority of loss control (corresponding to the loss optimization weight). Then, based on the current real-time load active power level (such as medium load) and the mains voltage fluctuation range (such as normal voltage), the corresponding target combined operating condition can be matched, and the dynamic optimization parameters under that operating condition can be obtained to ensure that the parameters are adapted to the current operating state.

[0123] It should be noted that the dynamic optimization parameters can be quantitatively adjusted according to patterns. For example, under the conditions of "heavy load, high harmonics, high voltage, and large fluctuations", the dynamic total harmonic distortion (THD) threshold is tightened from 4.5% to 4.0% (early warning of anomalies), and the loss optimization weight is increased from 0.4 to 0.6 (paying more attention to loss control); under the conditions of "light load, low harmonics, normal voltage, and small fluctuations", the dynamic THD threshold is relaxed to 4.8% (reducing unnecessary adjustments), and the loss optimization weight is reduced to 0.3 (paying more attention to harmonic suppression).

[0124] For example, under the operating condition of "medium load (50% rated power), medium harmonic distortion (3%), normal voltage (220V), and small frequency fluctuation (50Hz±0.2Hz)," the dynamic optimization parameters are set as follows: dynamic total harmonic distortion threshold of 4.5% and loss optimization weight of 0.5. When the current load active power is 50% and the mains voltage is 220V, the target combination operating condition is this condition, and the corresponding dynamic optimization parameters are the values ​​mentioned above.

[0125] S340. Update the dynamic optimization parameters to the bi-objective optimization function. Solve the updated bi-objective optimization function using a dynamic programming algorithm to determine the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter. The dynamic optimization parameters include the dynamic total harmonic distortion (THD) threshold and loss optimization weights.

[0126] In this implementation, the determined dynamic optimization parameters (including the dynamic total harmonic distortion rate threshold and loss optimization weight) can be updated into the bi-objective optimization function to adjust the target priority of harmonic suppression and loss control.

[0127] In this implementation, the updated function can be solved using a dynamic programming algorithm to obtain the optimal firing angle of the twelve-pulse rectifier bridge (controlling harmonic content) and the optimal switching state of the 11th filter (controlling loss), ensuring that the harmonic suppression requirements can be met and the filter loss can be effectively controlled under the current operating conditions.

[0128] It should be noted that the target combination of operating conditions is the historical operating condition matched with the current real-time parameters. For example, if the current load active power is 75% (heavy load) and the mains voltage is 235V (high voltage), then the target combination of operating conditions is "heavy load, medium harmonics, high voltage, large fluctuations", and the corresponding dynamic optimization parameters are: dynamic total harmonic distortion threshold of 4.2% and loss optimization weight of 0.6.

[0129] For example, the current load active power is 80% (heavy load), the mains voltage is 240V (high voltage), and the target combination of operating conditions is "heavy load, high harmonics, high voltage, large fluctuations". The dynamic optimization parameters are: dynamic total harmonic distortion threshold of 4.0%, and loss optimization weight of 0.6. After updating the dual-objective optimization function, the optimal firing angle of the twelve-pulse rectifier bridge is 12° (reducing the firing angle to reduce harmonics), and the optimal switching state of the 11th filter is in operation (shunting harmonics). At this time, the measured value of the total harmonic distortion of the input current is 3.8% (<4.0%), and the filter operating loss is 110W (<preset loss of 120W), which meets the dual-objective requirements.

[0130] This implementation method acquires historical operating data of different combinations of uninterruptible power supplies (UPS) operating conditions, identifies the target time period and extracts relevant data before and after it, and analyzes the change curves to determine the impact of each combination of operating conditions on harmonic suppression and loss control. Then, it determines the dynamic optimization parameters, updates them to the bi-objective optimization function, and solves the function. This makes harmonic suppression more closely match actual operating conditions, effectively reducing the probability of the total harmonic distortion rate of the input current exceeding the standard and improving the stability of harmonic suppression. Furthermore, it allows loss control to adapt to different operating conditions, precisely adjusting the loss optimization weights and reducing the occurrence of 11th order filter operating losses exceeding the preset threshold.

[0131] This implementation method determines the dynamic optimization parameters for each combination of operating conditions based on historical data. It adjusts the solution frequency and the trigger angle step size by combining the load current change rate with the current change rate constraint, making the optimization more targeted, reducing invalid solutions, improving optimization efficiency, and ensuring the operational stability of the twelve-pulse rectifier bridge to adapt to dynamic changes in operating conditions.

[0132] In some implementations, the above method further includes: determining the first duration of the first time period, the second duration of the second time period, and the data update duration based on the target duration of the target time period. The data update duration is a preset multiple of the sum of the first duration, the target duration, and the second duration.

[0133] In this implementation, three associated data time periods can be defined based on the specific duration of the target time period (i.e., the duration during which the total harmonic distortion rate of the input current is greater than or equal to 4.5% and the filter loss is greater than or equal to the preset loss): the first time period is used to collect "pre-fluctuation data" before the target time period, reflecting the stable state before the fluctuation of the operating condition; the second time period is used to collect "post-fluctuation data" after the target time period, reflecting the adjustment state after the fluctuation; the data update duration covers the complete process before, during and after the fluctuation, ensuring data continuity.

[0134] In this implementation, the data update duration needs to be set to a preset multiple of the sum of the first duration, the target duration, and the second duration. This ensures that enough relevant data is collected, covering the entire cycle of operating condition changes while avoiding deviations in pattern analysis due to insufficient data.

[0135] It should be noted that the first time period is the "historical baseline period" of the target time period, used to compare the parameter differences before and after the fluctuation; the second time period is the "recovery verification period", used to observe the recovery of the adjusted working conditions; and the data update duration is the "extension period" of the three, to ensure the integrity of the data chain.

[0136] For example, if the target duration of the target time period is 10 minutes (i.e., the harmonic exceedance lasts for 10 minutes), the first duration can be set to 5 minutes (to collect stable data for the 5 minutes before the fluctuation) and the second duration to 5 minutes (to collect recovery data for the 5 minutes after the fluctuation). The sum of the three is 20 minutes. If the preset multiplier is 2, the data update duration is 40 minutes, covering the complete cycle from 5 minutes before the fluctuation to 5 minutes after the fluctuation.

[0137] It should be noted that the preset multiplier setting needs to balance "data integrity" and "computational efficiency": the more frequent the operating conditions (such as data center UPS), the more the preset multiplier can be increased (such as 2~3) to ensure that enough fluctuation samples are collected; the more stable the operating conditions (such as small office area UPS), the more the preset multiplier can be decreased (such as 1.5) to avoid data redundancy.

[0138] This implementation method determines the first duration of the first time period, the second duration of the second time period, and the data update duration based on the target duration of the target time period. The data update duration is a preset multiple of the sum of the first three. Then, corresponding data is collected to determine the change curve and the influence law of each combination of operating conditions. Subsequently, a dual-objective optimization function for updating dynamic optimization parameters is determined, making data collection more targeted and complete, improving the accuracy of influence law, making dynamic optimization parameters more suitable for operating conditions, and enhancing the effect of harmonic suppression and loss control. It also ensures sufficient and correlated data, improves the reliability of law analysis, makes dynamic optimization parameters more accurate, and optimizes control effect better.

[0139] This implementation method determines the target time period, then determines the first duration, the second duration, and the data update duration according to the target duration. Data from each time period is collected to form a complete data chain. The influence patterns of each combination of working conditions are clarified through the change curves, making the analysis of influencing factors more comprehensive, accurately locating the reasons for non-compliance, making the dynamic optimization parameter setting more targeted, and reducing the probability of exceeding the standard.

[0140] Figure 10 A flowchart illustrating the sixth uninterruptible power supply control method based on twelve-pulse rectification provided in this application embodiment is shown below. Figure 10 As shown, in some implementations, in the above-mentioned S330, the influence law of each combination of operating conditions on the harmonic suppression effect and loss control effect is determined according to the peak value, fluctuation amplitude and the time to reach the target under different combination of operating conditions, including S331 to S332. S331 to S332 will be explained in detail below.

[0141] S331. Based on the harmonic distortion rate curve and loss variation curve corresponding to each combined operating condition, determine the curve characteristics corresponding to each combined operating condition. Among them, the curve characteristics include the curve peak value, the curve fluctuation amplitude, and the curve reaching the target time.

[0142] In this implementation, for each combined operating condition, the curves of harmonic distortion rate versus time and loss versus time can be obtained first. Then, key features that reflect the harmonic suppression effect and loss control effect under the operating condition can be extracted from these two curves. These features are the basis for subsequent analysis of the impact of the operating condition on the control effect.

[0143] In this implementation, the curve features include three dimensions: the curve peak value is the maximum value of the curve within a set time period, reflecting the most severe level of harmonics or losses under this operating condition; the curve fluctuation amplitude is the difference between the maximum and minimum values ​​of the curve, reflecting the degree of fluctuation of harmonics or losses; and the curve compliance time is the length of time that the curve value meets the preset threshold requirements, reflecting the continuous stability of the effect.

[0144] For example, for a harmonic distortion rate curve, the peak value is the maximum harmonic distortion rate within the target time period; the fluctuation amplitude is the maximum value minus the minimum value within that time period; and the time for the curve to meet the target is the total time during which the harmonic distortion rate is less than 4.5%. For a loss variation curve, the peak value is the maximum loss value, the fluctuation amplitude is the difference between the maximum and minimum values, and the time for the curve to meet the target is the total time during which the loss is less than the preset loss.

[0145] S332. Through correlation mining algorithms, determine the quantitative correlation between combined working conditions and curve features.

[0146] In this implementation, a correlation mining algorithm can be used to analyze the numerical relationship between parameters (such as load active power level and mains voltage effective value range) and curve characteristics (such as curve peak value and fluctuation amplitude) of combined operating conditions, and obtain a quantitative correlation degree. For example, for every 10% increase in the load power level, the corresponding harmonic distortion rate curve peak value increases by 15%. In this way, the specific impact of different operating conditions on harmonic suppression and loss control can be clearly defined.

[0147] For example, correlation mining algorithms may include the Pearson correlation coefficient method and the decision tree algorithm.

[0148] This implementation method first determines the curve peak values ​​and other characteristics of each combined operating condition. Then, it quantifies the association between the combined operating conditions and the characteristics through a correlation mining algorithm. Finally, it adjusts the total harmonic distortion rate threshold and dual-objective weight of the input current by a fixed ratio according to the conditions satisfied by the curve characteristics. This provides a quantitative basis for parameter adjustment, avoids blind setting, improves the scientific nature of dynamic optimization parameters, and enhances the overall control stability.

[0149] In some implementations, in the above-mentioned S330, the dynamic optimization parameters corresponding to each combination of operating conditions are determined according to the influence law, including: when the peak value of the harmonic distortion rate curve is greater than or equal to the preset peak value, and the time for the loss change curve to meet the standard is less than the preset time for the curve to meet the standard, the total harmonic distortion rate threshold of the input current is multiplied by a first proportional value; wherein, the first proportional value is less than 1.

[0150] In this implementation, dynamic optimization parameters are set for different operating conditions based on the previously obtained influence patterns. When the peak value of the harmonic distortion rate curve of a certain operating condition exceeds the peak value of the preset curve, and the time for the loss change curve to meet the standard is less than the time for the preset curve to meet the standard, it indicates that the harmonics are prone to exceed the standard and the loss control is unstable under this operating condition. At this time, the total harmonic distortion rate threshold of the input current can be multiplied by a first proportional value less than 1 to reduce the threshold and control the harmonics in advance to avoid exceeding the standard.

[0151] For example, if the preset curve peak value is 4% and the preset curve compliance time is 70% of the total time, when the peak value of the harmonic distortion rate curve of a certain combination of working conditions reaches 4.2% and the loss compliance time only accounts for 60% of the total time, the total harmonic distortion rate threshold of the input current is adjusted from 4.5% by multiplying by 0.8 to 3.6% to more strictly control harmonics.

[0152] With this implementation, when the peak value of the harmonic distortion rate curve exceeds the preset value and the loss meets the standard for an insufficient period of time, the total harmonic distortion rate threshold of the input current is multiplied by the first proportional value to reduce the threshold in a targeted manner, accurately locate the risk conditions of harmonic exceedance, and effectively improve the accuracy and effectiveness of harmonic suppression.

[0153] In some implementations, in S330 above, based on the influence law, the dynamic optimization parameters corresponding to each combined operating condition are determined, including: when the peak value of the harmonic distortion rate curve is greater than or equal to the preset peak value, and the time for the loss change curve to reach the target is less than the preset target time, the total harmonic distortion rate threshold of the input current is multiplied by a first proportional value. When the fluctuation amplitude of the loss change curve is less than the preset fluctuation amplitude, and the time for the curve to reach the target is greater than or equal to the preset target time, the weight of the first target is multiplied by a second proportional value, and the weight of the second target is multiplied by a third proportional value. The second proportional value is greater than 1, and the third proportional value is less than 1.

[0154] In this implementation, when the fluctuation amplitude of the loss change curve of a certain combined operating condition is less than the fluctuation amplitude of the preset curve, and the time to achieve the target reaches or exceeds the time to achieve the target of the preset curve, it indicates that the loss control is stable and effective under this operating condition. At this time, the weight of the first objective (total harmonic distortion rate of input current less than 4.5%) can be increased by multiplying it by a second proportional value greater than 1, while the weight of the second objective (minimizing the operating loss of the 11th filter) can be decreased by multiplying it by a third proportional value less than 1, so as to achieve a dynamic balance of dual-objective optimization.

[0155] For example, if the preset curve fluctuation amplitude is 5% of the loss value and the preset curve compliance time is 80% of the total time, when the loss fluctuation amplitude of a certain combination of working conditions is only 3% and the compliance time accounts for 90% of the total time, the first target weight is multiplied by 1.2 and the second target weight is multiplied by 0.7, so that the harmonic suppression effect is prioritized under the condition of stable loss.

[0156] In this implementation, the first proportional value is less than 1, such as 0.8-0.9; the second proportional value is greater than 1, such as 1.1-1.2; and the third proportional value is less than 1, such as 0.7-0.8. These proportional values ​​are set based on the quantitative analysis of the working condition characteristics to ensure the scientific nature and effectiveness of the parameter adjustment.

[0157] This implementation method establishes a quantitative correlation between combined operating conditions and curve characteristics through a correlation mining algorithm. When the fluctuation amplitude of the loss change curve is small and the target duration is long, the first target weight is multiplied by a second proportional value, and the second target weight is multiplied by a third proportional value. This method enables loss control to adapt to the characteristics of the operating conditions and prioritizes harmonic suppression when the loss is stable, thereby achieving a dynamic balance of dual-objective optimization.

[0158] In some implementations, the above method further includes: determining the product of the data update duration and the proportion of the target achievement duration as the preset curve target achievement duration. Specifically, the proportion of the target achievement duration is determined by the curve peak value and the curve fluctuation amplitude.

[0159] In this implementation, the preset curve compliance time can be calculated by combining the update data duration used for analyzing historical operation (i.e., the time span of collecting historical operation data) and the compliance time ratio (reflecting the degree of requirement of the curve characteristics for compliance time). This calculation method ensures that the preset value is fully associated with the time range of historical data and the actual needs of the curve characteristics, avoiding the problem of fixed preset values ​​being out of touch with actual operation, and improving the rationality of the preset curve compliance time.

[0160] For example, if the update time of a certain combination of operating conditions is 72 hours (i.e., analyzing the historical operating data of the past 3 days), and the proportion of the standard compliance time determined by the curve characteristics is 0.5, then the preset standard compliance time of the curve is 72 × 0.5 = 36 hours, which means that the harmonic suppression or loss control under this operating condition must reach the qualified standard within 36 hours.

[0161] In this implementation, the compliance time ratio can be determined comprehensively by the peak value of the curve (such as the maximum value of the total harmonic distortion rate of the input current and the maximum value of the operating loss of the 11th filter) and the fluctuation amplitude (such as the difference between the maximum and minimum values ​​of the curve within the data update time). The higher the peak value of the curve, the greater the risk of harmonic exceedance or loss overload, and the smaller the compliance time ratio can be set to require the system to adjust to the compliance state more quickly; the larger the fluctuation amplitude of the curve, the more drastic the parameter changes, and the compliance time ratio can also be reduced accordingly to ensure that the compliance time can adapt to the fluctuation situation.

[0162] It should be noted that the percentage of time required to meet the standard can be quickly determined using an empirical value table. For example, an empirical value table can be created by pre-calculating the percentage values ​​corresponding to different combinations of peak values ​​and fluctuation amplitudes of the curve: when the peak value of the harmonic distortion rate curve is 4.5% (close to the 4.5% threshold) and the fluctuation amplitude is 0.5% (small fluctuation), the percentage of time required to meet the standard is set to 0.6; when the peak value is 5% (exceeding the threshold) and the fluctuation amplitude is 0.8% (large fluctuation), the percentage is set to 0.4; and when the peak value is 4% (below the threshold) and the fluctuation amplitude is 0.3% (very small fluctuation), the percentage is set to 0.7. In practical applications, the corresponding percentage of time required to meet the standard can be obtained simply by looking up the table based on the peak value and fluctuation amplitude of the current curve.

[0163] This implementation method ensures that the preset curve's target duration is fully correlated with data features, avoiding deviations caused by fixed values, improving the matching degree between preset values ​​and actual working conditions, and providing accurate basis for subsequent analysis; it also improves the accuracy of curve feature extraction, thereby enhancing the reliability of the quantitative correlation between combined working conditions and curve features, making the influence patterns more consistent with reality.

[0164] This application also provides an uninterruptible power supply control system based on twelve-pulse rectification, including a unit for implementing the above-described uninterruptible power supply control method based on twelve-pulse rectification.

[0165] Figure 11 A schematic diagram of the logic structure of an uninterruptible power supply control system based on twelve-pulse rectification, provided for an embodiment of this application, is shown below. Figure 11 As shown, the system 1 of this embodiment includes a processing unit 11, a storage unit 12, and a transceiver unit 13. The processing unit 11 is used to process data, the storage unit 12 is used to store data, and the transceiver unit 13 is used to send and receive data. The processing unit 11, the storage unit 12, and the transceiver unit 13 cooperate with each other to implement the above-described method. The beneficial effects of the embodiments of this application have been described in the above-described method and will not be repeated here.

[0166] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0167] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0168] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0169] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0170] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0171] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, 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.

[0172] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0173] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A control method for an uninterruptible power supply based on twelve-pulse rectification, characterized in that, The method includes: Obtain the real-time load demand parameters and real-time mains input parameters of the uninterruptible power supply. The real-time load demand parameters include the load active power requirement and the load current harmonic content requirement. The real-time mains input parameters include the mains voltage RMS value and the mains frequency. A rectifier-filter coordinated control model is constructed, and a dual-objective optimization function and constraints are determined. The rectifier-filter coordinated control model includes the output current characteristics of the twelve-pulse rectifier bridge, the impedance characteristics of the 11th-order filter, and their interaction characteristics. The interaction characteristics characterize the interaction between the output current characteristics of the twelve-pulse rectifier bridge and the impedance characteristics of the 11th-order filter. The first objective of the dual-objective optimization function is to ensure that the total harmonic distortion (THD) of the uninterruptible power supply's input current is less than 4.5%, and the second objective is to minimize the operating losses of the 11th-order filter, including copper losses and iron losses. The constraints include that the maximum output current of the twelve-pulse rectifier bridge is less than its rated current, and the switching state of the 11th-order filter is less than its design capacity. The optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter are obtained by solving the bi-objective optimization function using a dynamic programming algorithm. Based on the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter, the firing circuit parameters of the twelve-pulse rectifier bridge and the switching action of the 11th filter are controlled. The optimal firing angle is used to control the harmonic content of the rectifier bridge output current, and the optimal switching state includes operation or shutdown to control the operating loss of the filter.

2. The method according to claim 1, characterized in that, The method further includes: Obtain the real-time input current total harmonic distortion rate and 11th filter loss value of the uninterruptible power supply; When the total harmonic distortion rate of the uninterruptible power supply input current is greater than or equal to 4.5%, the collaborative control model of the twelve-pulse rectifier bridge and the 11th filter is reconstructed and solved again; when the operating loss of the 11th filter is greater than or equal to the preset loss, the optimization objective weight of the bi-objective optimization function is redefined; whereby the preset loss is 1.2 times the rated loss of the 11th filter.

3. The method according to claim 2, characterized in that, When the operating loss of the 11th filter is greater than or equal to a preset threshold, the optimization objective weights of the bi-objective optimization function are redefined, including: When the operating loss of the 11th filter is greater than or equal to a preset threshold, the operating loss range of the 11th filter is determined; wherein, the first operating loss range is 1.2 to 1.5 times the rated loss of the 11th filter, the second operating loss range is 1.5 to 2 times the rated loss of the 11th filter, and the third operating loss range is more than 2 times the rated loss of the 11th filter. When the operating loss range is within the first operating loss range, the weight of the first objective is determined to be 0.5~0.6 and the weight of the second objective is determined to be 0.4~0.5; when the operating loss range is within the second operating loss range, the weight of the first objective is determined to be 0.3~0.4 and the weight of the second objective is determined to be 0.6~0.7; when the operating loss range is within the third operating loss range, the weight of the first objective is determined to be 0.2~0.3 and the weight of the second objective is determined to be 0.7~0.

8.

4. The method according to claim 3, characterized in that, The method further includes: Obtain the real-time load current parameters of the uninterruptible power supply; calculate the load current change rate corresponding to the real-time load current parameters using the differential method; When the load current change rate is greater than or equal to the preset load current change rate, the solution frequency of the dynamic programming algorithm is set to the first solution frequency; when the load current change rate is less than the preset load current change rate, the solution frequency of the dynamic programming algorithm is set to the second solution frequency; wherein, the second solution frequency is less than the first solution frequency.

5. The method according to claim 4, characterized in that, The method further includes: Based on the rated current and thermal stability parameters of the twelve-pulse rectifier bridge, determine the upper limit constraint of the output current change rate of the twelve-pulse rectifier bridge. When the rate of change of current corresponding to the firing angle of the twelve-pulse rectifier bridge obtained by the dynamic programming algorithm exceeds the upper limit constraint of the rate of change of output current, the firing angle adjustment step size is increased to reduce the rate of change of output current of the twelve-pulse rectifier bridge.

6. The method according to claim 5, characterized in that, The method further includes: Acquire historical operating data of the uninterruptible power supply under different combinations of operating conditions within the data update period; among which, different combinations of operating conditions correspond to different load active power levels, load current harmonic content ranges, mains voltage effective value ranges, and mains frequency fluctuation ranges; historical operating data includes the firing angle, output current, and rated current operating ratio of the twelve-pulse rectifier bridge, the switching status of the 11th filter, copper loss data, iron loss data, and measured value of total harmonic distortion rate of input current; Determine the target time period in historical operating data where the measured total harmonic distortion rate of the input current is greater than or equal to 4.5% or the operating loss of the 11th filter is greater than or equal to the preset loss; acquire the pre-fluctuation data in the preset first time period before the target time period and the post-fluctuation data in the preset second time period after the target time period; wherein, the pre-fluctuation data and the post-fluctuation data both include load parameters, mains parameters, harmonic data and loss data; Based on data before the fluctuation, data for the target time period, and data after the fluctuation, determine the variation curves of load active power, RMS mains voltage, total harmonic distortion rate of input current, and operating loss of the 11th filter; based on the peak value, fluctuation amplitude, and time to reach the target under different combined operating conditions, determine the influence law of each combined operating condition on harmonic suppression and loss control effects; based on the influence law, determine the dynamic optimization parameters corresponding to each combined operating condition; based on the load active power level and mains voltage fluctuation range, determine the target combined operating condition and the corresponding dynamic optimization parameters. The dynamic optimization parameters are updated to the bi-objective optimization function; the updated bi-objective optimization function is solved by dynamic programming algorithm to determine the optimal firing angle of the twelve-pulse rectifier bridge and the optimal switching state of the 11th filter; among them, the dynamic optimization parameters include the dynamic total harmonic distortion rate threshold and the loss optimization weight.

7. The method according to claim 6, characterized in that, The method further includes: Based on the target duration of the target time period, determine the first duration of the first time period, the second duration of the second time period, and the data update duration; wherein, the data update duration is a preset multiple of the sum of the first duration, the target duration, and the second duration.

8. The method according to claim 7, characterized in that, Based on the peak value, fluctuation amplitude, and time to reach the target value of the variation curves under different operating conditions, the influence of each operating condition combination on the harmonic suppression and loss control effects is determined, including: Based on the harmonic distortion rate curve and loss variation curve corresponding to each combined operating condition, the curve characteristics corresponding to each combined operating condition are determined; among which, the curve characteristics include the curve peak value, the curve fluctuation amplitude, and the curve reaching the target time. The quantitative correlation between combined working conditions and curve features is determined by correlation mining algorithms; Based on the influence patterns, the dynamic optimization parameters corresponding to each combination of operating conditions are determined, including: When the peak value of the harmonic distortion rate curve is greater than or equal to the preset peak value, and the time for the loss change curve to meet the standard is less than the preset time for the curve to meet the standard, the first proportional value is multiplied by the total harmonic distortion rate threshold of the input current. When the fluctuation amplitude of the curve corresponding to the loss change curve is less than the preset curve fluctuation amplitude, and the curve reaches the target duration is greater than or equal to the preset curve reaches the target duration, the weight of the first target is multiplied by the second proportion value, and the weight of the second target is multiplied by the third proportion value; wherein, the first proportion value is less than 1, the second proportion value is greater than 1, and the third proportion value is less than 1.

9. The method according to claim 8, characterized in that, The method further includes: The product of the data update duration and the proportion of the target duration is determined as the preset target duration for the curve; the proportion of the target duration is determined by the curve peak value and the curve fluctuation amplitude.

10. A control system for an uninterruptible power supply based on twelve-pulse rectification, characterized in that, Includes units for implementing the method of any one of claims 1 to 9.

Images