An AC / DC dummy load testing method capable of stepless generation of reactive power

Abstract

The application provides a kind of AC / DC dummy load test method of stepless reactive power generation, it is related to electric power test technical field.The method responds to test instruction and reconfigures internal circuit as AC or DC load topology, captures the voltage and current instantaneous value of the output end of the uninterrupted power supply under test, calculates the target value based on instantaneous reactive power theory and generates current instruction.In AC mode, thyristor and insulated gate bipolar transistor are cooperatively controlled, the fundamental phase difference is established, and the amplitude deviation and harmonic are corrected.In DC mode, current reference is generated through proportional-integral adjustment, and adjustable DC output is realized through pulse width modulation.The heat energy management unit automatically adjusts the heat dissipation intensity according to the temperature, realizes stepless simulation of reactive power and continuous adjustment of load, and improves the test accuracy and stability of uninterrupted power supply.

Description

Technical Field

[0001] This application relates to the field of power testing technology, and in particular to an AC / DC dummy load testing method that can steplessly generate reactive power. Background Technology

[0002] Data centers operate a large number of servers and network devices, which require a stable and reliable power supply. Therefore, data centers are equipped with critical power supply equipment such as uninterruptible power supplies (UPS), high-voltage DC power supplies, and inverter drives for server cooling fans and air conditioning systems. These devices must undergo rigorous testing before being put into use or during regular maintenance to ensure proper operation. The AC / DC dummy load testing method, which can steplessly generate reactive power, is a key technology for testing the performance of power equipment such as UPS.

[0003] Currently, existing AC / DC dummy load testing solutions commonly employ an RLC impedance load cell composed of resistors combined with simple power device control, simulating AC / DC loads by adjusting impedance parameters. However, this solution has the following shortcomings: First, the load cell is large and its power density is insufficient, making it difficult to match the increasingly powerful intelligent computing server racks. Second, it can only roughly adjust the power, failing to accurately simulate the complex real-time power factor changes of a real server during operation, especially since server power supplies generate dynamic reactive power and harmonics, and existing solutions cannot accurately match the reactive power requirements under different operating conditions. Third, modern data center power supply systems often include both AC and DC power distribution, while traditional load cells typically only simulate a single type of load, requiring multiple different types of testing equipment, resulting in cumbersome operation and low testing efficiency. Fourth, the cooling method is singular; traditional air cooling is inadequate for high power density testing, easily leading to overheating during prolonged testing, resulting in insufficient test stability and significant data deviation.

[0004] Therefore, existing technologies suffer from technical problems such as poor accuracy and stability of AC / DC dummy load testing, inability to simulate reactive loads, separate AC / DC testing, and insufficient heat dissipation capacity. Summary of the Invention

[0005] The purpose of this application is to provide an AC / DC dummy load testing method that can steplessly generate reactive power, so as to solve the problems of poor accuracy and stability of AC / DC dummy load testing in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an AC / DC dummy load testing method capable of steplessly generating reactive power, comprising: In response to test commands, the internal circuitry is reconfigured into an AC load topology or a DC load topology, and the instantaneous voltage and current values ​​at the output of the uninterruptible power supply under test are captured. The digital signal processor calculates the target reactive power value of the IT load to be simulated based on the instantaneous voltage and current values ​​using instantaneous reactive power theory, and generates a current command signal representing the desired current vector based on the target reactive power value. When the internal circuit is reconfigured into an AC load topology, the thyristor and the insulated gate bipolar transistor connected in series with the thyristor are controlled in coordination according to the current command signal to generate an AC current that flows through the energy consumption unit and conforms to the current command signal, and the AC current is applied to the AC output terminal of the uninterruptible power supply under test. When the internal circuit is reconfigured into a DC load topology, a DC current reference is generated based on the load voltage reference value and the DC output voltage value of the uninterruptible power supply under test. The insulated gate bipolar transistor is controlled by a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit, thereby forming an adjustable DC load. The heat energy converted from the current flowing through the energy consumption unit is simultaneously exported through the air-cooled and liquid-cooled heat dissipation structures, and the heat dissipation intensity of the heat dissipation system is automatically adjusted according to the real-time monitored temperature.

[0007] Optionally, the step of coordinating the control of the thyristor and the insulated-gate bipolar transistor connected in series with the thyristor according to the current command signal to generate an alternating current flowing through the energy-consuming unit and conforming to the current command signal specifically includes: Based on the reactive power target value in the current command signal, the conduction phase of the thyristor is controlled to establish a fundamental phase difference corresponding to the reactive power target value between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit, thereby generating a preliminary alternating current containing the fundamental phase difference flowing through the energy consumption unit. While maintaining the fundamental phase difference without directional reversal, the amplitude deviation and harmonic components of the initial AC current are corrected using the insulated gate bipolar transistor connected in series with the thyristor, with the current command signal as a reference, to obtain the AC current that conforms to the current command signal.

[0008] Optionally, the step of generating a DC current reference based on the load voltage reference value and the DC voltage value output by the uninterruptible power supply under test, and controlling the insulated gate bipolar transistor through a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit, specifically includes: The load voltage reference value is set according to the rated output DC voltage of the uninterruptible power supply under test carried in the test command. The load voltage reference value can be adjusted in real time by the test command during the test. The DC output voltage value of the uninterruptible power supply under test is compared with the load voltage reference value, and the difference is processed by a proportional-integral regulator to generate the DC current reference. The pulse width modulation signal is generated based on the DC current reference, and the insulated gate bipolar transistor is driven to operate according to the duty cycle of the pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit.

[0009] Optionally, the step of controlling the conduction phase of the thyristor according to the reactive power target value in the current command signal, and establishing a fundamental phase difference corresponding to the reactive power target value between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit, further includes: Based on the reactive power target value in the current command signal, the required phase-shifting firing angle within each AC half-wave is calculated, and a firing angle command is generated. The zero-crossing point of the output voltage of the uninterruptible power supply under test is captured in real time, and a synchronization pulse signal is generated. The phase-shift triggering algorithm calculates the delay time from the zero-crossing point to the trigger point within each AC half-wave based on the trigger angle command and the synchronization pulse signal, and generates a trigger pulse sequence. The trigger pulse sequence is sent to the phase-shift trigger drive circuit, which uses an optocoupler to achieve strong and weak current isolation. When the trigger pulse arrives, it drives the thyristor to conduct at a non-zero crossing point, forming a continuously variable fundamental phase difference between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit.

[0010] Optionally, the step of using the insulated-gate bipolar transistor connected in series with the thyristor to correct the amplitude deviation and harmonic components of the preliminary alternating current based on the current command signal, and obtaining the alternating current conforming to the current command signal, further includes: The current command signal in the stationary coordinate system is converted to the rotating coordinate system and compared with the feedback value of the converted initial AC current to obtain the current deviation value. The current deviation value is processed by a proportional resonant regulator to generate a compensation voltage vector; A first pulse width modulation signal is synthesized based on the compensation voltage vector using a space vector pulse width modulation algorithm or a sinusoidal pulse width modulation algorithm, and sent to the insulated gate bipolar transistor to drive it to correct the initial alternating current.

[0011] Optionally, it also includes: Based on the mathematical model of the electronic load, the current command signal at the current moment and the instantaneous value of the AC current collected in real time are substituted into the mathematical model to calculate the duty cycle required for the next switching cycle. A second pulse width modulation signal is generated based on the calculated duty cycle and sent to the insulated gate bipolar transistor instead of the first pulse width modulation signal, so that the actual alternating current at the next moment tracks the current command signal within one switching cycle.

[0012] Optionally, when in AC output mode and the test command requires simulating a pure active load, and the thermal time constant of the energy consumption unit is greater than 10 times the zero-crossing trigger control cycle, such that the temperature fluctuation amplitude of the energy consumption unit under intermittent cycle conduction does not exceed a preset temperature threshold, the method of coordinating the control of the thyristor and the insulated gate bipolar transistor is replaced by zero-crossing trigger control of the thyristor, specifically including: Based on the active power target value in the test command, the ratio of the number of complete AC cycles to the number of cycles to be turned on to the number of cycles to be turned off in each zero-crossing trigger control cycle is calculated, and the on / off ratio command is generated. A switching control sequence is generated based on the switching ratio command. The switching control sequence is used to determine the on or off state of each complete AC cycle within the zero-crossing trigger control cycle. The on / off control sequence is sent to the zero-crossing trigger drive circuit. When the on / off control sequence indicates on and the AC voltage zero-crossing point is detected, the thyristor is driven to turn on at the voltage zero-crossing point. When the on / off control sequence indicates off, the thyristor is driven to turn off naturally at the next voltage zero-crossing point, so that the average active power flowing through the energy consumption unit is continuously adjustable.

[0013] Optionally, the zero-crossing trigger drive circuit uses an optocoupler with zero-crossing detection function. When the on / off control sequence indicates on and the optocoupler detects the AC voltage zero-crossing point, the bidirectional thyristor in the optocoupler is turned on, so that the thyristor is turned on at the voltage zero-crossing point.

[0014] Optionally, when in AC output mode and the test command requires simulating reactive load, and the test command carries a low harmonic requirement flag, the control of the thyristor in the coordinated control adopts a hybrid control method of phase-shift triggering and zero-crossing triggering, specifically including: Based on the reactive power target value and active power target value in the current command signal, the trigger angle command required for phase-shift triggering and the on / off ratio command required for zero-crossing triggering are calculated respectively, and hybrid control parameters are generated. The hybrid control parameters define the duration ratio of the phase-shift triggering working period to the zero-crossing triggering working period within a hybrid control cycle. The duration ratio is determined according to the ratio of the reactive power target value to the active power target value. The higher the proportion of the reactive power target value, the higher the proportion of the phase-shift triggering working period. The zero-crossing point of the output voltage of the uninterruptible power supply under test is captured in real time, and a synchronization pulse signal is generated. Based on the hybrid control parameters and the synchronization pulse signal, each hybrid control cycle is divided into a first time period and a second time period. During the first time period, the delay time is calculated according to the trigger angle command and the synchronization pulse signal and a trigger pulse sequence is generated to drive the thyristor to conduct at a non-zero crossing point, thereby forming a fundamental phase difference between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit. During the second time period, an on / off control sequence is generated according to the on / off ratio command, driving the thyristor to turn on or turn off naturally at the voltage zero crossing point, so that the average active power flowing through the energy consumption unit is continuously adjustable. By adjusting the ratio of the first time period to the second time period, and ensuring that the switching between the first time period and the second time period occurs at the voltage zero-crossing point, the AC current after mixed triggering is applied to the AC output terminal of the uninterruptible power supply under test.

[0015] Optionally, after the hybrid triggering, the process further includes a two-stage correction of the initial alternating current, wherein the two-stage correction is executed sequentially, specifically including: The current command signal and the initial AC current feedback value that actually flows through the energy consumption unit after hybrid triggering are received by the space vector modulation algorithm. A compensation voltage vector is generated, and a first pulse width modulation signal is synthesized according to the compensation voltage vector and sent to the insulated gate bipolar transistor connected in series with the thyristor to perform the first stage correction on the initial AC current. When the AC current deviation corrected in the first stage enters the preset steady-state error band, the deadbeat current prediction control algorithm receives the current command signal and the instantaneous value of the AC current collected in real time, calculates the duty cycle required for the next switching cycle based on the mathematical model of the electronic load, generates a second pulse width modulation signal to replace the first pulse width modulation signal and sends it to the insulated gate bipolar transistor to complete the second stage correction, and stops generating the first pulse width modulation signal during the second stage correction.

[0016] The AC / DC dummy load testing method for stepless reactive power generation provided in this application has the following advantages: By reconstructing the internal circuit into an AC or DC load topology in response to test commands, it can simultaneously adapt to both AC and DC test conditions on a single platform, eliminating the need for multiple different types of test equipment and solving the problems of separate AC and DC tests and cumbersome operation in existing technologies; By using a digital signal processor to directly calculate the reactive power target value from the instantaneous values ​​of voltage and current based on instantaneous reactive power theory and generate a current command signal, it provides a precise control reference for the subsequent coordinated control of thyristors and insulated-gate bipolar transistors; Through the coordinated operation of thyristor conduction phase control and series correction of the insulated-gate bipolar transistor in AC mode, the thyristor is responsible for establishing the fundamental phase difference to achieve reactive power generation. In the initial simulation of the load, the insulated-gate bipolar transistor (IGBT) corrects the amplitude deviation and harmonic components of the initial AC current while maintaining the fundamental phase difference without directional reversal. This enables stepless continuous adjustment of reactive power and accurately reproduces the complex operating conditions of real-time power factor changes in IT equipment. In DC mode, the difference between the actual DC voltage value and the load voltage reference value is processed by a proportional-integral regulator to generate a DC current reference. The IGBT is then controlled by a pulse-width modulation signal, enabling continuous adjustment of the DC load to adapt to different DC testing requirements. By simultaneously dissipating heat through air cooling and liquid cooling structures and automatically adjusting the heat dissipation intensity according to the real-time temperature, the heat dissipation requirements under high power density testing can be effectively met, avoiding overheating that could affect test stability and equipment lifespan.

[0017] Furthermore, when in AC output mode and the test command requires simulating a pure active load and the load thermal inertia meets the preset conditions, a zero-crossing trigger control method is adopted. The thyristor turns on and off at the zero-crossing point of the voltage, and the output current waveform maintains a complete sine wave without generating harmonic pollution. By adjusting the proportion of the conduction cycle, the active power can be continuously adjusted to adapt to pure active load scenarios, further improving the relevance and accuracy of the test.

[0018] Furthermore, when continuous reactive power regulation is required in a low-harmonic environment, a hybrid control method combining phase-shift triggering and zero-crossing triggering is adopted. Reactive power regulation is achieved through phase-shift triggering in the first time period of the hybrid control cycle, and low-harmonic active power regulation is achieved through zero-crossing triggering in the second time period. The combination of the two methods enables the dummy load to achieve stepless generation of reactive power while maintaining a low level of harmonic pollution, breaking the limitations of a single triggering method and adapting to complex test scenarios. Attached Figure Description

[0019] Figure 1 A schematic diagram of the overall process of an AC / DC dummy load test method that can steplessly generate reactive power, provided for an embodiment of this application; Figure 2A schematic diagram of the AC load topology circuit provided in the embodiments of this application; Figure 3 A schematic diagram of a DC load topology circuit provided in an embodiment of this application; Figure 4 This is a schematic diagram of a relay matrix switching circuit provided in an embodiment of this application; Figure 5 A schematic diagram of the structure of coordinated control of thyristor phase-shift triggering and insulated-gate bipolar transistor correction in AC mode provided in an embodiment of this application; Figure 6 This is a schematic diagram of a key waveform for phase-shift triggering provided in an embodiment of this application; Figure 7 This is a schematic diagram of the hysteresis current control in DC mode provided in an embodiment of this application; Figure 8 This is a schematic diagram of the hysteresis current control waveform provided in an embodiment of this application. Detailed Implementation

[0020] In uninterruptible power supply (UPS) performance testing, AC / DC dummy load testing is crucial for ensuring test accuracy. However, existing testing methods have significant shortcomings: the RLC impedance load box combined with simple power device control results in low power density and only allows for coarse power adjustment, failing to accurately simulate the complex real-time power factor changes of a real server; the separation of AC and DC testing leads to cumbersome operation; and a single air-cooling method is insufficient for high power density testing, easily overheating during prolonged testing, resulting in large deviations and poor stability in test data.

[0021] To address the aforementioned issues, this application proposes an AC / DC dummy load testing method capable of steplessly generating reactive power. This method allows for flexible switching between AC and DC load modes according to testing requirements. A digital signal processor calculates the target reactive power value based on instantaneous reactive power theory and generates a current command signal. Stepless adjustment of reactive power and continuous adjustability of the AC / DC load are achieved through the coordinated control of thyristors and insulated-gate bipolar transistors. A combination of air cooling and liquid cooling is used for heat dissipation, with the heat dissipation intensity automatically adjusted based on real-time temperature.

[0022] To enable those skilled in the art to better understand the present application, the following detailed description is provided in conjunction with the accompanying drawings and specific embodiments. The core of this application is to provide an AC / DC dummy load testing method capable of steplessly generating reactive power. A flowchart of one specific embodiment is shown below. Figure 1 As shown, the method includes the following steps S101 to S105.

[0023] S101, in response to the test command, reconfigures the internal circuit into an AC load topology or a DC load topology, and captures the instantaneous voltage and current values ​​at the output of the uninterruptible power supply under test.

[0024] Specifically, in response to the test command, based on the current operating mode of the uninterruptible power supply (UPS) under test and the load type requirements in the test command, the internal circuit is reconfigured into an AC load topology or a DC load topology, and the instantaneous voltage and current values ​​at the output terminal of the UPS are captured by the data acquisition module. The instantaneous voltage value refers to the specific value of the UPS output voltage at a given moment, and the instantaneous current value refers to the specific value of the UPS output current at a given moment.

[0025] When the device receives a test command from the host computer, the digital signal processor (DSP) automatically determines whether to reconfigure the topology to AC or DC load based on the test type specified in the command and the current operating mode of the uninterruptible power supply (UPS). For example, when the test command requires a load test on a 380V / 50Hz or 220V / 50Hz AC UPS, the DSP determines it to be an AC load topology; when the test command requires a load test on a 240V DC UPS or a 48V high-voltage DC power distribution system, the DSP determines it to be a DC load topology; when the test command requires sequentially performing AC and DC compatibility tests on the same UPS, the DSP switches between AC and DC load topologies according to the sub-condition sequence in the test command.

[0026] Specifically, refer to Figure 2 The diagram shows an AC load topology circuit. The AC load topology consists of a thyristor, an insulated gate bipolar transistor, and an energy dissipation unit connected in series at the AC output terminal of the uninterruptible power supply under test to form a loop. A voltage sensor is connected in parallel at the output terminal of the power supply under test to collect the instantaneous voltage value, and a current sensor is connected in series in the loop to collect the instantaneous current value.

[0027] exist Figure 2 In AC mode, the AC output of the power supply under test sequentially passes through a thyristor for phase-shifting conduction, an insulated-gate bipolar transistor for amplitude and harmonic correction, and an energy dissipation unit, thus forming a complete circuit. Voltage and current sensors acquire instantaneous values ​​in real time and send them to a digital signal processor. These instantaneous values ​​serve as inputs for the theoretical calculation of instantaneous reactive power and the generation of current commands.

[0028] In AC mode, the thyristor performs phase-shifting triggering to establish the fundamental phase difference between voltage and current. The insulated gate bipolar transistor corrects the amplitude deviation and harmonic components of the current without directional reversal of the fundamental phase difference. The energy consumption unit adopts a thick-film heating element, and its heat exchange surface is simultaneously fitted with air-cooled heat sink fins and liquid-cooled flow channels.

[0029] Reference Figure 3 The schematic diagram of the DC load topology circuit shown is provided. The DC load topology consists of an insulated-gate bipolar transistor and an energy dissipation unit connected in series at the DC output terminal of the uninterruptible power supply under test. The DC load topology does not contain thyristors. The voltage sensor and current sensor are connected in the same way as the AC load topology. The insulated-gate bipolar transistor operates according to the duty cycle of the pulse width modulation signal, thereby generating a continuously adjustable DC current to simulate an adjustable DC load.

[0030] exist Figure 3 In DC mode, the circuit does not contain thyristors. The DC output of the power supply under test directly passes through an insulated-gate bipolar transistor (IGBT) and an energy-dissipating unit to form a complete circuit. The IGBT is controlled by a pulse-width modulation (PWM) signal. This transistor operates according to a preset duty cycle, enabling continuous adjustment of the DC current, thereby simulating an adjustable DC load.

[0031] If the uninterruptible power supply under test is currently in AC output mode and the test command requires simulating an AC load, the digital signal processor sends a control signal to the internal relay matrix to control the relays to switch their closed state, thus reconstructing the internal circuit into an AC load topology. If the uninterruptible power supply under test is currently in DC output mode, it switches to a DC load topology through the relay group.

[0032] Reference Figure 4 The schematic diagram of the relay matrix switching circuit shown includes an internal relay matrix comprising AC path relays. and DC path relay ,in Connected in series in the branch between the output terminal of the power supply under test and the anode of the thyristor. The branch connected in series at the output of the power supply under test, bypassing the thyristor and directly connecting to the input of the insulated-gate bipolar transistor; digital signal processor output. control signals and Control signals are mutually exclusive, meaning that only one relay closes at any given time.

[0033] When the digital signal processor determines that it is in AC mode, the output makes... closure, The control signal is disconnected, and the output of the power supply under test is... Connect to an AC load topology; when the digital signal processor determines it is in DC mode, the output... closure, The control signal is disconnected, and the output of the power supply under test is... The DC load topology bypasses the thyristor. The switching action of the relay matrix is ​​completed within a few milliseconds after the test command arrives. No power switching devices operate during the switching process, thereby avoiding electrical shock to the uninterruptible power supply under test.

[0034] exist Figure 4 In the middle, the switching logic is as follows: In AC mode, the digital signal processor enables the switch Closing, opening and closing Disconnect, current flows through Thyristors and insulated-gate bipolar transistors flow to the energy consumption unit.

[0035] In DC mode, the digital signal processor enables closure, Disconnect, current flows through The energy flows directly into the insulated-gate bipolar transistor, bypassing the thyristor, and then into the energy-consuming unit.

[0036] and Mutual exclusion at any time ensures that only one path is connected to the power supply under test, avoiding dual path conflicts.

[0037] Meanwhile, the device uses a built-in high-precision data acquisition module to capture the instantaneous voltage and current values ​​at the output of the uninterruptible power supply under test in real time. The acquisition frequency meets the requirements of real-time testing, providing reliable input data for the reactive power calculation of the digital signal processor in subsequent steps.

[0038] For example, the equipment uses a DSP28335 digital signal processor as its control core, and the host computer sends test commands via RS485 communication. Upon receiving the command, the digital signal processor determines the operating mode of the uninterruptible power supply (UPS) based on the command content and the operating mode of the UPS under test, and immediately sends a control signal to the internal relay matrix to complete topology reconstruction. The equipment's built-in ADS1256 high-precision analog-to-digital converter (ADC) chip serves as the data acquisition module, with a sampling frequency set to 10kHz. Voltage and current sensors are connected to the output terminals of the UPS under test, respectively, to capture the instantaneous values ​​of the output voltage and current in real time, acquiring data every 100 microseconds. The acquired values ​​are then converted from analog to digital and transmitted to the digital signal processor.

[0039] S102. The digital signal processor calculates the target reactive power value of the IT load to be simulated based on the instantaneous voltage and current values ​​using instantaneous reactive power theory, and generates a current command signal representing the desired current vector based on the target reactive power value.

[0040] Specifically, in traditional electrical theory, reactive power is defined based on a complete AC cycle (20 milliseconds in a 50Hz AC system). Its calculation requires integrating and averaging the voltage-current product within this cycle, resulting in a slow response time that is difficult to meet real-time control requirements. Instantaneous reactive power theory, proposed by Akaki et al. in 1983, eliminates the need for integration over a complete cycle. Instead, it directly derives the instantaneous reactive power value at any sampling moment from the instantaneous voltage and current values ​​through algebraic operations. This theory enables real-time feedback and closed-loop control of reactive power, forming the theoretical basis for the "stepless generation of reactive power" achieved in this application.

[0041] The digital signal processor receives the instantaneous voltage and current values ​​transmitted from the data acquisition module in step S101, calls relevant algorithms based on instantaneous reactive power theory, and calculates the target reactive power value for the IT load to be simulated based on the acquired instantaneous voltage and current values. Then, based on this target reactive power value, a corresponding current command signal is generated. This signal carries the desired current vector information and serves as the reference for thyristor conduction phase control and insulated-gate bipolar transistor current correction in subsequent steps. The specific calculation process includes the following five steps.

[0042] The first step is the Clarke transform.

[0043] The purpose of the Clarke transform is to convert the three electrical quantities of a three-phase balanced system into two equivalent orthogonal electrical quantities, thereby reducing three-dimensional calculations to two dimensions. This reduces the computational burden on the digital signal processor and provides a foundation for subsequent coordinate transformations. Geometrically, the Clarke transform projects three voltage vectors in three-phase space along directions 120° apart onto two mutually perpendicular coordinate systems. On the β and β axes. Let the instantaneous values ​​of the three-phase output voltages of the uninterruptible power supply under test be respectively... , , The instantaneous values ​​of the three-phase output current are respectively , , , through 3 -2 The transformation converts three-phase quantities to -β Two-phase stationary coordinate system:

[0044]

[0045] When the uninterruptible power supply under test is a single-phase output, the original electrical signal has only one phasor, making it impossible to directly apply the instantaneous reactive power theory based on three-phase systems. Therefore, a virtual two-phase method is used: the measured single-phase voltage is... Single-phase current direct as Axial components (i.e.) = , = ), and through Hilbert transform or FIR-based all-pass filter to , Generates a 90° phase shift β-axis virtual component (i.e. , This application preferably uses an all-pass filter based on FIR, because its phase shift is strictly 90° across the entire frequency band and does not introduce amplitude attenuation. The typical filter order is 63 or 127, and the sampling rate should be more than 100 times the fundamental frequency.

[0046] The second step is to calculate the instantaneous active power. and instantaneous reactive power :

[0047]

[0048] The third step is to determine the target value of reactive power for the IT load that needs to be simulated. .

[0049] The value is given directly by the test command, or based on the active power target value in the test command. and power factor target value Reverse calculation:

[0050] In particular, The convention for positive and negative signs is as follows: >0 indicates inductive (current lags voltage). A value less than 0 indicates capacitive (current leads voltage). The test command explicitly includes an inductive / capacitive characteristic flag to avoid issues arising solely from | The ambiguity regarding the nature of the load cannot be distinguished. The thyristor phase-shift triggering method in this application supports inductive simulation by default; if the test command requires simulation of capacitive load, it needs to be achieved through phase lead compensation in the IGBT correction stage.

[0051] Fourth step, based on and Reverse calculation Current command signal in the -β coordinate system:

[0052]

[0053] Step 5: Obtain the three-phase current command signal by inverse Clarke transform:

[0054]

[0055]

[0056] Numerical substitution calculation for S102: Taking a three-phase 380V / 50Hz AC system as an example, assuming the current time The instantaneous values ​​of the three-phase output voltage and current of the uninterruptible power supply under test are as follows (voltage phase is taken as...). (60° after rising from zero): =269V =-269V、 =0V; =12.2A、 =-12.2A、 =0A (pure active load condition).

[0057] Substituting into the Clarke transform formula, we get Voltage and current components in the -β coordinate system:

[0058]

[0059]

[0060]

[0061] Substitute into the instantaneous power formula:

[0062]

[0063] The value ≈0 confirms that this operating condition is a pure active load, consistent with the initial assumption, verifying the correctness of the algorithm's link calculation. If the test command requires simulating the power factor... =0.8 (emotional), active target =1000W IT load, reverse calculation of reactive power target: =1000×tan(arccos(0.8))=750var (positive value, inductive). Substituting into the current command formula, we get... ≈1.29A ≈-3.03A, obtained as a three-phase current command signal after inverse Clarke transformation. ≈1.05A ≈-2.67A ≈1.62A, and output as a current command signal representing the desired current vector to step S103.

[0064] S103. When the internal circuit is reconfigured into an AC load topology, the thyristors and insulated gate bipolar transistors are controlled in coordination according to the current command signal to generate an AC current that conforms to the current command signal and apply the AC current to the AC output terminal of the uninterruptible power supply under test.

[0065] The core control principle of this step is "division of control": the thyristor is responsible for establishing the fundamental phase difference, and the insulated gate bipolar transistor (hereinafter referred to as IGBT) is responsible for correcting the amplitude deviation and harmonic components of the initial alternating current. The two do not conflict with each other in terms of control and complement each other.

[0066] Specifically, the thyristor is triggered by a phase shift, causing the initial alternating current flowing through the energy consumption unit to form a continuously variable fundamental phase difference relative to the output voltage of the power supply under test. The magnitude of this phase difference is determined by the firing angle command. (Calculated from the reactive power target value in the current command signal) to achieve basic adjustment of reactive power; Under the premise of maintaining the above fundamental phase difference without directional reversal (i.e. without making the inductive load capacitive or vice versa), the IGBT uses the current command signal as a reference to correct the amplitude deviation of the d-axis (active component) of the initial AC current and the high-order harmonic components (mainly the 5th, 7th, 11th and 13th) introduced by the intermittent conduction of the thyristor.

[0067] Since the fundamental phase difference (corresponding to the q-axis current component) established by the thyristor is retained as an outer loop constraint in the IGBT correction control loop, the IGBT only fine-tunes the d-axis current tracking and high-frequency harmonic cancellation through PWM action, while the correction amplitude of the q-axis current component is strictly constrained by the limiter to ensure that the directionality of the initial fundamental phase difference is not destroyed, thereby avoiding conflict between the control objectives of the two devices.

[0068] Reference Figure 5 The diagram shows the structure of the coordinated control of thyristor phase-shift triggering and insulated gate bipolar transistor correction in AC mode. The specific implementation of step S103 is divided into two interconnected parts: the left side S103-A group (thyristor phase-shift triggering branch) and the right side S103-B group (IGBT amplitude / harmonic correction branch).

[0069] The left-side group S103-A receives the reactive power target value from the current command signal generated in step S102 as input. Through four sub-steps S103-A1 to S103-A4, it sequentially completes the firing angle calculation, synchronization signal generation, delay time calculation, and thyristor driving, outputting the initial AC current flowing through the energy consumption unit. The right-side group S103-B, using the current command signal from step S102 as a reference, corrects the d-axis amplitude deviation and higher harmonic components of the initial AC current, while maintaining the directionality of the fundamental phase difference established by group S103-A. Through three sub-steps S103-B1 to S103-B3, it sequentially completes coordinate transformation and deviation calculation, proportional resonance adjustment, and PWM synthesis. The two groups work together to ensure the accuracy and reliability of the final current applied to the AC output terminal of the uninterruptible power supply under test.

[0070] (I) S103-A group: Thyristor phase shift triggering to establish fundamental phase difference Group S103-A Figure 5 Corresponding to the left branch, the "phase-shift triggering" refers to controlling the triggering timing of the thyristor (delaying the timing from the zero-crossing point of the AC voltage for a certain period of time) to create a phase difference between the current and the voltage. The longer the delay time, the greater the phase difference and the greater the reactive power. This process goes through the following four sub-steps in sequence.

[0071] In stage S103-A1, the phase-shifting triggering algorithm determines the reactive power target value in the current command signal generated in step S102. and active power target value Calculate the required phase-shift firing angle within each AC half-wave. This generates the trigger angle command. Following the example in section S102.2, =750var =1000W, =arctan(0.75)≈36.87°.

[0072] In stage S103-A2, the synchronization signal detection circuit captures the zero-crossing point of the output voltage of the uninterruptible power supply under test in real time and generates a synchronization pulse signal. The "zero-crossing point" refers to the moment when the instantaneous value of the AC voltage crosses the zero level. There are two zero-crossing points in each complete AC cycle (20 milliseconds at 50Hz) (corresponding to the voltage changing from negative to positive and from positive to negative, respectively). The "synchronization pulse signal" is a short pulse (typically 1~10 microseconds) generated by the zero-crossing detection circuit at each zero-crossing point. It is used to provide a time reference for the phase-shift triggering algorithm and ensure that the occurrence time of the trigger pulse is precisely synchronized with the voltage waveform.

[0073] In stage S103-A3, the phase-shifting trigger algorithm calculates the delay time from the zero-crossing point to the trigger point within each AC half-wave based on the trigger angle command and the synchronization pulse signal, and generates a trigger pulse sequence:

[0074] Where f is the power grid frequency (typically 50Hz or 60Hz). In a 50Hz AC system, =36.87° corresponds to the delay time =36.87°×( (180°) / 314.16≈2.05ms. That is, after the synchronization pulse arrives at each voltage zero-crossing point, the phase-shift triggering algorithm waits 2.05 milliseconds to generate a trigger pulse.

[0075] In stage S103-A4, the trigger pulse sequence is sent to the phase-shift trigger drive circuit. This circuit uses an optocoupler to achieve strong and weak current isolation, separating the weak current signal from the thyristor signal and preventing interference from the strong current signal to the weak current control circuit. When the trigger pulse arrives, it drives the thyristor to conduct at a non-zero-crossing point, creating a continuously variable fundamental phase difference between the initial AC current flowing through the energy consumption unit and the voltage waveform. ≈ .

[0076] Figure 6 The timing relationship of each key signal in the phase-shift trigger is further illustrated. The first waveform is the AC voltage waveform u(t) output by the power supply under test, which has a zero-crossing point every 10 milliseconds (the lower half-wave period of 50Hz) (marked as a dot in the figure); the second waveform is the synchronization pulse signal, which generates a short pulse at each voltage zero-crossing point; the third waveform is the trigger pulse sequence, which is delayed relative to the synchronization pulse. A trigger pulse is generated approximately 2.05ms later (corresponding to...) =36.87°); The fourth waveform is the initial AC current i(t) flowing through the energy consumption unit, which starts flowing through the thyristor from the trigger point to the next zero crossing (turn-off), showing a characteristic of lagging behind the voltage waveform by about 36.87°.

[0077] (II) S103-B Group: IGBTs correct the amplitude deviation and harmonic components of the initial AC current based on current commands. Group S103-B Figure 5 Corresponding to the right branch, and constrained by maintaining the fundamental phase difference established by group S103-A from directional reversal, the d-axis amplitude deviation and higher harmonic components of the initial AC current are corrected in the dq rotating coordinate system. The "dq rotating coordinate system" is a system based on the fundamental angular frequency of the power grid. A synchronously rotating coordinate system with a speed of 314.16 rad / s, from the observer's perspective, renders the fundamental sinusoidal quantity "frozen" as a stationary DC quantity (because the rotation of the coordinate system itself cancels out the sinusoidal change in the electrical quantity), thus allowing for zero steady-state error tracking of the DC quantity using a conventional PI controller. In the dq coordinate system, the d-axis current component... Corresponding active current (in phase with voltage), q-axis current component Corresponding reactive current (perpendicular to voltage phase).

[0078] In stage S103-B1, the current command signal in the stationary coordinate system is converted to the dq rotating coordinate system that rotates synchronously with the grid voltage to obtain the d-axis component of the command. and q-axis components Simultaneously, the preliminary AC current feedback value acquired in real time by S103-A4 is also transformed to the same rotating coordinate system to obtain the feedback. and In particular, since the q-axis component has been established by the phase-shifting trigger of the thyristors in group S103-A, the correction action of the IGBT in this step is subject to the mandatory constraint of "maintaining the fundamental phase difference from directional reversal".

[0079] Specifically, let the initial q-axis current component established by group S103-A be... (when feeling emotional) >0, capacitive If <0), then the q-axis current component after constraint correction by the q-axis correction limiter. satisfy:

[0080] in The limiting threshold is typically set to... =0.5×| (That is, the correction amount of the IGBT to the q-axis current does not exceed 5% of the initial value), and strictly ensures that and Same sign (if the calculation results have different signs, the smallest value with the same sign will be forcibly taken, typically 0.05 × | |). Set so small Limiting is intended to ensure that the correction of the IGBT only plays a "fine-tuning" role, that is, the influence of the IGBT on reactive power is strictly limited to within ±5%, and the main value of reactive power is still determined by the α angle triggered by the phase shift of the thyristor.

[0081] For example, let =+5A (sensibility), =5%×5=0.25A, then the IGBT can adjust the q-axis current component within the range of [4.75A, 5.25A], corresponding to a reactive power fluctuation of ±5%.

[0082] If the theoretical q-axis correction amount calculated by the IGBT correction stage =+0.18A, satisfying | |≤ The actual value taken is this. like =+0.5A exceeds Then the amplitude is limited to +0.25A; like =-0.6A will cause the q-axis to flip to capacitive, triggering directional protection to force a change. =0.25A instead of allowing flipping.

[0083] Under this constraint, the division of control between the thyristor and IGBT is clear: the thyristor's phase-shift triggering is responsible for determining the main value of reactive power (error ≤ ±5%), while the IGBT focuses on refining the amplitude of the d-axis (active component) and suppressing various harmonic components with the remaining 95% control margin. Their control objectives do not conflict. Therefore, the main correction target is the d-axis component. Deviations and harmonics.

[0084] In stage S103-B2, the d-axis current deviation will be... = The signal is fed into a proportional resonant regulator to generate a compensation voltage vector. The aforementioned "proportional resonance (PR) regulator" is a proportional element... Based on this, at the specified resonant frequency A resonant peak is superimposed at the resonant frequency, which theoretically has infinite gain. A sinusoidal quantity at that frequency can achieve zero steady-state error tracking. The transfer function of the PR regulator is:

[0085] Among them, the resonant frequency Take the fundamental angular frequency of the AC voltage being measured (50Hz corresponds to) =314 rad / s, corresponding to 60 Hz =377rad / s); Cutoff frequency Used to set the width of the resonance peak, typically 5~10 rad / s; proportionality coefficient = · ( The system's desired current loop bandwidth is typically 1500~3000 rad / s. (Input filter inductor); resonant gain Typical values ​​are 50-200. For major harmonics such as the 5th, 7th, 11th, and 13th harmonics, [the value can be adjusted accordingly]. Based on this, multiple resonant units are connected in parallel (with resonant frequencies of 5). 7 11 13 Achieving simultaneous suppression of multiple harmonics, fundamental resonant unit Typically, 100~200 is used for higher order resonant units. The values ​​are decreased sequentially (50-100 for the 5th time, 30-60 for the 7th time, 20-40 for the 11th time, and 10-30 for the 13th time) to ensure that the phase margin PM of the Bode plot of the system after the multi-resonance parallel connection is ≥45° and the amplitude margin GM is ≥6dB.

[0086] In stage S103-B3, according to the compensation voltage vector The IGBT is driven by a first pulse width modulation signal synthesized using either Space Vector Pulse Width Modulation (SVPWM) algorithm (three-phase scenario) or Sinusoidal Pulse Width Modulation (SPWM) algorithm (single-phase scenario). SVPWM is based on the synthesis principle of 6 effective voltage vectors and 2 zero vectors in a three-phase inverter. By adjusting the duration of adjacent effective vectors, it synthesizes any desired space voltage vector, offering advantages such as high DC bus voltage utilization (approximately 15.5% higher than SPWM) and low harmonic content. SPWM generates a PWM signal by comparing a sinusoidal modulation wave with a triangular carrier wave, suitable for single-phase systems or applications where DC bus utilization is not critical. The SVPWM / SPWM output duty cycle is related to the compensation voltage and the DC bus voltage before the IGBT. The relationship is:

[0087] in, The DC bus voltage before the IGBT (provided by the internal auxiliary rectifier unit, typically 400V or 800V), The scope is limited to [ / 2,+ / 2] to ensure that "overmodulation" is not achieved. For example =400V At 120V, 0.5 + 120 / 400 = 0.8, meaning the IGBT conducts for 80 microseconds and turns off for 20 microseconds in each switching cycle, ensuring an average output voltage of 320V per switching cycle, which meets the requirements of the compensation voltage command. The IGBT operates according to the duty cycle of the PWM signal, correcting the initial AC current to ensure accurate d-axis amplitude tracking while maintaining the fundamental phase difference established by group S103-A. Furthermore, higher harmonic components are effectively suppressed. The corrected alternating current is applied to the AC output terminal of the uninterruptible power supply under test.

[0088] (III) Optional: Deadbeat current predictive control based on electronic load mathematical model Optionally, based on group S103-B, deadbeat current predictive control based on the mathematical model of the electronic load can be used to further improve the correction accuracy. The term "deadbeat" refers to the control system ensuring that the controlled variable (actual current) accurately reaches the target value (current command) within the shortest possible time (one switching cycle in this application, typically 100 microseconds) without residual deviation. Its working mechanism is: based on the mathematical model of the controlled object (electronic load circuit in this application), and according to the current measurement value at the current moment... and the target current value at the next moment By working backwards, the voltage value that the IGBT should output in the next switching cycle can be derived. This allows for the calculation of the corresponding PWM duty cycle. This control method is a hybrid "feedforward + feedback" control, offering a faster response than a pure feedback PR regulator, but it is less sensitive to model parameters (inductance). ,resistance The accuracy of ) is highly dependent.

[0089] The electronic load circuit of this application is simplified to an RL equivalent circuit, wherein the instantaneous voltage at the output terminal of the uninterruptible power supply under test is... The current flowing through the energy consumption unit is The equivalent voltage at the IGBT output terminal is The total inductance of the circuit is The total resistance is Then the continuous-time state equation is:

[0090] During the sampling period The following uses forward Euler discretization to obtain a discrete mathematical model of the electron load:

[0091] The objective of deadbeat predictive control is to ensure that the actual current in the next switching cycle catches up with the commanded current, i.e. = The inverse solution yields the IGBT output voltage command for the next switching cycle:

[0092] Corresponding duty cycle =[ +0.5· ] / .parameter , The value is measured once during equipment manufacturing using an LCR tester and then fixed in the DSP's program storage area. Data is collected in real time by a DC bus voltage sensor. This is the switching sampling period for the DSP (typically 10μs~100μs). For example, =100V =4.7A、 =5.0A、 =2mH、 =0.1Ω =100μs, When the voltage is 400V, substituting into the formula yields... =100 0.47 6 = 93.53V, duty cycle =(93.53+200) / 400=0.734, meaning that the IGBT will conduct for 73.4 microseconds in the next 100 microsecond switching cycle, making the average output voltage of the switching cycle equal to 93.53V, and the current value predicted for the next moment is 5.0A, thus achieving deadbeat tracking.

[0093] Specifically, in the above formula and The calibration accuracy directly affects the steady-state performance of deadbeat control, especially in practical engineering. and Due to factors such as temperature drift, magnetic saturation, and long-term aging, the current may slowly deviate from the calibration value, causing steady-state current oscillations. To improve robustness, this application preferably adds a disturbance observer (DOB) or extended Kalman filter (EKF) in addition to the deadbeat predictive controller to estimate the current in real time. and deviation and And it is compensated in real time during duty cycle calculation. The typical update cycle of DOB or EKF is 10 to 100 times. As a simpler alternative, a first-order low-pass filter can be used to adjust the duty cycle. The rapidly changing components are smoothed (the cutoff frequency is taken as 1 / 5 to 1 / 3 of the system's expected bandwidth), sacrificing some dynamic response in exchange for steady-state robustness.

[0094] Optionally, the method also includes zero-crossing trigger control in a purely active load scenario: Optionally, when in AC output mode and the test command requires simulating a pure active load, and the thermal time constant of the energy consumption unit is more than 10 times the zero-crossing trigger control cycle, such that the temperature fluctuation amplitude of the energy consumption unit under intermittent cycle conduction does not exceed the preset temperature threshold, the method of coordinating the control of the thyristor and IGBT in step S103 can be replaced by zero-crossing trigger control of the thyristor.

[0095] This application provides a zero-crossing triggering control method as an alternative to the S103-A / B cooperative control method to achieve better spectrum performance. It should be noted that in a purely active load scenario ( =0), using the S103-A / B collaborative control mode (at this time, the phase-shifting trigger angle α=0, that is, the full half-wave conduction, and the IGBT adjusts the current amplitude through the PWM duty cycle), although it can output a current in phase with the voltage, the continuous adjustment of active power depends on the decay of the PWM duty cycle of the IGBT in each switching cycle. This will superimpose high-frequency harmonic components near the switching frequency (typically in the 10kHz~20kHz frequency band) on the current waveform, causing interference to the power quality assessment of the uninterruptible power supply under test.

[0096] In contrast, the zero-crossing triggering method regulates active power through a "full-cycle on / off" mechanism: the thyristor is always on and off at the voltage zero-crossing point. During the on-cycle, the current flowing through the energy-consuming unit is a complete sinusoidal half-wave, perfectly in phase with the voltage and without any harmonic components; during the off-cycle, the current is zero. By adjusting the on / off ratio of "number of on cycles / total number of cycles" per unit time, continuous regulation of the average active power can be achieved. This method offers significantly better spectral purity than the cooperative control method, but at the cost of "discontinuous" average power output in cycles, with instantaneous power fluctuating between "rated value" and "zero." Therefore, the energy-consuming unit must possess sufficiently large thermal inertia to prevent its temperature fluctuations from being significantly affected by the discontinuous cycles. Based on the above reasons, this application sets "the thermal time constant of the energy consumption unit is more than 10 times the zero-crossing trigger control cycle" as the preset condition for adopting the zero-crossing trigger mode. That is, when the test scenario has high requirements for the purity of the current spectrum and the thermal inertia of the energy consumption unit is sufficient to withstand the cycle discontinuous condition, the zero-crossing trigger mode is enabled; otherwise, the S103-A / B cooperative control mode is still used to ensure temperature stability.

[0097] The "zero-crossing triggering" (also known as cycle control) is an alternative thyristor control method to phase-shift triggering. Its characteristic is that the thyristor's turn-on and turn-off times are strictly controlled at the AC voltage zero-crossing point. When on, it outputs a complete sine half-wave; when off, it does not conduct for the entire cycle. Because the thyristor always switches at the voltage zero point, there is no forced current cutoff, thus preventing the generation of low-order harmonics in the frequency spectrum, resulting in a clean output current spectrum. The "thermal time constant" is a physical quantity describing the "inertia" of the temperature change of the heating element, defined as the ratio of the thermal energy stored in the heating element to the heat dissipation power per unit temperature difference. The larger the value, the greater the inertia of the temperature change of the heating element, so that the temperature only fluctuates slightly around the average value under intermittent cycle conduction.

[0098] Specifically, the thermal time constant of the heating element Determined by the following formula:

[0099] Where m is the mass of the heating element (kg), is the specific heat capacity of the heating element material (J / (kg·K)), and h is the convective heat transfer coefficient of the heating element surface (W / (m²·K)). Effective heat exchange area (m²). Zero-crossing trigger control cycle. Typically, the cycle number is taken as an integer multiple of the power grid cycle. =0.1 seconds to 1 second. The load thermal inertia meets the preset conditions, specifically defined as follows: .

[0100] For example, the physical parameters of the thick-film heating element used in this application are: mass m = 0.5 kg, specific heat capacity... =900J / (kg·K) (taking alumina ceramic matrix as an example), surface convection heat transfer coefficient h=50W / (m²·K), effective heat transfer area =0.02m², substituting into the formula for the thermal time constant, we get =(0.5×900) / (50×0.02)=450 seconds. Zero-crossing trigger control cycle set. =0.2 seconds (corresponding to 10 50Hz cycles), then / =2250 times, far exceeding the lower limit requirement of 10 times, meeting the preset thermal inertia conditions. Under these conditions, the surface temperature fluctuation amplitude of the heating element under intermittent cycle conduction is about 0.044℃, which is far less than the preset temperature threshold (typically 5% of the average temperature, i.e., 5℃), ensuring a high degree of stability in active power output.

[0101] The specific zero-crossing trigger control process includes: calculating the ratio of the number of complete AC cycles to the number of cycles to be turned on to the number of cycles to be turned off within each zero-crossing trigger control cycle based on the active power target value in the test command, and generating an on / off ratio command; generating an on / off control sequence based on the on / off ratio command; sending the on / off control sequence to the zero-crossing trigger drive circuit; when the on / off control sequence indicates on and an AC voltage zero-crossing point is detected, driving the thyristor to turn on at the voltage zero-crossing point; when the on / off control sequence indicates off, driving the thyristor to turn off naturally at the next voltage zero-crossing point. For example, the active power target... =700W, rated power When the power is 1000W, the on / off ratio =0.7, in Within a 200ms control cycle, the circuit conducts for 7 cycles (140ms) and turns off for 3 cycles (60ms), resulting in an average active power of 700W without harmonic pollution.

[0102] Optionally, the zero-crossing trigger drive circuit uses an optocoupler with zero-crossing detection function. When the on / off control sequence indicates that the circuit is on and the optocoupler detects the AC voltage zero-crossing point, the bidirectional thyristor in the optocoupler is turned on, so that the thyristor is turned on at the voltage zero-crossing point.

[0103] Optionally, the method also includes a hybrid control scheme of phase-shift triggering and zero-crossing triggering under reactive load scenarios: Specifically, when in AC output mode and the test command requires simulating reactive load, and the test command carries a low harmonic requirement flag, the thyristor control in step S103 adopts a hybrid control method of phase-shift triggering and zero-crossing triggering. The "hybrid control" is a composite control method that time-division multiplexes phase-shift triggering and zero-crossing triggering. Its core idea is to divide a relatively long hybrid control cycle into two time periods, using phase-shift triggering (to achieve reactive power regulation, but introduces harmonics) and zero-crossing triggering (to achieve active power regulation, without harmonics) respectively. By adjusting the ratio of the two time periods, stepless generation of reactive power can be achieved while maintaining a low level of harmonic pollution.

[0104] Specifically, based on the reactive power target value in the current command signal. and active power target value Calculate the firing angle command required for phase-shift triggering respectively. On / off ratio instruction required for zero-crossing triggering ( (Based on rated active power), hybrid control parameters are generated. These hybrid control parameters define a phase-shift triggering operating period (the first time period) within a hybrid control cycle. ) and zero-crossing trigger working period (second time period) (time ratio) It shall be determined according to the following principles:

[0105] That is, the higher the proportion of the reactive power target value, the higher the proportion of the phase-shifted triggering working period. When When r approaches zero, it is equivalent to a complete zero-crossing trigger; when When r approaches zero, it approaches infinity, which is equivalent to full phase-shift triggering. For example... =500var =1000W, hybrid control cycle When the time is 200ms, r = 0.5, that is... : =1:2, specific division =200×1 / 3≈66.7ms =200×2 / 3≈133.3ms; Since the switching time must be strictly aligned to the voltage zero-crossing point (one zero-crossing point every 10ms at 50Hz), the actual value is rounded down to... =70ms (corresponding to 7 half-waves) =130ms (corresponding to 13 half-waves). That is, within each 200ms mixed control cycle, phase shift triggering is performed in the first 70ms and zero-crossing triggering is performed in the last 130ms. The switching of the two time periods occurs at the voltage zero-crossing point to avoid the instantaneous current surge during switching.

[0106] The specific process is as follows: A synchronization pulse signal is generated by capturing the zero-crossing point of the output voltage of the uninterruptible power supply (UPS) under test in real time. Based on the hybrid control parameters and the synchronization pulse signal, each hybrid control cycle is divided into a first time period T1 and a second time period T2. During the first time period T1, the delay time is calculated based on the trigger angle command and the synchronization pulse signal, and a trigger pulse sequence is generated to drive the thyristor to conduct at non-zero-crossing points. A fundamental phase difference is formed between the voltage waveform at the output of the UPS and the current waveform flowing through the energy consumption unit, thereby regulating reactive power. During the second time period T2, an on / off control sequence is generated based on the on / off ratio command, driving the thyristor to conduct at voltage zero-crossing points or naturally turn off, making the average active power flowing through the energy consumption unit continuously adjustable. Simultaneously, since the thyristor switches at zero-crossing points, no additional harmonic pollution is generated. By adjusting the ratio of T1 to T2, and ensuring that the switching between T1 and T2 occurs at voltage zero-crossing points, the AC current after hybrid triggering is applied to the AC output of the UPS.

[0107] Optionally, the method also includes a two-stage current fine-tuning control following hybrid triggering: Specifically, after the hybrid triggering, the process also includes a two-stage fine-tuning control of the initial AC current. This two-stage fine-tuning control is executed sequentially rather than in parallel. The two correction methods, SVPWM / SPWM+PR regulator (hereinafter referred to as feedback control) and deadbeat predictive control (hereinafter referred to as feedforward control), each have their own characteristics: feedback control is insensitive to model parameters and can respond quickly under large deviation conditions, but its steady-state accuracy is limited by the regulator bandwidth; feedforward control is based on a precise mathematical model solved in reverse, resulting in high steady-state accuracy, but it is sensitive to parameter drift and may oscillate under large deviations or parameter inaccuracies. If both methods drive the IGBT in parallel, a control conflict will occur (the same IGBT receiving two PWM commands cannot physically execute them). Therefore, this application adopts a two-stage method with sequential switching: feedback control is used for a fast response in the initial stage of large deviations, and feedforward control is switched to improve steady-state accuracy after entering the steady-state error band.

[0108] The first stage (rapid correction stage for large deviations): The space vector modulation algorithm receives the current command signal and the initial AC current feedback value actually flowing through the energy consumption unit after hybrid triggering. It converts the current command from the stationary coordinate system to the rotating coordinate system and compares it with the initial AC current feedback value. The output difference is processed by the proportional resonant regulator to generate a compensation voltage vector. Based on the compensation voltage vector, a first pulse width modulation signal is synthesized and sent to the IGBT connected in series with the thyristor, completing the first stage correction. This stage focuses on handling initial large deviations (typical deviation > 5% of rated current), with fast response and insensitivity to the model.

[0109] Stage switching condition: When the AC current deviation corrected in the first stage enters the preset steady-state error band ± (typical When the current is within 2% of the rated current, the system switches to the second stage. The alternating current is a time-varying sinusoidal quantity, and the instantaneous difference between it and the sinusoidal command signal is | The sinusoidal wave inherently fluctuates within its fundamental period. Directly using instantaneous absolute values ​​to determine whether it has entered a steady state can lead to misjudgments due to sinusoidal fluctuations. This application employs two criteria to avoid this problem: Criterion one (preferred) will... and Transform to the dq rotating coordinate system and calculate the steady-state error along the d-axis. When | |≤ Switching is triggered after N consecutive switching cycles (typically N=10); criterion two involves calculating the root mean square (RMS) of the instantaneous deviation within one fundamental cycle, and when this RMS value is ≤ Switching is triggered after N consecutive fundamental frequency cycles.

[0110] The second stage (steady-state refinement stage): The deadbeat current predictive control algorithm receives the current command signal and the real-time acquired instantaneous value of AC current, and calculates the duty cycle required for the next switching cycle based on the discrete mathematical model of the electronic load given above. A second pulse width modulation (PWM) signal is generated and sent to the IGBT to replace the first PWM signal, completing the second-stage correction. This stage features high accuracy and small steady-state error (typically ≤0.5% of rated current). Cut-back condition (optional): When a large deviation disturbance is detected in the steady-state stage (e.g., | |>3· After M consecutive switching cycles (typically M=5), the system switches back to the first stage to quickly respond to load surges; after the disturbance is eliminated, it switches back to the second stage. It is important to note that at any given time, the PWM signal received by the IGBT comes from only one of the first or second pulse width modulation signals, and these two signals are executed sequentially rather than being sent simultaneously. During the execution of the second stage, the space vector modulation algorithm continues to operate, but its output first PWM signal is clock-gated and not sent to the IGBT, and vice versa.

[0111] Numerical example: Rated current =10A, steady-state error band =2%×10=0.2A, Switching condition duration N=10 switching cycles, Switching cycle When =100μs, after the first stage of feedback control starts, if the conditions are met for 10 consecutive switching cycles from k+30 to k+39, | If |≤0.2A, then a switch to the second stage will be triggered during the (k+40th)th switching cycle; during the execution of the second stage, if | A 0.6A current for five consecutive switching cycles triggers a back-switch.

[0112] In summary, the selection of the three control methods (cooperative control, zero-crossing triggering, and hybrid control) in the AC mode provided in this application is jointly determined by the load characteristic parameters and load condition parameters in the test command, and the specific selection logic is shown in Table 1: Table 1. Selection Logic Table for Three Control Modes in AC Mode

[0113] When the test command issued by the host computer changes during the test, the digital signal processor switches to the new operating mode at the next hybrid control cycle boundary or the next AC half-wave zero crossing point to avoid the instantaneous current surge during the switch.

[0114] Specifically, the criterion for determining "compliance with the current command signal" in this application is: the instantaneous value of the actual alternating current under steady-state conditions. With current command signal In a fundamental frequency cycle The root mean square error within the range does not exceed the preset tracking accuracy. ,Right now:

[0115] in, A settable tracking accuracy threshold, typically =1%· For scenarios requiring higher testing accuracy, It can be set to 0.5%. Or even lower; for scenarios with high requirements for test stability and sensitivity to response speed, It can be relaxed to 2%. This criterion applies to all three control modes of this application. The digital signal processor calculates at the end of each hybrid control cycle. It is then uploaded to the host computer via remote communication for operator verification.

[0116] S104. When the internal circuit is reconfigured into a DC load topology, a DC current reference is generated based on the load voltage reference value and the DC output voltage value of the uninterruptible power supply under test. The insulated gate bipolar transistor is controlled by a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit, thereby forming an adjustable DC load.

[0117] Specifically, the load voltage reference value is a preset standard voltage value in DC output mode; the DC current reference is a standard DC current value generated by comparing the DC output voltage value of the uninterruptible power supply under test with the load voltage reference value, and it serves as the target reference for controlling the DC current output; an adjustable DC load refers to a load type in which the magnitude of the DC current can be continuously adjusted. The implementation structure of this step is as follows: Figure 7 As shown, it can be divided into three closed-loop elements: outer loop (voltage outer loop), inner loop (hysteresis current control loop), and feedback loop.

[0118] The implementation process of the outer loop (voltage outer loop) is as follows: load voltage reference value The test command carries the setting based on the rated DC output voltage of the uninterruptible power supply under test. For example, when the rated output of the power supply under test is 48V DC, =48V; when the rated output is 240V DC =240V. During the test, The DC output voltage of the uninterruptible power supply under test can be adjusted in real time by a host computer via test commands (typically within ±20% of the rated voltage) to verify the load capacity of the power supply under test at different voltage operating points. The DC voltage of the power supply under test is acquired in real time by the sensor. and The difference A DC current reference is generated after processing by the DC voltage outer loop PI regulator. :

[0119] The "PI controller" is short for proportional-integral controller, and its output consists of two parts: the proportional term and the integral term. · (t) provides an immediate response to the current deviation; the larger the deviation, the larger the output correction. (Integral term) ·∫ (τ)dτ accumulates the historical deviations to ensure that the deviations are completely eliminated under steady-state conditions, i.e., zero steady-state error. This is the voltage outer loop proportionality factor (typically 0.1~10A / V). The integral coefficient is typically 1~100 A / (V·s)). The parameters are designed based on the voltage response bandwidth of the power supply under test and can be obtained through engineering tuning or the Ziegler-Nichols method. In the DSP digital implementation, the continuous-time form of the above equation needs to be discretized. This application uses an incremental PI algorithm, the specific expression of which is as follows:

[0120]

[0121] in, This refers to the DSP's sampling period (typically 100μs~1ms). Incremental PI versus positional PI has the following advantages: First, it eliminates the need to store the accumulated integral value, avoiding integral saturation after prolonged operation; second, during mode switching or faults, the output surge is only the current increment, not the accumulated value; third, it requires less DSP computation. Numerical example: =48V =1ms =2、 =10, current measured value =46V deviation =2V, the previous moment =1.5V, substituting into the incremental PI formula, we get Δ =2×(2-1.5)+10×0.001×2=1.02A, if the previous moment =8A, then at the current time =8 + 1.02 = 9.02A.

[0122] The implementation process of the inner loop (hysteresis current control loop) is as follows: Optionally, the pulse width modulation signal is generated according to the DC current reference to drive the IGBT to operate. This application preferably adopts a hysteresis current control method. The "hysteresis current control" is a switching control method based on dual threshold judgment, where the DC current reference... Set a threshold line at the top and bottom, respectively. + (Upper threshold) and (Lower threshold) The hysteresis width is defined as follows: when the actual current drops below the lower threshold, the IGBT is turned on to increase the current; when the actual current rises above the upper threshold, the IGBT is turned off to decrease the current. The trajectory of the actual current fluctuates back and forth between the upper and lower threshold lines, forming a loop-like trajectory, hence the name "hysteresis". This control method has the engineering advantages of simple structure, fast dynamic response, and insensitivity to parameter changes.

[0123] Specifically, the step of controlling the insulated gate bipolar transistor with the pulse width modulation signal further includes the following three sub-steps: S104-1 The instantaneous value of the DC current flowing through the energy consumption unit is collected in real time. and Compare; S104-2 when When the output signal is activated, it drives the IGBT to conduct, causing the DC current to rise; S104-3 When When the IGBT is turned off, the output shutdown signal drives the IGBT to turn off, causing the DC current to decrease. The instantaneous DC current value refers to the specific value of the DC current flowing through the energy consumption unit at a certain moment, which is collected in real time; the hysteresis width is a preset current deviation threshold, used to avoid frequent IGBT switching on or off due to small current fluctuations, and to ensure the stability of current regulation.

[0124] The operating waveform of hysteresis current control is as follows: Figure 8 As shown, the vertical axis represents direct current. The horizontal axis represents time t. Baseline. (9.02A) has upper and lower thresholds respectively. (9.22A) and lower threshold (8.82A), hysteresis width 2 =0.4A. The actual current (blue) exhibits a sawtooth pattern, fluctuating between two threshold lines: when the lower threshold is reached, the IGBT turns on, causing the current to rise; when the upper threshold is reached, the IGBT turns off, causing the current to fall. The IGBT switch status bar displays the corresponding ON / OFF switching sequence. The average actual current is the reference value of 9.02A.

[0125] exist Figure 8 In the middle, when the direct current The drop reached the lower threshold When the insulated-gate bipolar transistor turns on, the current rises; when The rise reached the upper threshold When the insulated-gate bipolar transistor (IGBT) is turned off, the current decreases. The actual current fluctuates between the two threshold lines, and the average value is the DC current reference value. Switching frequency The calculation formula is: .

[0126] Hysteresis width The principle for determining the value: Use 0.5% to 5% of the rated DC current. Use a smaller value when high testing accuracy is required. =0.5%~1% At this point, the current fluctuation is small, but the IGBT switching frequency is high and the loss increases; a larger value is taken when the test requires high stability and is sensitive to IGBT losses. =2%~5% For example: typical hysteresis width at a rated DC current of 10A. =0.2A (i.e., 2%). Switching frequency for hysteresis current control. With hysteresis width Circuit inductance DC bus voltage The following theoretical relationships exist:

[0127] From this, we can deduce that: given the maximum allowable switching frequency of an IGBT... (Determined by IGBT switching losses and heat dissipation capacity, typically) =10kHz~50kHz), then the minimum value of the hysteresis width Should meet .For example =400V =2mH、 At 20kHz, ≥2.5A, usually taken as in engineering =1.5· As a safety margin, the feedback loop is implemented as follows: the output current of the energy consumption unit is collected by the current sensor and sent back to the hysteresis comparator to form a complete closed loop, ensuring that the actual current always fluctuates around the reference value within the hysteresis width range.

[0128] S105. The heat energy converted from the current flowing through the energy consumption unit is simultaneously discharged through the air-cooled heat dissipation structure and the liquid-cooled heat dissipation structure, and the heat dissipation intensity of the heat dissipation system is automatically adjusted according to the real-time monitored temperature.

[0129] Among them, the air-cooled heat dissipation structure is a heat dissipation component that uses airflow generated by a fan to remove the shallow heat from the surface of the energy-consuming unit, and its heat transfer coefficient is [missing information]. Typical values ​​are 50~200 W / (m²·K), suitable for rapid removal of shallow, low-density heat; liquid-cooled heat dissipation structures are heat dissipation components that absorb heat through the circulation of coolant (typically deionized water or ethylene glycol aqueous solution) within the flow channel, and their heat transfer coefficient is... Typical values ​​are 3000~10000W / (m²·K), approximately 50 times that of air cooling, making it suitable for efficient removal of deep, high-density heat. Simultaneous heat removal refers to the two structures operating simultaneously and coordinating heat dissipation.

[0130] Regardless of whether it's the AC mode in step S103 or the DC mode in step S104, the current after power regulation ultimately flows into the energy consumption unit. The energy consumption unit converts electrical energy into heat energy. Both air cooling and liquid cooling structures are activated simultaneously. Air cooling is responsible for the initial removal of heat from the surface of the heating element, while liquid cooling is responsible for the deep removal of high-density heat from the core of the heating element. The two work together to adapt to high power density testing scenarios (typically ≥10W / cm²). The device monitors key temperatures in real time through a temperature acquisition module and transmits the data to the control core. The control core adjusts the heat dissipation intensity according to preset thresholds. If the temperature is too high, the fan speed is increased and the coolant flow rate is increased; if the temperature is too low, it is appropriately reduced to ensure that the heat dissipation intensity matches the actual needs.

[0131] Optionally, the energy consumption unit adopts a thick-film heating element. The heat-conducting substrate of the thick-film heating element is a ceramic substrate (typically alumina Al2O3 or aluminum nitride AlN) or a metal substrate (typically a copper surface insulating oxide layer). Its heat exchange surface is simultaneously fitted with an air-cooled heat dissipation fin structure and a liquid-cooled flow channel structure. The thermal time constant of the thick-film heating element is greater than 2 seconds to achieve high power density energy consumption and efficient heat dissipation.

[0132] This application provides a schematic diagram of a specific implementation of an AC / DC dummy load testing system capable of steplessly generating reactive power. The system may include: The topology reconfiguration module is used to respond to test commands and reconfigure the internal circuit into an AC load topology or a DC load topology according to the current operating mode of the uninterruptible power supply under test and the load type requirements in the test command. The data acquisition module is used to capture the instantaneous voltage and current values ​​at the output terminal of the uninterruptible power supply under test. The instruction generation module, which includes a digital signal processor, is used to calculate the target reactive power value of the IT load to be simulated based on the instantaneous voltage and current values ​​using instantaneous reactive power theory, and to generate a current instruction signal representing the desired current vector based on the target reactive power value. The AC load control module is used to coordinate the control of thyristors and insulated gate bipolar transistors connected in series with the thyristors according to the current command signal when the internal circuit is reconfigured into an AC load topology. This generates an AC current that flows through the energy consumption unit and conforms to the current command signal, and applies the AC current to the AC output terminal of the uninterruptible power supply under test. The DC load control module is used to generate a DC current reference based on the load voltage reference value and the DC output voltage value of the uninterruptible power supply under test when the internal circuit is reconfigured into a DC load topology. It controls the insulated gate bipolar transistor through a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit. The heat dissipation adjustment module is used to convert the heat energy converted from the current flowing through the energy consumption unit into heat energy, and simultaneously export it through the air-cooled heat dissipation structure and the liquid-cooled heat dissipation structure, and automatically adjust the heat dissipation intensity of the heat dissipation system according to the real-time monitored temperature.

[0133] Optionally, the AC load control module further includes a thyristor control submodule and an insulated-gate bipolar transistor (IGBT) correction submodule: the thyristor control submodule is used to control the conduction phase of the thyristor according to the reactive power target value in the current command signal, to establish a fundamental phase difference corresponding to the reactive power target value between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit, and to generate a preliminary AC current containing the fundamental phase difference flowing through the energy consumption unit; the IGBT correction submodule is used to correct the amplitude deviation and harmonic components of the preliminary AC current based on the current command signal, while maintaining the fundamental phase difference from directional reversal.

[0134] Optionally, the DC load control module further includes a voltage outer loop submodule and a current inner loop submodule: the voltage outer loop submodule includes a proportional-integral (PI) regulator, used to process the difference between the DC output voltage value of the uninterruptible power supply (UPS) under test and the load voltage reference value to generate the DC current reference, and the load voltage reference value is set and adjustable by the test command based on the rated DC output voltage of the UPS under test; the current inner loop submodule is used to generate the pulse width modulation (PWM) signal according to the DC current reference, driving the insulated-gate bipolar transistor (IGBT) to operate according to the duty cycle of the PWM signal, generating a continuously adjustable DC current to the energy consumption unit. Optionally, the current inner loop submodule further employs a hysteresis current control method.

[0135] The AC / DC dummy load test system capable of steplessly generating reactive power in this application embodiment is used to implement the aforementioned AC / DC dummy load test method capable of steplessly generating reactive power. Therefore, the specific implementation of the AC / DC dummy load test system capable of steplessly generating reactive power can be found in the embodiment section of the AC / DC dummy load test method capable of steplessly generating reactive power above. The specific implementation can be referred to the description of the corresponding embodiment, and will not be repeated here.

[0136] The foregoing has provided a detailed description of the AC / DC dummy load testing method and system capable of steplessly generating reactive power, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for testing AC / DC dummy loads capable of steplessly generating reactive power, characterized in that, include: In response to test commands, the internal circuitry is reconfigured into an AC load topology or a DC load topology, and the instantaneous voltage and current values ​​at the output of the uninterruptible power supply under test are captured. The digital signal processor calculates the target reactive power value of the IT load to be simulated based on the instantaneous voltage and current values ​​using instantaneous reactive power theory, and generates a current command signal representing the desired current vector based on the target reactive power value. When the internal circuit is reconfigured into an AC load topology, the thyristor and the insulated gate bipolar transistor connected in series with the thyristor are controlled in coordination according to the current command signal to generate an AC current that flows through the energy consumption unit and conforms to the current command signal, and the AC current is applied to the AC output terminal of the uninterruptible power supply under test. When the internal circuit is reconfigured into a DC load topology, a DC current reference is generated based on the load voltage reference value and the DC output voltage value of the uninterruptible power supply under test. The insulated gate bipolar transistor is controlled by a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit, thereby forming an adjustable DC load. The heat energy converted from the current flowing through the energy consumption unit is simultaneously exported through the air-cooled heat dissipation structure and the liquid-cooled heat dissipation structure, and the heat dissipation intensity of the heat dissipation system is automatically adjusted according to the real-time monitored temperature. Specifically, when in AC output mode and the test command requires simulating reactive load, and the test command carries a low harmonic requirement flag, the control of the thyristor in the coordinated control adopts a hybrid control method of phase-shift triggering and zero-crossing triggering, specifically including: Based on the reactive power target value and active power target value in the current command signal, the trigger angle command required for phase-shift triggering and the on / off ratio command required for zero-crossing triggering are calculated respectively, and hybrid control parameters are generated. The hybrid control parameters define the duration ratio of the phase-shift triggering working period to the zero-crossing triggering working period within a hybrid control cycle. The duration ratio is determined according to the ratio of the reactive power target value to the active power target value. The higher the proportion of the reactive power target value, the higher the proportion of the phase-shift triggering working period. The zero-crossing point of the output voltage of the uninterruptible power supply under test is captured in real time, and a synchronization pulse signal is generated. Based on the hybrid control parameters and the synchronization pulse signal, each hybrid control cycle is divided into a first time period and a second time period. During the first time period, the delay time is calculated according to the trigger angle command and the synchronization pulse signal and a trigger pulse sequence is generated to drive the thyristor to conduct at a non-zero crossing point, thereby forming a fundamental phase difference between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit. During the second time period, an on / off control sequence is generated according to the on / off ratio command, driving the thyristor to turn on or turn off naturally at the voltage zero crossing point, so that the average active power flowing through the energy consumption unit is continuously adjustable. By adjusting the ratio of the first time period to the second time period, and ensuring that the switching between the first time period and the second time period occurs at the voltage zero-crossing point, the AC current after mixed triggering is applied to the AC output terminal of the uninterruptible power supply under test.

2. The method according to claim 1, characterized in that, The step of coordinating the control of the thyristor and the insulated-gate bipolar transistor connected in series with the thyristor according to the current command signal to generate an alternating current flowing through the energy-consuming unit and conforming to the current command signal specifically includes: Based on the reactive power target value in the current command signal, the conduction phase of the thyristor is controlled to establish a fundamental phase difference corresponding to the reactive power target value between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit, thereby generating a preliminary alternating current containing the fundamental phase difference flowing through the energy consumption unit. While maintaining the fundamental phase difference without directional reversal (i.e. without causing the inductive load to become capacitive or the capacitive load to become inductive), the amplitude deviation and harmonic components of the initial AC current are corrected using the insulated gate bipolar transistor connected in series with the thyristor, with the current command signal as a reference, to obtain the AC current that conforms to the current command signal.

3. The method according to claim 1, characterized in that, The step of generating a DC current reference based on the load voltage reference value and the DC voltage value output by the uninterruptible power supply under test, and controlling the insulated gate bipolar transistor with a pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit specifically includes: The load voltage reference value is set according to the rated output DC voltage of the uninterruptible power supply under test carried in the test command. The load voltage reference value can be adjusted in real time by the test command during the test. The DC output voltage value of the uninterruptible power supply under test is compared with the load voltage reference value, and the difference is processed by a proportional-integral regulator to generate the DC current reference. The pulse width modulation signal is generated based on the DC current reference, and the insulated gate bipolar transistor is driven to operate according to the duty cycle of the pulse width modulation signal to generate a continuously adjustable DC current to the energy consumption unit.

4. The method according to claim 2, characterized in that, The step of controlling the conduction phase of the thyristor according to the reactive power target value in the current command signal, and establishing a fundamental phase difference corresponding to the reactive power target value between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit, further includes: Based on the reactive power target value in the current command signal, the required phase-shifting firing angle within each AC half-wave is calculated, and a firing angle command is generated. The zero-crossing point of the output voltage of the uninterruptible power supply under test is captured in real time, and a synchronization pulse signal is generated. The phase-shift triggering algorithm calculates the delay time from the zero-crossing point to the trigger point within each AC half-wave based on the trigger angle command and the synchronization pulse signal, and generates a trigger pulse sequence. The trigger pulse sequence is sent to the phase-shift trigger drive circuit, which uses an optocoupler to achieve strong and weak current isolation. When the trigger pulse arrives, it drives the thyristor to conduct at a non-zero crossing point, forming a continuously variable fundamental phase difference between the voltage waveform at the output terminal of the uninterruptible power supply under test and the current waveform flowing through the energy consumption unit.

5. The method according to claim 2, characterized in that, The step of using the insulated-gate bipolar transistor connected in series with the thyristor to correct the amplitude deviation and harmonic components of the preliminary alternating current based on the current command signal, and obtaining the alternating current that conforms to the current command signal, further includes: The current command signal in the stationary coordinate system is converted to the rotating coordinate system and compared with the feedback value of the converted initial AC current to obtain the current deviation value. The current deviation value is processed by a proportional resonant regulator to generate a compensation voltage vector; A first pulse width modulation signal is synthesized based on the compensation voltage vector using a space vector pulse width modulation algorithm or a sinusoidal pulse width modulation algorithm, and sent to the insulated gate bipolar transistor to drive it to correct the initial alternating current.

6. The method according to claim 5, characterized in that, Also includes: Based on the mathematical model of the electronic load, the current command signal at the current moment and the instantaneous value of the AC current collected in real time are substituted into the mathematical model to calculate the duty cycle required for the next switching cycle. A second pulse width modulation signal is generated based on the calculated duty cycle and sent to the insulated gate bipolar transistor instead of the first pulse width modulation signal, so that the actual alternating current at the next moment tracks the current command signal within one switching cycle.

7. The method according to claim 1, characterized in that, When in AC output mode and the test command requires simulating a pure active load, and the thermal time constant of the energy consumption unit is greater than 10 times the zero-crossing trigger control cycle, such that the temperature fluctuation amplitude of the energy consumption unit under intermittent cycle conduction does not exceed a preset temperature threshold, the method of co-controlling the thyristor and the insulated gate bipolar transistor is replaced by zero-crossing trigger control of the thyristor, specifically including: Based on the active power target value in the test command, the ratio of the number of complete AC cycles to the number of cycles to be turned on to the number of cycles to be turned off in each zero-crossing trigger control cycle is calculated, and the on / off ratio command is generated. A switching control sequence is generated based on the switching ratio command. The switching control sequence is used to determine the on or off state of each complete AC cycle within the zero-crossing trigger control cycle. The on / off control sequence is sent to the zero-crossing trigger drive circuit. When the on / off control sequence indicates on and the AC voltage zero-crossing point is detected, the thyristor is driven to turn on at the voltage zero-crossing point. When the on / off control sequence indicates off, the thyristor is driven to turn off naturally at the next voltage zero-crossing point, so that the average active power flowing through the energy consumption unit is continuously adjustable.

8. The method according to claim 7, characterized in that, The zero-crossing trigger drive circuit uses an optocoupler with zero-crossing detection function. When the on / off control sequence indicates that the circuit is on and the optocoupler detects the AC voltage zero-crossing point, the bidirectional thyristor in the optocoupler is turned on, so that the thyristor is turned on at the voltage zero-crossing point.

9. The method according to claim 1, characterized in that, After the hybrid triggering, the process also includes a two-stage correction of the initial AC current actually flowing through the energy consumption unit. The two-stage correction is executed sequentially, and specifically includes: The current command signal and the initial AC current feedback value that actually flows through the energy consumption unit after hybrid triggering are received by the space vector modulation algorithm. A compensation voltage vector is generated, and a first pulse width modulation signal is synthesized according to the compensation voltage vector and sent to the insulated gate bipolar transistor connected in series with the thyristor to perform the first stage correction on the initial AC current. When the AC current deviation corrected in the first stage enters the preset steady-state error band, the deadbeat current prediction control algorithm receives the current command signal and the instantaneous value of the AC current collected in real time, calculates the duty cycle required for the next switching cycle based on the mathematical model of the electronic load, generates a second pulse width modulation signal to replace the first pulse width modulation signal and sends it to the insulated gate bipolar transistor to complete the second stage correction, and stops generating the first pulse width modulation signal during the second stage correction.

Images