An automatic transfer switch grading selection control method based on surge current characteristic partitioning

Abstract

This invention discloses an automatic transfer switch selection and control method based on inrush current characteristic partitioning: First, the operation process of the automatic transfer switch is decomposed into time, a three-state timing control function model is established, a system-level simulation loop is constructed, and the results are verified by dual experimental measurements of inductive load switching waveforms and capacitive load inrush current data; Full matrix scanning of the inrush current coefficient is performed on four groups of RC loads with different capacitance parameters at nine switching times, revealing that the inrush current of weak capacitive loads tends to stabilize 20ms after power failure, while the inrush current of strong capacitive loads shows a rapid inflection point at 70ms after power failure; Using 20ms and 70ms as time thresholds, the capacitive load loop is divided into three control segments. Weak capacitive loads are configured with high-speed transfer switches to shorten the interruption time, strong capacitive loads are configured with fast transfer switches before the inrush current inflection point to reduce the inrush current using residual capacitor voltage, and after the inflection point, traditional automatic transfer switches are used to withstand large inrush current impacts.

Description

Technical Field

[0001] This invention belongs to the field of power quality control technology for low-voltage power distribution systems, specifically relating to a method for modeling and analyzing inrush current of automatic transfer switches for capacitive loads in low-voltage power supply systems and for differentiated switching control. Background Technology

[0002] In recent years, the focus of power quality research has gradually shifted from steady-state problems such as harmonic distortion and overvoltage to transient disturbances such as voltage sags and short-term voltage interruptions. The latter, due to their high frequency and wide impact, are widely recognized as the most prominent power quality risks currently facing power systems. These transient processes involve dynamic interactions across multiple time scales, placing high demands on the solution accuracy and model flexibility of simulation tools. The MATLAB / Simulink platform, with its modular modeling capabilities, microsecond-level transient solution accuracy, and support for hardware-in-the-loop verification, has been widely used in related fields.

[0003] Due to their unique electrical architecture and load characteristics, urban low-voltage power supply systems are particularly sensitive to transient power quality. These systems extensively utilize power electronic devices, such as transformer-frequency converters and precision control systems, which can trigger protection actions when the voltage drops to 85% of its rated value for more than 15ms. Urban low-voltage power supply systems also house a wide variety of capacitive loads, such as data center server power supplies, commercial building inverter air conditioning drives, communication base station power supplies, and large-scale LED inverter lighting drives. The front-end circuits of these devices generally employ a rectifier bridge + DC bus filter capacitor topology. During normal power supply, the filter capacitor is charged to a DC voltage close to the AC peak value. Once power is interrupted, the capacitor continues to discharge through the downstream load resistor, and the terminal voltage decays exponentially according to the RC time constant. When power is restored via a switchover, the capacitor, whose terminal voltage has significantly decreased, suddenly experiences a high-amplitude voltage from the power supply side. The voltage difference between the two drives a transient charging current with an amplitude far exceeding the rated value, which is the inrush current. Actual measurement data shows that the peak inrush current can reach nearly eight times the steady-state rated current. Excessive inrush current not only causes severe arcing and Joule thermal shock on switch contacts, accelerating material migration and mechanical fatigue in silver alloy contacts, but also exceeds the setting values ​​of the upstream circuit breaker's instantaneous trip unit or electronic protection, triggering malfunctions and expanding the power outage area. In extreme cases, it can even cause contact welding or irreversible damage to the load-side rectifier devices. Urban low-voltage loads have significant spatiotemporal concentration characteristics. During peak urban operating hours, if inrush current issues cause multiple critical equipment to shut down cascadingly or trip beyond their rated capacity, it will directly impact the city's public services and lifeline infrastructure. The large-scale system paralysis and secondary socio-economic risks caused during peak hours cannot be ignored.

[0004] In power system protection architecture, automatic transfer switch (ATS) is the most widely used emergency measure. Deployed on the distribution side, it switches on to the backup power supply after a 500-1000ms delay upon detecting a main power supply failure, ensuring that the load does not suffer a complete power outage. Short-term voltage interruptions are a type of power supply fault with a much higher disturbance level than voltage dips, posing a more direct threat to the continuity of power supply to critical loads. Current technical means to address short-term voltage interruptions also include UPS, DVR, and energy storage systems: UPS requires energy storage units and is limited by battery life and maintenance costs, resulting in limited continuous power supply time under high-power conditions; DVR relies on series voltage injection, and its compensation capability completely fails under 100% voltage interruption conditions; while energy storage systems have scalable capacity, their construction costs are high and their dynamic response stability is insufficient. These devices are either limited by capacity and cost or lack sufficient compensation capability under full voltage interruption conditions, making it difficult to simultaneously meet the millisecond-level response speed and the continuous power supply requirements of high-power loads.

[0005] Automatic transfer switches (ATS) have become a key technology for solving short-term power outage problems due to their millisecond-level rapid switching capability (typically 15-30ms). ATS can achieve rapid switching between primary and backup power sources without the need for energy storage components. Their dual-power switching mode is naturally compatible with urban low-voltage dual-busbar power supply architectures. With rated load switching current covering hundreds to thousands of amperes, ATS has become a core emergency device for ensuring power continuity in critical power supply locations such as subways, data centers, and hospitals.

[0006] A search revealed that Chinese invention patent CN106300376A discloses a surge-free switching device and its control method suitable for switching capacitive loads. The surge-free switching device includes a mechanical switch, a first unidirectional thyristor, a second unidirectional thyristor, a control unit, and a current-limiting element. The first unidirectional thyristor and the current-limiting element are connected in series to form a series circuit, which is connected in parallel with the second unidirectional thyristor. The main circuit terminals of the mechanical switch are connected in parallel with the second unidirectional thyristor. The control unit is connected to the control terminal of the mechanical switch and to the first and second unidirectional thyristors. The control unit pre-stores the operating time parameters of the mechanical switch. This invention has the advantages of short thyristor conduction time, strong overload capacity, high cost-effectiveness, long electrical life of the mechanical switch, and high reliability.

[0007] The technology of this invention is compared with that of the aforementioned patent as follows:

[0008] 1. The patent "Inrush-Free Switching Device and Control Method" focuses on eliminating inrush current at the hardware topology level. Its core lies in constructing an inrush-free switching device consisting of a current-limiting thyristor branch connected in parallel with a direct-through thyristor branch, and then in parallel with a mechanical switch branch. This achieves a sequence of actions when a capacitive load is connected: first, current-limiting pre-charging; then, full-power conduction; and finally, mechanical switch connection. This ensures that the capacitor's voltage is charged to near the power supply voltage via the current-limiting thyristor branch before closing, thereby eliminating the closing voltage difference and suppressing inrush current. In contrast, this patent introduces an automatic transfer switch model with time-series control. In the MATLAB / Simulink platform, key time thresholds are extracted through a full matrix scan of the capacitive load inrush current characteristics, resulting in a differentiated switch selection strategy based on the power outage duration. The two differ fundamentally in their core working mechanisms: the former relies on hardware topology-driven inrush current elimination, while this patent employs a system-differentiated control strategy combining time-series modeling and tiered selection. Their core working principles are completely different.

[0009] 2. The patented "Inrush-Free Switching Device and Control Method" does not consider the size of the load capacitance or the duration of the power outage; it eliminates inrush current by pre-charging the load through a current-limiting thyristor branch. This patent, however, measures the inrush current coefficient of different capacitive loads at various switching times and extracts the inrush current growth rate and jump time threshold of capacitive loads with different parameters using a curve inflection point identification method. Based on this, different capacitive circuits are selected: a high-speed switching switch (HTSE), a fast switching switch (QTSE), and a traditional automatic transfer switch. The two differ significantly in their time scale for inrush current handling and their switching strategies.

[0010] A search revealed that Chinese invention patent CN106711924A discloses a device and method for suppressing inrush current during capacitive load connection and overvoltage during connection. The device includes a circuit breaker for capacitive loads and a control module. The circuit breaker includes a first auxiliary switch, a second auxiliary switch, a main switch, a first resistor, and a second resistor. The method is as follows: When closing, the control module controls the first auxiliary switch to close first, connecting the first resistor in series with the circuit; then controls the second auxiliary switch to close, connecting the parallel resistance of the first and second resistors in series with the circuit; finally, controls the main switch to close, short-circuiting the current-limiting resistor, and then controls the first and second auxiliary switches to open, effectively limiting the inrush current during circuit breaker connection. When opening, the control module controls the second auxiliary switch to close first; then controls the main switch to open, connecting the second resistor in series with the circuit; finally, controls the second auxiliary switch to open, disconnecting the circuit, effectively limiting the overvoltage during connection. This invention can simultaneously solve the problems of effectively suppressing inrush current during capacitive load connection and overvoltage during capacitive load disconnection.

[0011] The technology of this invention is compared with that of the aforementioned patent as follows:

[0012] 1. The patent "A Device and Method for Suppressing Inrush Current During Capacitive Load Closing and Overvoltage During Interruption" focuses on the optimized design of parallel resistors when switching capacitors in a medium-voltage power distribution system circuit breaker. Its core is an auxiliary branch containing resistive components connected in parallel across the main switch of the circuit breaker. Through multi-stage closing and opening timing control, resistors of different resistance values ​​are switched sequentially, thereby using the resistive components to dissipate transient energy and suppress inrush current during closing and overvoltage during interruption. In contrast, this patent addresses the automatic transfer switch switching scenario in a low-voltage dual-power supply architecture. The switching circuit does not introduce any current-limiting or voltage-limiting components; instead, it uses the residual voltage decay law governed by the RC time constant of the capacitive circuit to classify load strength and match different types of automatic transfer switches. The two differ fundamentally in their applied voltage levels and suppression mechanisms: the former is a medium-voltage single-equipment suppression scheme that uses resistors to dissipate system transient energy, while this patent is a differentiated switching control method in a low-voltage dual-power supply architecture that uses selection and grading to avoid inrush current impact.

[0013] 2. The patent "A Device and Method for Suppressing Inrush Current and Interrupting Overvoltage of Capacitive Loads" uses fixed time parameters for two predetermined time periods: closing and opening. Its function is to allow reasonable intervals for the step-by-step operation of main and auxiliary switches, as well as the engagement and disengagement of current-limiting and voltage-limiting resistors. It provides a unique resistance matching combination only for medium-voltage reactive power compensation capacitor banks. This patent, however, extracts two key time thresholds with clear physical meanings—20ms and 70ms—through a full matrix scan. These thresholds respectively mark the critical point where the inrush current of a weak capacitive load enters a steady state and the inflection point where the inrush current growth rate of a strong capacitive load abruptly changes. These are used for selecting and classifying automatic transfer switches, ensuring that the power outage duration matches the switch's mechanical strength and short-time withstand current level. The former's time parameter is a delay interval reserved for the action of resistive suppression components, while the latter is a classification boundary extracted from the inrush current growth law. The two differ significantly in their physical meaning and engineering applications.

[0014] Upon retrieval, the US patent with publication number US10566787B2 discloses unique systems, methods, techniques and apparatuses for inrush current detection and reduction. One exemplary embodiment is a method for transmitting power to a load, which includes operating a solid-state switching device of a power distribution network device with a microcontroller-based controller. The solid-state switching device includes a gate and is structured to receive a signal with the gate to control the switching device to receive power from a power source and selectively provide power, including an output current, to the load; detecting an overcurrent condition in the output current; operating the solid-state switching device to determine that the load is a capacitive load in a charging condition in response to detecting an overcurrent condition; and operating the solid-state switching device to reduce the magnitude of the output current during the charging condition.

[0015] The technical comparison between the present invention and the above patent is as follows:

[0016] 1. The patent "INRUSH CURRENT DETECTION AND CONTROL WITH SOLID-STATESWITCHING DEVICES" focuses on inrush current detection and suppression of solid-state semiconductor switching devices (MOSFETs / IGBTs, etc.) in power distribution protection. It achieves inrush current detection and suppression by applying voltage signals of different amplitudes and timings to the gate through a microcontroller. In contrast, this patent uses an automatic transfer switch composed of instantaneously excitation-holding mechanical contactors. It does not rely on the gate controllability of semiconductor devices, but instead establishes a three-state timing control function model based on a sign function, using a two-stage criterion under a dual-power supply architecture to drive the main and backup contactors to switch on and off. The two differ fundamentally in their switching device mechanisms and control signal forms: the former belongs to an active current shaping method for solid-state semiconductors based on gate signal modulation, while this patent belongs to a passive switching control method for mechanical switches based on timing modeling and graded selection.

[0017] 2. The patent "INRUSH CURRENT DETECTION AND CONTROL WITH SOLID-STATESWITCHING DEVICES" employs a strategy of current classification and current reduction. It first distinguishes between capacitive inrush current and short-circuit faults by observing characteristics such as the peak output current, delay, or dv / dt of the voltage across the switch under a fixed-width gate pulse. Then, it gradually charges the capacitor using methods such as pulse widening or high-frequency current-limiting pulse trains. This patent, however, does not identify the cause of the inrush current. Instead, it pre-determines two key time thresholds of 20ms and 70ms through simulation and actual measurement, thus solidifying the matching relationship between the switch type, load capacitance, and power-off duration range in one step. The two patents differ significantly in the timing, feedback method, and implementation of inrush current handling. Summary of the Invention

[0018] This invention addresses the inrush current problem during automatic transfer switching (ATS) of capacitive loads in low-voltage power supply systems, proposing a segmented control method for ATS based on inrush current characteristic partitioning. This method focuses on overcoming the challenges of modeling the ATS operation process, constructing the low-voltage power distribution system network topology, and analyzing inrush current characteristics and formulating differentiated strategies.

[0019] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0020] An automatic transfer switch selection control method based on inrush flow characteristic zoning includes the following steps:

[0021] S1. Modeling of timing control functions for automatic transfer switching appliances;

[0022] S2. System-level simulation platform construction and switching performance verification;

[0023] S3, Simulation and experimental cross-validation of inrush current during capacitive load closing;

[0024] S4. System scanning and pattern extraction of inrush flow characteristics under multi-capacity operating conditions;

[0025] S5. Differentiated switching control strategy formulation and switch selection.

[0026] As a preferred technical solution of the present invention, step S1 is specifically as follows:

[0027] S11. Construct the timing control function model for the automatic transfer switch:

[0028] The automatic transfer switch includes a smart contactor QS1 located on the main power supply side and a smart contactor QS2 located on the backup power supply side. The main power supply is stepped down by transformer T1 and the backup power supply is stepped down by transformer T2 and then connected through QF1 and QF2 respectively. Each downstream load branch is equipped with an independent protection circuit breaker QF3~QFN.

[0029] S12. Motion Time Decomposition and Model Simplification:

[0030] The total operating time of the automatic transfer switch is It encompasses the entire process from detecting the main power supply deviation to completing the dual power supply switching, and its time structure is broken down into three parts:

[0031]

[0032] in, The inherent mechanical action time for contact switching. Set a delay manually. To account for the computation time of the voltage detection algorithm, the model will include... and Unified zeroing process, i.e. ;

[0033] S13. Mathematical modeling of the control function of automatic transfer switching appliances:

[0034] Based on the action time decomposition, the control function of the automatic transfer switch is established, and its mathematical expression is as follows:

[0035]

[0036] in, The instantaneous value of the main power supply voltage. The operating voltage threshold, This is the real-time delay starting from the moment QS1 is disconnected. The zero-position transition time, i.e. The symbolic function sgn is defined as:

[0037]

[0038] S14, Modal Partitioning and Transition Logic:

[0039] According to the control function The output value of the automatic switching device is divided into three modes based on its operating state:

[0040] when When QS1 is closed and QS2 is open, the system is in mains power supply mode;

[0041] when When QS1 and QS2 are both disconnected, the system is in a zero-position locked state.

[0042] when At this time: QS1 is open and QS2 is closed, the system is in standby power supply mode;

[0043] The transition process between the three modes mentioned above is as follows:

[0044] When the main power supply is normal , When QS1 is closed and QS2 is open, the load receives power from the main power supply. Fall to the following, When the signal changes from +1 to 0, QS1 disconnects, and the system then enters a zero-bit transition phase. Not yet counted Previously, the controller did not execute. The corresponding QS2 closing command forces the system to maintain the zero-position state of QS1 and QS2 open, pending... Full After the lock is released, When the value changes from 0 to -1, QS2 closes, and the backup power supply is officially connected to the load.

[0045] S15. Power quality assessment after switching:

[0046] After the automatic transfer switch completes the switching process, a voltage quality assessment is performed on the load side, as detailed below:

[0047] The total harmonic distortion (THD) is calculated by extracting the fundamental component amplitude V1 from the voltage waveform using a fast Fourier transform.

[0048]

[0049] Among them, V h The amplitude of the h-th harmonic is required to have a THD of no more than 5%.

[0050] As a preferred technical solution of the present invention: the interlocking station of the intelligent contactors QS1 and QS2 is a three-position station structure.

[0051] As a preferred technical solution of the present invention, step S2 is specifically as follows:

[0052] S21. Simulation Model Architecture Setup

[0053] The simulation model is built in Simulink as follows:

[0054] Both the main power contactor module Q1 and the backup power contactor module Q2 are driven by a common control bus. The bus signal generates a first subsystem and a second subsystem through two paths. The first subsystem collects the three-phase voltage from the main circuit, and feeds it into the MATLAB Function module after gain scaling. The second subsystem performs the same operation on the backup circuit. The two signals are combined and enter the controller module. In conjunction with the simulation clock, the controller module executes the operation logic of the control function of the automatic switching device, and finally outputs three modes of drive signals (+1 / 0 / -1) to control the on and off of the contactor.

[0055] S21, Inductive Load Simulation Verification

[0056] By setting the port voltage of the inductive load as the observation object, a power-off switching simulation is performed on the model:

[0057] Without the intervention of an automatic transfer switch, the voltage of the inductive load drops directly to zero after the main power supply is interrupted.

[0058] When the automatic transfer switch is engaged, and the transfer delay is set, the system voltage recovers to the rated range within a preset time.

[0059] As a preferred technical solution of the present invention: in step S21, the inductive load is set as a variable frequency drive motor, and the port voltage of the variable frequency drive motor is the object of observation.

[0060] As a preferred technical solution of the present invention, step S3 is as follows:

[0061] S31. Construction of Capacitive Load Experimental System

[0062] With a three-phase four-wire power supply as the input, the circuit passes through a disconnecting switch, a molded case circuit breaker, a single-phase circuit breaker, and an automatic transfer switch in sequence, and is connected to the capacitive load terminal consisting of three sets of parallel bridge rectifiers and a switchable RC array.

[0063] S32, Actual Cross-Validation

[0064] By changing the capacitance value at the capacitive load end and scanning step by step with the conversion time, the peak value of the AC / DC inrush current and the impact coefficient under each combination are recorded one by one, and compared with the measured results under the same working conditions to complete the secondary verification of the simulation model.

[0065] As a preferred technical solution of the present invention, step S4 is specifically as follows:

[0066] S41, Inrush Data Acquisition

[0067] After confirming the accuracy of the inrush dimension in the model, the power outage time was gradually extended from the lowest value to the highest value, and the peak inrush flow was recorded at each power outage time.

[0068] S42, Full Matrix Inrush Characteristics Scan

[0069] Using a simulation model verified by actual measurements, a full matrix scan was performed on four typical loads. The incremental changes of the impact coefficient between adjacent transition time levels were compared one by one. When the increment showed a significant jump within a certain time interval, the starting time of that interval was the critical inflection point of the inrush growth. At the same time, the strong and weak capacitive loads were classified according to whether the impact coefficient curve tended to flatten after 20ms. If the curve entered the steady state interval after 20ms, it was determined to be a weak capacitive load; if the curve continued to rise after 20ms, it was determined to be a strong capacitive load.

[0070] As a preferred technical solution of the present invention, step S5 is as follows:

[0071] Based on the analysis in step S4, key inrush flow characteristics are obtained, and a differentiated switching control strategy is formed accordingly:

[0072] Slow capacitive load switching scheme: For devices with input filter capacitors of 0.52mF and below, the inrush current enters a steady-state platform and the absolute amplitude reaches the preset threshold 20ms after power failure. The minimum switching delay of the automatic transfer switch is set to 20ms to skip the initial oscillation range of the inrush current. In this case, the high-speed transfer switch HTSE is selected.

[0073] Short-term power failure switching solution for highly capacitive loads: For devices with input filter capacitors of 13.2mF and above, the capacitors retain a considerable level of residual voltage when the power failure time does not exceed 70ms. The natural buffering effect of the residual voltage keeps the inrush current within a controllable range. In this case, a fast transfer switch (QTSE) is selected, and the switching is completed before the inrush current inflection point arrives.

[0074] Long-term power outage switching scheme for highly capacitive loads: After the power outage time exceeds 70ms, the inrush current has passed the growth inflection point and entered the rapid growth range. In this case, a traditional automatic transfer switch is selected to take over the closing.

[0075] For intermediate load sections: For intermediate load sections with capacitance values ​​between 0.52mF and 13.2mF, the inflection point of the inrush current curve can be calculated based on the actual RC time constant, and the above three-level strategies can be used for classification and selection.

[0076] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0077] This invention establishes an automatic transfer switch timing control model with a sign function as its core. The automatic transfer switch action is discretized into three quantifiable and adjustable state modes, making the switching timing in the simulation no longer a fixed parameter but a continuously adjustable research variable. Secondly, through cross-comparison of simulation and actual measurements, and using the slope abrupt change point identification method of the impact coefficient curve, the boundary law of the inrush current of weak capacitive loads tending to a steady state after 20ms and the inrush current of strong capacitive loads exhibiting a jump in growth rate at 70ms is quantitatively determined, providing two key time thresholds with engineering quantitative significance for capacitive load switching control. Finally, a selection and classification control scheme for HTSE / QTSE / traditional automatic transfer switches is presented, categorized by load capacitive strength and power outage duration. This limits the inrush current impact within the safety threshold of each type of switch, balancing power supply continuity and equipment operational reliability, and can directly serve the engineering design and operation and maintenance decisions of low-voltage power supply systems. Attached Figure Description

[0078] Figure 1 This diagram shows the network topology of a low-voltage power supply system and the working principle of an automatic transfer switch.

[0079] Figure 2 The diagram shows the structure of an automatic transfer switch simulation model built on the Simulink platform.

[0080] Figure 3 This is a comparison chart of the dynamic waveforms of the load voltage under conditions without an automatic transfer switch;

[0081] Figure 4 This is a comparison chart of the dynamic waveforms of the load voltage after the automatic transfer switch is engaged;

[0082] Figure 5 The results are the FFT spectrum analysis of the phase A voltage after switching.

[0083] Figure 6 This is a schematic diagram of the capacitive load experimental system.

[0084] Figure 7 Simulated waveform for a 10ms transition time under a 5Ω / 30mF capacitive load;

[0085] Figure 8 Simulated waveform for a 20ms transition time under a 5Ω / 30mF capacitive load;

[0086] Figure 9 The graph shows the results of a 10ms conversion experiment under a 5Ω / 30mF capacitive load.

[0087] Figure 10 The graph shows the results of a 20ms conversion experiment under a 5Ω / 30mF capacitive load.

[0088] Figure 11 The curves show the AC side inrush current coefficient as a function of the automatic transfer switch switching time for four typical capacitive loads (5Ω / 30mF, 5Ω / 13.2mF, 5Ω / 0.52mF, 5Ω / 0.3mF). Detailed Implementation

[0089] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0090] like Figure 1 As shown, this invention proposes an automatic transfer switch selection and control method based on inrush current characteristic zoning, with a low-voltage power supply system as the engineering background. Figure 1 As shown, the power supply system of this system consists of a first-section busbar (35 / 0.4kV), a second-section busbar (35 / 0.4kV), and an independent 0.4kV backup power system. During normal operation, incoming circuit breakers QF1 and QF2 are closed, while the bus tie circuit breaker QF3 remains open, and both busbars independently supply power to downstream loads. When a short-term voltage interruption occurs on the first-section busbar, the protection process is as follows: QF1 disconnects and isolates the faulty section; after a delay set by the automatic transfer switch on the first-section load feeder side, QF3 closes, and the second-section busbar supplies power to the first-section load. However, this process generates a large inrush current, which is the core problem that this invention aims to solve.

[0091] The specific steps of this invention are as follows:

[0092] Step S1: Model the timing control function of the automatic switching appliance, such as... Figure 1 As shown, the details are as follows:

[0093] Construct a timing control function model for automatic transfer switching equipment:

[0094] The automatic transfer switch consists of two sets of mechanically interlocked intelligent contactors: QS1 (main power side) and QS2 (standby power side). The main power supply is stepped down by transformer T1, and the standby power supply is stepped down by transformer T2 before being connected to the main power supply via QF1 and QF2 respectively. Each downstream load branch is equipped with an independent protective circuit breaker QF3~QFN. During normal operation, QS1 is closed to connect the main power supply, while QS2 remains open.

[0095] Intelligent contactors have two interlocking station designs: two-position and three-position. The former only has a main power supply position and a backup power supply position, with the contacts switching directly between the two positions. This allows for fast switching but carries the risk of asynchronous closing circulating current when the two power supplies are not synchronized. The latter adds a zero position between the two positions, requiring the contacts to first reside in the zero position and the main contact to completely extinguish the arc before closing the backup contact, providing a higher safety margin. In low-voltage power supply systems using dual busbar power supplies, a phase angle difference exists; therefore, a three-position station structure is chosen.

[0096] Based on the switching speed, automatic transfer switches can be divided into three levels. The key parameters of the three types of automatic transfer switches are shown in Table 1.

[0097]

[0098] Table 1

[0099] Low-voltage power supply systems cover a large number of capacitive loads and variable frequency drive motors, which have extremely stringent requirements for the speed and reliability of power switching. Different types of automatic transfer switches need to be selected appropriately according to the specific load characteristics.

[0100] This invention employs an instantaneous excitation holding type mechanical switch structure, with QTSE and HTSE fast switches as the main verification objects.

[0101] Action Time Decomposition and Modeling Simplification: Total Action Time of Automatic Transfer Switching Appliances The entire process, from detecting the main power supply deviation to completing the dual power supply switching, is broken down into three parts:

[0102]

[0103] in, The inherent mechanical action time for contact switching. Set a delay (i.e., the desired transition time) for the user. This refers to the computation time of the voltage detection algorithm. Because the selected instantaneous excitation holding contactor has extremely low mechanical inertia, The time has been compressed to less than 1ms; the single-phase derivative transformation detection algorithm used in this invention can capture voltage anomalies in an extremely short time. The time is no more than 1ms. During modeling, both are uniformly zeroed out, therefore... .

[0104] Mathematical Modeling of the Control Function of Automatic Transfer Switchgear: Based on the action-time decomposition, the control function of the automatic transfer switchgear is established, and its mathematical expression is as follows:

[0105]

[0106] in, The instantaneous value of the main power supply voltage. The operating voltage threshold is set to 70% of the rated voltage peak value, i.e., 187V, in this implementation. This is the real-time delay starting from the moment QS1 is disconnected. Zero-position transition time (i.e.) The symbolic function sgn is defined as follows:

[0107]

[0108] Three modes of partitioning and transition logic: based on control function The output value of the automatic switching device is divided into three modes based on its operating state:

[0109] when When QS1 is closed and QS2 is open, the system is in mains power supply mode;

[0110] when When QS1 and QS2 are both disconnected, the system is in a zero-position locked state.

[0111] when When QS1 is open and QS2 is closed, the system is in standby power supply mode.

[0112] The transition process between the three modes is as follows: When the main power supply is normal , When QS1 is closed and QS2 is open, the load receives power from the mains. Once... Fall to the following, The system transitions from +1 to 0, and QS1 disconnects. Afterwards, the system enters a zero-bit transition phase, and... Not yet counted Previously, the controller did not execute. The corresponding QS2 closing command forces the system to maintain the zero-position state of both QS1 and QS2 open. This interlocking mechanism has two functions: first, it isolates the potential phase angle difference between the two power supplies to prevent asynchronous closing circulating current; second, it provides the necessary mechanical action time for the contact mechanism to switch positions. Full After the lock is released, When the value changes from 0 to -1, QS2 closes, and the backup power supply is officially connected to the load.

[0113] Power Quality Assessment After Switching: After the automatic transfer switch completes the switching process, a voltage quality assessment of the load side is required. This is achieved by performing a Fast Fourier Transform (FFT) on the voltage waveform and extracting the fundamental component amplitude V1 to calculate the Total Harmonic Distortion (THD).

[0114]

[0115] Where V h The h-th harmonic amplitude should not exceed 5% according to GB / T 14549-1993.

[0116] In step S1, the operating voltage threshold Adjustable parameter set to 65%~80% of rated voltage peak value; zero-point transition time The parameter can be set to an adjustable value of 10ms to 1000ms.

[0117] This step will determine the total action time of the model. Decomposed into the inherent action time of the contact Manually set delay and voltage detection algorithm processing time The three stages are: .

[0118] Driven by the synergistic development of switch contact materials and voltage detection algorithms, the mechanical response has been compressed to a level far below that of artificial delay. Furthermore, this invention employs single-phase differentiation. The transformation detection algorithm greatly reduces the processing time of the voltage detection algorithm. Therefore, when building the model, it can be made This makes the manually set delay the only adjustable control parameter for the switching speed. Under this condition, a switching control function was constructed. ,in Characterizes the real-time voltage amplitude of the main power supply. The voltage threshold for switching action. The time for zero potential transition, This is a real-time delay variable from the moment the autonomous contact opens. Using this function, the automatic transfer switch switching behavior is divided into three discrete modes: normal power supply (…). ), backup power supply ( Simultaneously satisfy ) and zero-position locking ( Or when at the same time ).

[0119] Step S2: System-level simulation platform setup and switching performance verification, such as... Figures 2-5 As shown, the details are as follows:

[0120] Simulation model architecture construction: such as Figure 2As shown, a simulation model is built in Simulink: the main power contactor module Q1 (blue) and the backup power contactor module Q2 (green) are both driven by a common control bus. This bus signal generates subsystem 1 and subsystem 2 via two paths. Subsystem 1 collects the three-phase voltage from the main circuit, scales the gain, and feeds it into the MATLAB Function module; subsystem 2 performs the same operation on the backup circuit. The two signals merge and enter the controller module, which, in conjunction with the simulation clock, executes the control function logic of the automatic transfer switch, ultimately outputting +1 / 0 / -1 mode drive signals to control the contactor's on / off state. In practice, only the switching time (adjustable from 10ms to 1000ms) and the operating voltage threshold (187V in this implementation) parameters need to be modified to achieve the switching behavior of different models of automatic transfer switches.

[0121] Inductive load simulation verification: The port voltage of the variable frequency drive motor is used as the observation object, and the power-off switching simulation is performed on the model. Figure 3 This reflects the situation when there is no automatic transfer switch: after the main power is interrupted, the load voltage drops directly to zero, and the escalator loses power completely. Figure 4 The effect of activating the automatic transfer switch was demonstrated: after setting a 50ms transfer delay, the system voltage returned to the rated range in about 0.65s. Figure 5 The frequency domain characteristics of the restored phase A voltage are analyzed, with the fundamental amplitude being 230.1V, THD being only 0.32%, the 3rd harmonic content being the highest, and the 5th and 7th harmonics decreasing in that order. The levels of each harmonic are far below the national standard limits.

[0122] In step S2, the inductive load is a variable frequency drive motor, and the verification metrics include voltage sag rate, zero-position stage duration, and steady-state amplitude after recovery. This proves that the automatic transfer switch modeling is correct.

[0123] This step utilizes the MATLAB / Simulink environment to embed the automatic transfer switch control function into the system transient simulation framework, building an integrated simulation platform that includes a 380V dual-circuit three-phase AC power supply, a timing controller based on the MATLAB Function module, and an adjustable load module. The controller executes two-level criteria: first, it compares the voltage amplitude acquired from the load side with the preset operating voltage; second, it compares the zero-position timer reading with the preset transition time. The combined result of the two-level criteria outputs a +1 / 0 / -1 drive signal, thereby controlling the opening and closing of the main and backup contactors. The threshold values ​​for the switching time and operating voltage can be configured according to actual needs. Furthermore, a Fast Fourier Transform is performed on the switched voltage to extract the fundamental amplitude and calculate the Total Harmonic Distortion (THD). Based on this, it is determined whether the power supply quality after the switch meets the GB / T 14549-93 requirement that THD not exceed 5%.

[0124] Step S3: Cross-verification of capacitive load inrush current simulation and actual measurement, such as... Figures 6-10 As shown, the details are as follows:

[0125] Capacitive Load Experimental System Setup: Capacitive loads in low-voltage power supply networks include information technology and data center equipment (with built-in rectifiers and large-capacity filter capacitors), frequency converters and industrial control terminals (equipped with EMI filter capacitors and DC bus capacitors at the front end), commercial switching power supplies and LED driver systems, etc. Their common characteristic is that the input bridge rectifier converts AC to DC, and then large-capacity electrolytic capacitors filter and store energy. When the power supply is interrupted, the DC bus capacitor continues to discharge through the load resistor, and the residual voltage gradually decreases. However, when the power supply is automatically switched back on, the large potential difference between the low-residual-voltage capacitor and the power supply generates a closing inrush current far exceeding the rated value within a very short time.

[0126] The principle of the experimental system for dedicated capacitive loads is as follows: Figure 6 As shown in the figure. The system uses a three-phase four-wire AC power supply (L1, L2, and L3 are phase lines, and N is the neutral line). The AC input stage is configured from front to back with an isolating switch GK, a molded case circuit breaker MCCB, a single-phase circuit breaker CJ, and an automatic transfer switch. The power conversion section consists of three sets of parallel single-phase bridge rectifiers, with the output terminals connected to a common DC bus. The DC-side load is a switchable RC array, and the parallel topology contains 24 independent controllable branches, each equipped with a dedicated switching control switch. The single-phase load parameter configuration is shown in Table 2.

[0127]

[0128] Table 2

[0129] When all 24 branches are connected in parallel, the system is equivalent to a lumped-parameter resistive-capacitive load with a total resistance of approximately 5Ω and a total capacitance of approximately 30mF. By selectively switching different combinations of branches, the total capacitance can be adjusted from 30mF to various preset operating conditions such as 13.2mF, 0.52mF, and 0.3mF, covering a typical load range from strong capacitive to weak capacitive.

[0130] Based on the topology and component parameters, a corresponding transient simulation model was built in Simulink. The most severe inrush current condition (5Ω / 30mF, RC time constant) was used. The simulation model was used as a benchmark for verification. Under two conditions, with conversion times of 10ms and 20ms respectively, the four-channel waveforms (DC bus voltage, DC side current, AC input voltage, and AC input current) obtained from the simulation were recorded. The results are as follows: Figure 7-10 As shown.

[0131] Simulation results show that at a 10ms transition time, due to the extremely short power-off time and the relatively high residual capacitor voltage, the closing voltage difference is relatively small, and the inrush current peak is relatively mild. However, at a 20ms transition time, the residual capacitor voltage has significantly decayed, the closing voltage difference has increased, and the inrush current peak has significantly increased. The voltage drop depth, peak current inrush value, and waveform decay rhythm under both operating conditions are highly consistent with the analysis and expectations based on RLC transient circuit theory, verifying the accuracy of the simulation model in terms of capacitive load transient response.

[0132] In step S3, the configurable load array contains 24 independently controllable RC branches, with an equivalent total resistance of approximately 5Ω and a total capacitance of approximately 30mF when all are engaged; the capacitance levels cover four typical operating conditions: 30mF, 13.2mF, 0.52mF, and 0.3mF.

[0133] This step establishes a variable-parameter capacitive load. Due to the diverse types and wide range of capacitance parameters in low-voltage networks, an experimental platform with flexibly configurable capacitance values ​​is needed to simulate actual operating conditions with varying capacitive strengths. The system uses a three-phase four-wire power supply as input, which sequentially passes through a disconnector, molded case circuit breaker, single-phase circuit breaker, and automatic transfer switch, before connecting to a capacitive load consisting of three sets of parallel bridge rectifiers and a switchable RC array. By varying the capacitance value between 0.3mF and 30mF to select various capacitive operating conditions, and using transition times ranging from 10ms to 1000ms for step-by-step scanning, the peak AC / DC inrush current and impact coefficient are recorded for each combination. These results are then compared with measured results under the same operating conditions to complete the secondary verification of the simulation model.

[0134] Step S4: System scanning and pattern extraction of inrush flow characteristics under multi-capacity operating conditions, such as... Figure 11 As shown, the details are as follows:

[0135] Inrush data acquisition: After confirming the accuracy of the inrush dimension in the model, the power outage time was gradually extended from 10ms to 70ms, and the peak inrush current was recorded for each power outage duration. The measured and simulated data are summarized in Table 3, which shows the simulated inrush current data of a 5Ω / 30mF load under different power outage times.

[0136]

[0137] Table 3

[0138] Note: Impact factor = peak inrush current / rated operating current; the steady-state rated current in the table is 90.93A (measured) / 98A (simulation) for AC side and 92.83A / 94.56A for DC side. The largest deviation between simulation and measurement under the five conversion times appears in the DC side impact factor under the 10ms condition, with a relative error of about 5.6%. The deviations for the other groups are between 3% and 7%, and the overall deviation does not exceed 8%.

[0139] Full-matrix inrush current characteristic scan: A full-matrix scan was performed on four typical loads using a simulation model verified by actual measurements. The automatic transfer switch switching times were set to nine operating points: 10ms, 15ms, 20ms, 30ms, 70ms, 200ms, 500ms, 700ms, and 1000ms. The four loads were 5Ω / 30mF, 5Ω / 13.2mF, 5Ω / 0.52mF, and 5Ω / 0.3mF, respectively. The results are plotted on [the graph / plot]. Figure 11 The figure divides the horizontal axis into three intervals using two vertical lines, 20ms and 70ms, respectively, marking the operating ranges of HTSE, QTSE, and traditional automatic transfer switches.

[0140] In step S4, the capacitance parameter is adjusted from 0.3mF to 30mF, and the conversion time is set from 10ms to 1000ms.

[0141] This step involves system scanning and data analysis. Based on the impact coefficient-time curves of each load group, a critical time threshold is determined using the inflection point identification method of the impact coefficient-transition time curve. The incremental changes in the impact coefficient are compared between adjacent transition time levels. When the increment shows a significant jump within a certain time interval, the start time of that interval is the critical inflection point for inrush growth. Simultaneously, the load type is classified as strong or weak capacitive based on whether the impact coefficient curve flattens after 20ms—if the curve enters the steady-state region after 20ms, it is determined to be a weak capacitive load; if the curve continues to rise after 20ms, it is determined to be a strong capacitive load.

[0142] Step S5, Differentiated switching control strategy formulation and switch selection, is as follows:

[0143] Slow capacitive load switching scheme: For devices with input filter capacitors of 0.52mF or less (such as low-power auxiliary power supplies, signal sensors, etc.), the inrush current enters a steady-state platform and has a low absolute amplitude 20ms after power failure. The minimum switching delay of the automatic transfer switch is set to 20ms to skip the initial oscillation range of the inrush current. This operating condition is equipped with a high-speed transfer switch (HTSE) (switching time ≤20ms). The lower limit of the HTSE switching delay is set slightly higher than one half-cycle of the power frequency (i.e., >10ms) to avoid the amplification effect of the inrush current at extremely unfavorable closing phase angles.

[0144] Short-term power failure switching solution for highly capacitive loads: For devices with input filter capacitors of 13.2mF and above (such as high-power ticketing systems, communication control hosts, etc.), the capacitors retain a considerable residual voltage when the power failure time does not exceed 70ms (taking a 30mF load as an example, the residual voltage is still 63% of the initial value after 70ms). The natural buffering effect of the residual voltage keeps the inrush current within a controllable range (impact coefficient of about 4.5~7.7). A fast transfer switch QTSE (switching time 20ms~70ms) is selected to complete the switching before the inrush current inflection point arrives.

[0145] Long-term power outage switching scheme for highly capacitive loads: After a power outage exceeding 70ms, the inrush current has passed the inflection point and entered a rapid growth range. The residual capacitor voltage decreases significantly or even drops to zero. At the moment of closing, a strong inrush current with extremely high amplitude and extremely fast rise rate will be generated. At this time, fast-acting or high-speed switches, due to their compact contact structure and relatively limited short-time withstand current, are at risk of contact welding, insulation breakdown, or spring mechanism failure. In such cases, a traditional automatic transfer switch with larger contact spacing, more arc-extinguishing grids, and a higher rated short-time withstand current (usually 15 to 20 times the rated value) should be used to take over the closing.

[0146] For intermediate load sections: For intermediate load sections with capacitance values ​​between 0.52mF and 13.2mF, the inflection point of the inrush current curve can be calculated based on the actual RC time constant, and the selection can be made by referring to the above three-level strategies.

[0147] In step five, the steady-state time of inrush current for weak capacitive loads is 20ms, and the inflection point of the inrush current growth curve for strong capacitive loads is 70ms. Weak capacitive loads are configured with HTSE (transition time ≤ 20ms), strong capacitive loads are configured with QTSE (transition time 20ms~70ms) for short-time power outages, and strong capacitive loads are configured with conventional automatic transfer switches (rated short-time withstand current higher than that of fast-acting switches) for long-term power outages.

[0148] The control functions, simulation models, and strategy parameters in each of the above steps all adopt standardized interface design and are seamlessly integrated with the Simulink toolbox. The switching time and action voltage threshold can be modified online according to actual engineering needs to adapt to the switching behavior of different types of automatic transfer switches.

[0149] The specific implementation of the present invention is as follows:

[0150] 1. First, the switching performance of the automatic transfer switch timing control model was verified through power outage switching simulation. A power outage was manually set at t=0.6s, during which the voltage rapidly dropped to zero; this zero-voltage phase corresponds to the switch switching time. After setting a switching delay of 50ms, the voltage rapidly recovered to the rated operating voltage of 220V within 0.65s, with the dynamic response precisely matching the preset delay. FFT spectrum analysis was performed on the recovered A-phase voltage, and the results are as follows... Figure 5 As shown, the fundamental amplitude is 230.1V, the THD is only 0.32%, the main harmonic components are the 3rd, 5th and 7th harmonics, which decrease in order. The harmonic levels of each order are far below the national standard limits, and the voltage quality after compensation fully meets the standards. Figure 3 This serves as a comparative reference point when the load voltage drops directly to zero after a main power outage without an automatic transfer switch. Figure 4 This is to ensure that the system voltage returns to the rated range after the automatic transfer switch is engaged.

[0151] At this point, the timing control model and simulation platform of the automatic transfer switch have been verified through simulation. The verification object is an inductive load (frequency converter controlled motor). The purpose is to confirm the switching timing and voltage response accuracy of the automatic transfer switch model itself, so as to lay a reliable model foundation for the subsequent study of inrush current characteristics when capacitive loads are connected.

[0152] 2. Comparison of capacitive load inrush current simulation and experiment: based on the experimental principle of capacitive load. Figure 6 Based on the load parameter configuration in Table 2, a transient simulation model was built in Simulink. The most severe inrush current condition of 5Ω / 30mF was used as the baseline (RC time constant). =150ms), two conversion times, 10ms and 20ms, were selected for synchronous comparison between simulation and experiment. The waveform is as follows. Figure 7-10 As shown, the waveforms in each graph, from top to bottom, represent DC voltage, DC current, AC voltage, and AC current. Simulation results show that the longer the power outage time, the deeper the residual voltage decays, the greater the potential difference between the power supply side and the capacitor terminal when the circuit is closed, and the higher the inrush current peak value. This trend is completely consistent with the exponential decay-recharge model based on RLC transient circuit theory. Based on this, the power outage time was gradually extended from 10ms to 70ms, and the simulation data are summarized in Table 3.

[0153] 3. Inrush Current Characteristic System Scanning Analysis and Differentiated Strategy Formulation: Using a simulation model verified by actual measurements, inrush current scanning analysis was conducted on four typical capacitive loads: 5Ω / 30mF, 5Ω / 13.2mF, 5Ω / 0.52mF, and 5Ω / 0.3mF. By adjusting the switching time of the automatic transfer switch (at nine operating points: 10ms, 15ms, 20ms, 30ms, 70ms, 200ms, 500ms, 700ms, and 1000ms), the AC side inrush current response was collected, and the variation law of the inrush coefficient was plotted. Figure 11 The curves corresponding to the four load groups in the figure are marked with black (30mF), red (13.2mF), blue (0.52mF), and green (0.3mF), respectively. At the same time, the horizontal axis is divided into three segments according to the two time threshold lines of 20ms and 70ms, which correspond to the working range of HTSE, QTSE and traditional automatic transfer switch, respectively.

[0154] The parameters are explained as follows:

[0155] (1) Inrush current coefficient: defined as the ratio of peak inrush current to rated operating current. This coefficient directly measures the impact of the closing transient on the switch contacts and protection settings. The larger the coefficient, the further the inrush current deviates from the rated value, and the greater the threat to the switch, protection and load-side devices.

[0156] (2) RC time constant and capacitor residual voltage: =RC determines the rate at which the voltage across the capacitor decays after power is cut off. The larger the value (i.e., the larger the capacitance or resistance), the longer the residual voltage remains after a power outage. During short power outages, the closing voltage difference is small and the inrush current is gentle. However, if the power outage time exceeds a certain threshold... As residual pressure decays more rapidly, the inrush flow will increase rapidly.

[0157] (3) Inrush current growth inflection point: The position on the impact coefficient curve of a strong capacitive load where the curvature changes most significantly. Before the inflection point, the inrush current growth is gradual, and after the inflection point, the inrush current accelerates. The power outage time corresponding to this inflection point is the key dividing line for differentiated strategies. The key simulation parameters are shown in Table 4.

[0158] Table 4

[0159] The specific analysis steps are as follows:

[0160] Step 1: Inrush current characteristics analysis of weakly capacitive loads, corresponding to Figure 11 The two curves, one blue (0.52mF) and one green (0.3mF), show the impact factor of a weakly capacitive load, which almost stops changing after the transition time exceeds 20ms, exhibiting a pattern of rapid initial convergence followed by a constant state. The time constant of the small capacitor is also shown. =RC is very short (5Ω × 0.52mF = 2.6ms). After power failure, the residual voltage is discharged to near zero within tens of milliseconds. After that, no matter how long the wait is, the port voltage difference when closing the circuit is locked near the full amplitude of the supply voltage, and the inrush current will naturally not fluctuate significantly. Moreover, the smaller the capacitance value, the lower the peak current that the charging circuit can draw from the line, so the 0.3mF curve is always below the 0.52mF curve. In the range where the switching time is less than 20ms, the fluctuation of the impulse coefficient is due to the modulation effect of the closing phase angle on the inrush current when the capacitor has not been fully discharged. After 20ms, the residual voltage returns to zero, and this modulation effect disappears.

[0161] Step 2: Inrush current characteristic analysis of highly capacitive loads, corresponding to Figure 11 The two curves, one black (30mF) and one red (13.2mF), show the surge coefficient of the highly capacitive load continuously increasing, but the rate of increase significantly accelerates at 70ms. During the short-term power outage (<30ms), the large capacitor has sufficient energy storage, a long discharge time constant, and the residual voltage remains at a considerable level, resulting in a small closing voltage difference and a mild inrush current. In this range, the inrush current of the 30mF load is actually lower than that of the 13.2mF load. This is because a larger capacitor can maintain a higher residual voltage and leave less voltage difference space for the inrush current in the same short time. Once the power outage exceeds 70ms, the residual voltage collapses significantly faster, and the voltage drop upon closing approaches the full amplitude. This, combined with the large transient current peak that the large capacitor can handle, pushes the inrush current up a steep slope, causing an observable jump in the curve's slope at 70ms. After this inflection point, the 30mF curve rises faster than the 13.2mF curve, and the difference between the two gradually widens, eventually converging to their respective asymptotic values ​​under long-term power outage conditions.

[0162] Step 3: Formulate differentiated switching control strategy. Using 20ms and 70ms, two time thresholds verified by actual testing, and following the configuration logic of using HTSE for weak capacitive loads, QTSE for strong capacitive short-term power outages, and traditional automatic transfer switches for strong capacitive long-term power outages, the inrush current impact generated by each type of capacitive load within the actual power outage duration falls within the safety envelope of the configured switch.

[0163] Step 4: Conclusion Analysis. By comparing the inrush current coefficient variation curves of four sets of loads with different capacitance parameters under nine switching times and the simulated inrush current data of the 5Ω / 30mF condition (Table 3), the following conclusions are drawn: The established automatic transfer switch timing control function model has been verified by both simulation and actual measurement. The simulated waveform and the measured waveform are in good agreement in terms of voltage drop rate, zero-position duration, and steady-state value after recovery. The load voltage THD after switching is only 0.32%, far below the national standard limit of 5%. The maximum relative error between the simulated and measured inrush current data under the five switching times does not exceed 8%. The surge coefficient of weakly capacitive loads (0.52mF and below) tends to a steady state after a power outage exceeding 20ms, with a relatively low absolute surge current amplitude. For strongly capacitive loads (13.2mF and above), the surge coefficient increases non-linearly with the power outage time, with a sharp inflection point at 70ms. For example, with a 5Ω / 30mF load, when the power outage time increases from 10ms to 70ms, the AC surge coefficient rises from 3.63 to 7.69, and the peak surge current increases from 330A to 700A, an increase of approximately 112%. Configuring an HTSE for weakly capacitive loads can shorten the power outage time to less than 20ms; configuring a QTSE for short-term power outages of strongly capacitive loads utilizes the residual voltage buffer effect of capacitors to achieve rapid power restoration; configuring a traditional automatic transfer switch for long-term power outages of strongly capacitive loads can avoid contact welding or malfunction of protection devices due to insufficient short-term withstand current in fast-acting switches.

[0164] In step 1, the inrush current steady-state time threshold for weak capacitive loads is adjustable; in step 2, the inrush current inflection point position for strong capacitive loads is determined based on the actual RC time constant; in step 3, the conversion time range for each segment can be finely adjusted according to the actual engineering situation.

[0165] In summary, the differentiated switching control method disclosed in this invention addresses the inrush current problem generated by capacitive loads during automatic transfer switch switching in low-voltage power supply systems. It constructs a complete technical route encompassing automatic transfer switch timing modeling → dual-track verification through simulation and experimental testing → comprehensive scanning of inrush current characteristics → formulation of differentiated strategies.

[0166] 1. An automatic transfer switch timing control function model with the symbol function as the core was established. The switching action process was discretized into three quantifiable state modes: main power supply, zero-position blocking and backup power supply. This made the switching timing in the simulation a continuously adjustable research variable. After dual simulation and experimental verification with inductive and capacitive loads, the model accurately reproduced the dynamic characteristics of the actual system.

[0167] 2. Through systematic simulation scanning of four sets of loads with different capacitance parameters under nine switching times, the key difference between the inrush current of weak capacitive load tending to a steady state after more than 20ms and the inrush current of strong capacitive load showing a jump in the growth rate inflection point at 70ms was quantitatively revealed, providing a quantifiable criterion for inrush current prediction and switch selection of capacitive loads.

[0168] 3. Based on the above principles, a differentiated switching control strategy with three threshold levels (20ms / 70ms) is proposed. This strategy matches three types of switches—HTSE, QTSE, and traditional automatic transfer switches—to loads with different capacitive parameters, achieving a balance between suppressing inrush current and shortening power outage time. This method can directly serve the engineering design and operation and maintenance decisions of low-voltage power supply systems.

[0169] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. An automatic transfer switch selection control method based on inrush flow characteristic zoning, characterized in that, Includes the following steps: S1. Modeling of timing control functions for automatic transfer switching appliances; S2. System-level simulation platform construction and switching performance verification; S3, Simulation and experimental cross-validation of inrush current during capacitive load closing; S4. System scanning and pattern extraction of inrush flow characteristics under multi-capacity operating conditions; S5. Differentiated switching control strategy formulation and switch selection.

2. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 1, characterized in that, Step S1 is as follows: S11. Construct the timing control function model for the automatic transfer switch: The automatic transfer switch includes a smart contactor QS1 located on the main power supply side and a smart contactor QS2 located on the backup power supply side. The main power supply is stepped down by transformer T1 and the backup power supply is stepped down by transformer T2 and then connected through QF1 and QF2 respectively. Each downstream load branch is equipped with an independent protection circuit breaker QF3~QFN. S12. Motion time decomposition and model simplification: The total operating time of the automatic transfer switch is It encompasses the entire process from detecting the main power supply deviation to completing the dual power supply switching, and its time structure is broken down into three parts: ； in, The inherent mechanical action time for contact switching. Set a delay manually. To account for the computation time of the voltage detection algorithm, the model will include... and Unified zeroing process, i.e. ; S13. Mathematical modeling of the control function of automatic transfer switching appliances: Based on the action time decomposition, the control function of the automatic transfer switch is established, and its mathematical expression is as follows: ； in, The instantaneous value of the main power supply voltage. The operating voltage threshold. This is the real-time delay calculated from the moment QS1 is disconnected. The zero-position transition time, i.e. The symbolic function sgn is defined as: ； S14, Modal Partitioning and Transition Logic: According to the control function The output value of the automatic switching device is divided into three modes based on its operating state: when When QS1 is closed and QS2 is open, the system is in mains power supply mode; when When QS1 and QS2 are both disconnected, the system is in a zero-position locked state. when At this time: QS1 is open and QS2 is closed, the system is in standby power supply mode; The transition process between the three modes mentioned above is as follows: When the main power supply is normal , When QS1 is closed and QS2 is open, the load receives power from the main power supply. Fall to the following, When the signal changes from +1 to 0, QS1 disconnects, and the system then enters a zero-bit transition phase. Not yet counted Previously, the controller did not execute. The corresponding QS2 closing command forces the system to maintain the zero-position state of QS1 and QS2 open, pending... Full After the lock is released, When the value changes from 0 to -1, QS2 closes, and the backup power supply is officially connected to the load. S15. Power quality assessment after switching: After the automatic transfer switch completes the switching process, a voltage quality assessment is performed on the load side, as detailed below: The total harmonic distortion (THD) is calculated by extracting the fundamental component amplitude V1 from the voltage waveform using a fast Fourier transform. ； Among them, V h The amplitude of the h-th harmonic is required to have a THD of no more than 5%.

3. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 2, characterized in that, The interlocking station of the intelligent contactors QS1 and QS2 is a three-position station structure.

4. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 1, characterized in that, Step S2 is as follows: S21. Simulation model architecture construction; The simulation model is built in Simulink as follows: Both the main power contactor module Q1 and the backup power contactor module Q2 are driven by a common control bus. The bus signal generates a first subsystem and a second subsystem through two paths. The first subsystem collects the three-phase voltage from the main circuit, and feeds it into the MATLAB Function module after gain scaling. The second subsystem performs the same operation on the backup circuit. The two signals are combined and enter the controller module. In conjunction with the simulation clock, the controller module executes the operation logic of the control function of the automatic switching device, and finally outputs three modes of drive signals (+1 / 0 / -1) to control the on and off of the contactor. S21. Inductive load simulation verification; By setting the port voltage of the inductive load as the observation object, a power-off switching simulation is performed on the model: Without the intervention of an automatic transfer switch, the voltage of the inductive load drops directly to zero after the main power supply is interrupted. When the automatic transfer switch is engaged, and the transfer delay is set, the system voltage recovers to the rated range within a preset time.

5. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 4, characterized in that, In step S21, the inductive load is set as a variable frequency drive motor, and the port voltage of the variable frequency drive motor is the object of observation.

6. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 1, characterized in that, Step S3 is as follows: S31. Construction of capacitive load experimental system; With a three-phase four-wire power supply as the input, the circuit passes through a disconnecting switch, a molded case circuit breaker, a single-phase circuit breaker, and an automatic transfer switch in sequence, and is connected to the capacitive load terminal consisting of three sets of parallel bridge rectifiers and a switchable RC array. S32, Actual cross-validation; By changing the capacitance value at the capacitive load end and scanning step by step with the conversion time, the peak value of the AC / DC inrush current and the impact coefficient under each combination are recorded one by one, and compared with the measured results under the same working conditions to complete the secondary verification of the simulation model.

7. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 1, characterized in that, Step S4 is as follows: S41, Inrush data acquisition; After confirming the accuracy of the inrush dimension in the model, the power outage time was gradually extended from the lowest value to the highest value, and the peak inrush flow was recorded at each power outage time. S42, Full-matrix inrush characteristic scan; Using a simulation model verified by actual measurements, a full matrix scan was performed on four typical loads. The incremental changes of the impact coefficient between adjacent transition time levels were compared one by one. When the increment showed a significant jump within a certain time interval, the starting time of that interval was the critical inflection point of the inrush growth. At the same time, the strong and weak capacitive loads were classified according to whether the impact coefficient curve tended to flatten after 20ms. If the curve entered the steady state interval after 20ms, it was determined to be a weak capacitive load; if the curve continued to rise after 20ms, it was determined to be a strong capacitive load.

8. The automatic transfer switch selection control method based on inrush flow characteristic zoning according to claim 1, characterized in that, Step S5 is as follows: Based on the analysis in step S4, key inrush flow characteristics are obtained, and a differentiated switching control strategy is formed accordingly: Slow capacitive load switching scheme: For devices with input filter capacitors of 0.52mF and below, the inrush current enters a steady-state platform and the absolute amplitude reaches the preset threshold 20ms after power failure. The minimum switching delay of the automatic transfer switch is set to 20ms to skip the initial oscillation range of the inrush current. In this case, the high-speed transfer switch HTSE is selected. Short-term power failure switching solution for highly capacitive loads: For devices with input filter capacitors of 13.2mF and above, the capacitors retain a considerable level of residual voltage when the power failure time does not exceed 70ms. The natural buffering effect of the residual voltage keeps the inrush current within a controllable range. In this case, a fast transfer switch (QTSE) is selected, and the switching is completed before the inrush current inflection point arrives. Long-term power outage switching scheme for highly capacitive loads: After the power outage time exceeds 70ms, the inrush current has passed the growth inflection point and entered the rapid growth range. In this case, a traditional automatic transfer switch is selected to take over the closing. For intermediate load sections: For intermediate load sections with capacitance values ​​between 0.52mF and 13.2mF, the inflection point of the inrush current curve can be calculated based on the actual RC time constant, and the above three-level strategies can be used for classification and selection.

Images