1. Introduction For a long time, China's 6-10kV medium-voltage distribution network has used either an ungrounded neutral point or a grounded neutral point via an arc suppression coil. In recent years, with the continuous expansion of the medium-voltage distribution network capacity and the large-scale replacement of overhead lines with cable lines in urban distribution networks (the ground capacitance of cable lines is approximately 30 times that of overhead lines), the ground capacitance current of the power grid has increased significantly. When the neutral point is ungrounded, the ground capacitance current will flow through the entire system at the grounding point, causing insulation damage to non-faulty phases and resulting in phase-to-phase short circuits, posing a significant threat to the power grid. In a neutral-point grounding system using an arc-suppression coil, the total current passing through the grounding point during a single-phase ground fault is the vector sum of the inductor current and the total system capacitance current, i.e., [img=16,23]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/54-2.jpg[/img]. However, because the ground capacitance current of the power grid varies with the power grid's operating mode and is also affected by the environment, to effectively compensate for the ground capacitance current, the inductance value of the arc-suppression coil should be adjusted according to the changes in capacitance current to achieve optimal compensation. The automatic tracking compensation device introduced in this paper can calculate the capacitive current in real time based on the measured value using the corresponding algorithm, and adjust the arc suppression coil through the automatic control system, thereby reducing the grounding residual current to within a safe range. 2. Composition of the Device The automatic tracking compensation arc suppression device consists of two main parts: the arc suppression coil body and the control system. The arc suppression coil body includes a grounding transformer, an adjustable arc suppression coil, and a series resistor. The control part consists of a microcontroller and related chips. 2.1 Arc Suppression Coil Part The arc suppression coil part consists of a grounding transformer, an arc suppression coil, and a series resistor, as shown in Figure 1. [img=220,263]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/54-3.jpg[/img] Under normal circumstances, a single-phase ground fault in an ungrounded system without an arc suppression coil is prone to causing other faults, such as phase-to-phase short circuits, etc., but the unbalanced voltage during normal operation is generally small. After the arc suppression coil is installed, the grounding current decreases when a single-phase short circuit occurs, but the system unbalance voltage under normal operating conditions often increases. The system unbalance voltage (i.e., neutral point displacement voltage) under normal operating conditions after the arc suppression coil is installed is [img=370,203]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/54-4.jpg[/img] where the three phases of the power grid are grounded. For the arc suppression coil grounding system, it is generally desirable for the detuning degree V to be as small as possible. However, as can be seen from formula (1), after the arc suppression coil is installed, as the detuning degree V decreases, the neutral point displacement voltage U_N often increases and may exceed the safe range. Therefore, it is necessary to install a series resistor to reduce the neutral point displacement voltage under normal conditions. When a series resistor exists, the unbalanced voltage of the power grid under normal operating conditions is: [img=335,46]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/55-1.jpg[/img] In equation (2), 3C0 is the total capacitance of the line to ground. By comparing equation (1) and equation (2), it can be seen that by connecting a series resistor with an appropriate resistance value, the neutral point displacement can be controlled within the specified range of 15%. When a single-phase ground fault occurs in the system, the series resistor can be cut off by the control system to achieve a better compensation effect. The series resistor is made of enameled nickel-chromium resistance wire, and the resistance value is stable. If the damping resistor is too large, the current of the regulating circuit will be too small during normal operation, which will not meet the requirements of the microcomputer input signal, reduce the regulating sensitivity, and increase the grounding residual current. 2.2 Hardware and Software Design of Automatic Control System 2.2.1 Working Principle of Automatic Control System a) Monitoring of Neutral Point Displacement Voltage In the device, the change of UN is used to determine whether there is a single-phase grounding. When the change of UN exceeds a certain value ηU?, the control system will automatically measure the capacitor current. However, the value of UN is affected by many factors (such as changes in insulation leakage, changes in the parameters in equation (1), etc.). In this system, it is adopted to collect data at fixed time intervals, analyze its changes, and measure the capacitor current if it is determined that the capacitance to ground has changed according to the measured value. b) Real-time Measurement of Capacitor Current When it is determined that a single-phase grounding has occurred based on the change value of UN, the capacitor current should be measured in real time and the arc suppression coil should be adjusted accordingly to implement full compensation as much as possible. In this device, the neutral point displacement current phase angle method is used to measure the capacitor current. The gear is adjusted once during system initialization, and the capacitor current is calculated according to the measured value. The current Ii is measured at the initial setting of the arc suppression coil, and the current phase angle is θi. The setting is then adjusted again to measure the current value Ii+1, and the current phase angle is θi+1. The phase voltage UΦ is also measured, where Φ is the angle between the current and the voltage after the setting is adjusted once. The two measured phase angles are substituted into θ=θi+1-θi to calculate the phase difference between the two measurements [1]. The Fourier short data window algorithm is used in the system to obtain the phase signal value. To achieve accurate calculation, the following calculation programs are all written in C and nested in assembly language. [img=336,136]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/55-3.jpg[/img] [img=341,42]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/55-2.jpg[/img]  The capacitor current [WTBX]I[WTBZ]C is calculated using formula (5) based on the input quantity.  The detuning degree is calculated using formula (6).  c) Compensation for ground capacitance current  When it is determined to be a single-phase ground fault, the microcontroller control system will immediately disconnect the series resistor and adjust the arc suppression coil's position based on the capacitance current value obtained from the above calculation. Then measure the capacitor current again. If the calculated capacitor current value is the same, continue adjusting the arc suppression coil. Repeat this control process until the capacitance current to ground is reduced to below 5A. 2.2.2 Automatic Control System Hardware Structure (as shown in Figure 2) The system hardware structure mainly consists of four parts: a) Data Processing Unit The data processing unit is the core part of the microcomputer protection device. Its main components include CPU, memory, clock circuit, etc. In this system, the 80C196KC is used. In this system, the 196 chip and the XILINX 9572 chip together constitute the core of the system and complete most of the functions such as data temporary storage, logical judgment, timing separation and bus driving. The memory DS1644 (RAM) and 27256 (EPROM) are used to store programs, sampled data, intermediate calculation results and settings. EPROM is used to store programs and settings, and RAM is used to store sampled data, intermediate calculation results and other values ​​that need to be accessed at any time. The DS1644 used in this system is a clock chip with non-volatile SRAM (including a crystal oscillator and lithium battery). The highest 8 bytes of the 32KB RAM store time information, including year, month, day, weekday, hour, minute, second, and control bits, in BCD code format. The watchdog chip used in the system is X5043, which has an optional watchdog overflow timer and 4kbits of EEPROM. This EEPROM has write protection; it can only be written to correctly after a write enable command is entered. Once the write operation is completed, the write protection is automatically reset, greatly improving the reliability of the system. b) Analog Input Unit The analog input unit converts the analog current and voltage outputs from the secondary side of the current transformer and voltage transformer into digital quantities that the data processing unit can accept. This includes signal conditioning, low-pass filtering, sample and hold, and A/D conversion. This system uses the AD7891 chip, which features multiplexing and sample-and-hold functionality. It is an 8-channel, 12-bit data acquisition system with parallel output and a single-channel conversion time of 1.6μs, meeting the system's sampling speed requirements. Because the system's external bus uses an 8-bit configuration, while the AD7891 is a 12-bit parallel output, there is a bus width conversion issue. Considering the negative impact of frequent bus width conversions, it was decided to keep the system bus width constant. Each time the AD7891 is read, the lower 8 bits are read directly, while the higher 4 bits are latched in the CPLD. These 4 bits are then read from another address (the decoding circuitry used is implemented internally within the CPLD). [img=447,199]http://zszl.cepee.com/cepee_kjlw_pic/files/wx/jdq/2002-2/56-1.jpg[/img] c) Switch Output Unit The switch output unit mainly outputs the processed output from the microcomputer in the required form to perform a control function. Because it may introduce external interference into the device, the relay and the microcomputer system need to be optically isolated and use an independent power supply. The expansion function of the output ports in the system is integrated into the CPLD and connected to the address/data bus inside the CPLD. d) Communication Unit: The keyboard, monitor, parallel port, serial port, and printer mainly serve as the interface for human-machine interaction, acting as a bridge for operators to control and monitor the device's operating status. They can also communicate with the communication management unit via the serial port. The keyboard is directly connected to the CPU's P1 port as an external interrupt. The display uses an MGLS12864T LCD module with a 128×64 dot matrix, capable of displaying text, graphics, and text-graphic composite modes. The font size is 8×8. The LCD is a low-speed device, requiring a wait cycle when the CPU accesses it. The device uses the MAXIM RS-485 opto-isolated interface chip MAX1480B. A single +5V power supply on the logic side is sufficient to power both sides of the interface via a DC/DC converter. Communication is half-duplex with a transmission rate of up to 250kbps, and includes a transmission rate limiting circuit and overload protection. 2.2.3 Automatic Control System Main Program Flow (as shown in Figure 3) [b]3 Conclusion[/b] Through principle explanation and related tests, it is shown that this device can monitor the neutral point displacement voltage and perform real-time measurement and compensation of the capacitor current, ensuring the system always operates in optimal condition. While enhancing the system's overvoltage capability, it significantly reduces the adjustment operations of the original arc suppression coil, further improving operational safety and reliability, and reducing the workload of operation and dispatch personnel. It can greatly reduce the power frequency reactive current at the fault point and reduce the probability of a single-phase ground fault developing into a phase-to-phase short circuit.