Analysis of DC system grounding detection method and improvement of detection device

      The DC system of power plants and substations is an independent power supply, which is not affected by generators, plant power, station transformers and changes in system operation mode. It provides reliable and stable uninterrupted power supply for the control circuit, signal circuit, relay protection, automatic device and emergency lighting of the power system. It also provides operating power for the opening and closing of circuit breakers. Therefore, the DC power supply system plays a vital role in the safe operation of the substation and is the prerequisite for the safe operation of the substation. The DC power supply system in the power system adopts the ground insulation operation mode. When a point of grounding occurs in the DC system, it does not cause direct harm and can continue to operate, but the potential danger is very large. An alarm must be given immediately and searched; otherwise, when another point of grounding occurs, there may be very serious consequences.

1. About the grounding of DC system and its hazards

1.1 What is grounding of DC system

    DC power supply is a power supply with polarity, that is, the positive pole and the negative pole of the power supply. The "ground" of DC power supply is just a concept of neutral point for DC circuit. If the insulation resistance between the positive or negative pole of the DC power supply system and the ground is reduced to a certain set value, or lower than a certain specified value, then we say that the DC system has a positive grounding fault or a negative grounding fault.

1.2 Why is the DC system grounded?

    The DC system of power plants and substations is connected to many devices and complex circuits. In the long-term operation process, due to changes in the environment, climate changes, aging of cables and connectors, problems with the equipment itself, etc., DC system grounding is inevitable. Especially during the construction or expansion of power plants and substations, due to various problems in construction and installation, it is inevitable that hidden dangers of DC system grounding faults will remain.

1.3 Classification and hazards of DC system grounding

    Due to the complex connection of the DC system feeder network, it can be divided into positive grounding and negative grounding according to grounding polarity; according to the type of grounding, it can be divided into direct grounding (also known as metal grounding or full grounding) and indirect grounding (also known as non-metal grounding or semi-grounding); according to the grounding situation, it can be divided into single-point grounding, multi-point grounding, loop grounding and insulation reduction. According to research, positive grounding may cause the circuit breaker to trip incorrectly. Since the circuit breaker tripping coils are all connected to the negative power supply, when positive grounding occurs, it may cause the circuit breaker to trip. Negative grounding may cause the circuit breaker to refuse to trip.

    Operation practice also found that DC grounding will not only cause relay protection to malfunction or fail to operate, but even cause DC-controlled equipment to malfunction or fail to operate, and even damage the equipment, resulting in serious consequences such as large-scale power outages and system collapse. For example, on July 6, 2000, in heavy rain, a DC grounding occurred in a 220kV hub substation in Hebei. The 273-1 electric knife switch automatically opened during operation without any signal. After investigation and analysis, it was found that the 273-1 knife switch control box was damp due to water seepage, and it was confirmed that the knife switch was malfunctioned due to DC two-point grounding. In a 500kV hub substation, due to heavy snow, the DC two-point grounding caused the station's 380V AC power control switch to trip. In a 220kV hub substation in Guangxi, the circuit breaker malfunctioned due to DC two-point grounding. These are all power grid faults caused by DC system insulation problems.

2. Common methods for finding and eliminating DC system faults

There are many detection technologies for ground faults in DC systems, and the implementation principles are also different. They can be summarized into the following methods:

2.1 DC bus bridge method

The detection device developed using the bridge method is relatively simple, which is equivalent to adding two balancing resistors to the positive and negative DC bus to form a balanced bridge; it only gives an alarm for the ground fault of the DC system, and cannot indicate the straight line and grounding resistance value where the fault is located, and the function is too monotonous. When field maintenance personnel troubleshoot, they usually use the manual pulling method. Pull open the DC load circuits supplied by the DC panel in turn for a short time. When a certain circuit is cut off, the fault disappears, indicating that the fault is in the circuit. The operability is relatively poor, especially for important loads, short-term power off is not allowed. Therefore, the detection device using this method is only suitable for DC cabinet systems in very low-end distribution rooms.

2.2 Low-frequency signal injection method

    After a ground fault occurs in the DC system, a low-frequency signal is injected between the fault bus and the ground. The low-frequency current flows out of the signal generator, flows through the ground fault feeder, and returns from the grounding point. Use a clamp-type current detector to detect each feeder. Find the grounded feeder and then find the grounding point.

    This method successfully achieves the search for DC grounding points without power outages, but its detection accuracy and sensitivity are greatly affected by the distributed capacitance of the DC system. The maximum capacitance of the feeder branch can reach several microfarads. When the probe is measured at a certain point, due to the presence of capacitive current, it will be difficult for the operator to determine whether it is capacitive current or grounding resistance current. Using this method to detect grounding resistance often results in misjudgment or large errors in the measured and calculated grounding resistance.

2.3 Frequency conversion signal injection method

Frequency conversion signal injection is actually still low-frequency signal input, but the injected signal is a low-frequency signal with alternating frequency. Then, the clamp-type current probe is used to detect the change in the branch resistive current amplitude to determine the grounding branch and the fault point. By injecting a signal with constant amplitude and variable frequency, the resistive current in the feeder branch is indirectly calculated. However, through on-site use inspection, the effect is still not ideal, and the reason is still distributed capacitance. In addition, the injection of low-frequency signals will increase the voltage ripple factor of the DC system.

2.4 Magnetic modulation DC leakage current detection method 

    Using magnetic modulation leakage current sensor, the positive and negative lines of the DC system feeder branch pass through the CA ring core, and the current provided to the load, there is a triangular wave constant current excitation current in the CA leakage current sensor, excitation winding, detection winding; when the current on the positive and negative lines is not equal, the leakage current sensor has a leakage current size direction signal output. Due to the principle of balanced bridge, the magnetic modulation leakage current sensor can only monitor asymmetric DC grounding faults, and is powerless when the positive and negative insulation resistances drop equally or their values are close DC screen