[b]0 Introduction[/b] Cross-linked polyethylene (XLPE) cables have seen rapid development in recent years due to their good insulation performance, ease of manufacturing, and convenient installation. With the implementation of urban and rural power grid upgrades, the utilization rate of power cables will increase significantly. Therefore, it is essential to maintain and utilize existing power equipment effectively and improve power supply reliability. The operating condition of cables directly affects the safe operation of the power system and the reliability of power supply. In the past, my country widely used preventative testing methods, which involved periodic power outages, a form of offline testing. However, with the development of power supply, this traditional method of power outage testing is increasingly unable to meet the actual needs of power production and supply. Therefore, researching online monitoring technology for power cables is crucial for timely and reasonable maintenance, repair, and replacement of cables, ensuring reliable cable operation. In recent years, many researchers have proposed new live-line inspection methods, which are very effective in detecting early signs of cable insulation degradation. 1. Non-energized testing method for power cable performance With the development of urban construction, power cables have become increasingly important in urban power grid supply, and have gradually replaced overhead transmission lines in some urban areas; at the same time, with the increase in the number of cables and the extension of operating time, cable faults have become more and more frequent. Due to the concealment of cable lines, the incomplete operating data of individual operating units and the limitations of testing equipment, it is very difficult to find cable faults[1]. Power cable faults can be divided into two types according to their nature: series (broken line) faults and parallel (short circuit) faults. The latter can be divided into main insulation faults (with metal shielding) and outer sheath (outer sheath) faults (without metal shielding) according to whether there is a metal sheath or shielding outside the insulation. Main insulation faults can be divided into three types according to the insulation resistance Rf of the fault point: ① metallic short circuit (low resistance) faults, where Rf varies depending on the instrument and method selection, generally Rf<10 Z0 (Z0 is the cable wave impedance); ② high resistance faults; ③ intermittent (flashover) faults. There are no absolute boundaries between the three; the distinction is mainly based on the field testing methods and is related to the equipment's capacity and internal resistance. In the past decade, my country's urban power grids have widely adopted XLPE power cables. Based on cable faults, various testing methods exist both domestically and internationally. 1.1 Bridge Method and Low-Voltage Pulse Reflection Method: Before the 1970s, the bridge method and low-voltage pulse reflection method were widely used worldwide for power cable fault testing. Both are accurate for low-resistance faults but not suitable for high-resistance faults. Therefore, they are often combined with the burning resistance reduction (burn-through) method, which involves increasing the current to burn through the fault, reducing its insulation resistance so that the bridge method or low-voltage pulse method can be used for measurement. The burn-through method has adverse effects on the cable's main insulation and is now rarely used. 1.2 High-Voltage DC Flash Test Method and Impulse Flash Test Method: These methods test intermittent faults and high-resistance faults respectively. Both can be further divided into current flash test methods and voltage flash test methods, with different sampling parameters and each having its advantages and disadvantages. The voltage sampling method has a high measurability rate, clear and easy-to-interpret waveforms, and a dead zone half that of the current method. However, the wiring is complex, and excessive voltage division poses a danger to personnel and instruments. The current sampling method, on the other hand, is simple to wire but suffers from significant waveform interference, making it difficult to identify and resulting in a large blind zone. These two methods are currently the mainstream methods used in domestically produced high-resistance fault testers, primarily from manufacturers such as Xi'an Sifang, Shandong Kehui, and Wuhan High Voltage Research Institute. High-voltage current and voltage flash measurement methods have largely solved the problem of high-resistance cable faults and are widely used in my country's power sector, with extensive experience in their application. However, these instruments have blind zones, and the waveforms are sometimes not clear enough, requiring manual judgment, which sometimes fails. The accuracy and error of these instruments are relatively large. 1.3 The Secondary Pulse Method: This testing technique emerged in the 1990s. Because low-voltage pulses are accurate and easy to use, combined with high-voltage generator-based impulse flashover technology, a low-voltage pulse is triggered by an internal device at the moment the arc ignites at the fault point. This pulse undergoes a short-circuit reflection at the flashover point (where the arc resistance is very low), and the waveform is stored in the instrument. After the arc extinguishes, a normal low-voltage measurement pulse is sent to the cable. This pulse does not create a path at the fault point (high resistance) and reaches the cable end directly, where it undergoes an open-circuit reflection. By comparing the two low-voltage pulse waveforms, the location of the fault point (breakdown point) can be easily determined. The instrument can automatically match and calculate the distance to the fault point. The advent of the secondary pulse method has made high-resistance cable fault testing very simple, becoming the most advanced testing method. For the secondary pulse method, the principles of products from both Baur (Austria) and Seba (Germany) are the same, differing only in implementation: the former emphasizes the coordination of arc initiation and trigger pulses, using an internal communication device to dampen the inrush current and increase its width; while the latter employs a specialized arc stabilizer, emphasizing extended arc time to ensure the low-voltage pulse arrives during arc initiation. Compared to domestically produced high-voltage current or voltage testers, this method has the following advantages: ① Integrated design, compact structure; simply connect to power, ground, and the cable under test to perform various testing methods; simple wiring, easy switching, and safe and reliable. ② High degree of automation, achieving automatic matching, automatic protection, automatic judgment, and automatic calculation; it can also print or save graphs to a floppy disk for computer data analysis. ③ No blind zone: Considering the length of the instrument's feeder and the external high-voltage cable lead, a "tm" test is introduced during instrument debugging. First, the time "tm" it takes for the pulse wave to travel through the instrument to the end of the lead in each method is measured and input into the memory system. When testing the cable, the instrument automatically sets the origin (starting point) at the "tm" time of that method. Because "tm" is a fixed value and independent of the wave velocity selection, the time "tm" for the pulse to travel through the instrument and lead remains constant regardless of the wave velocity selected. The "tm" time point in the measured waveform is the beginning of the tested cable, thus eliminating the concept of a blind zone during measurement. ④ High accuracy: Using the Baur IRG300 echo meter with a sampling frequency of 200 MHz, and a wave velocity of 160 m/μs, the accuracy reaches 0.4 m. Due to the high degree of automation, accuracy, and ease of operation of this instrument, it overcomes the shortcomings of current and voltage impulse methods, effectively solving the difficulties of high-resistance fault testing. As long as the wave velocity is selected correctly, the measurement results are very accurate. [b]2 Methods for Live-Line Testing of Power Cable Insulation Performance [2-4][/b] Currently, research on live-line testing methods is widely conducted both domestically and internationally, resulting in the proposal of various methods. In actual operation, it has been found that most power cable faults are caused by insulation degradation. There are many reasons for this cable degradation (including electrical degradation, thermal degradation, chemical degradation, mechanical degradation, and even degradation caused by rodents and insects), but electrical degradation is the most significant. Its main degradation forms are: ① partial discharge degradation; ② electrical treeing degradation; ③ water treeing degradation. Studies have shown that in solid-insulated cables below 33 kV, water treeing degradation is the primary cause of insulation degradation. However, regardless of the type of degradation, it can lead to a decrease in insulation resistance, an increase in leakage current, and a larger dielectric loss (tgδ). This results in a larger AC loss current under operating voltage, increasing the DC component in the current flowing through the insulation. Therefore, online monitoring of cable insulation can be used to determine degradation signals and whether the cable insulation can continue to operate. The degradation signals of cable insulation are generally extremely small. For example, the DC component current generated by dendritic degradation is in the nA range, and the largest is only in the μA range. Therefore, based on extensive research on the fundamental physical processes of degradation of polymer insulation materials, foreign countries have studied and adopted corresponding monitoring measures for degradation signals. There are many methods for online monitoring of cable insulation, such as the DC current method, DC voltage superposition method, AC voltage superposition method, low-frequency AC superposition method, etc. 2.1 DC Current Method When water tree degradation occurs in a cable under AC voltage, the current contains a DC component, and there is a certain relationship between the length of the tree degradation and the DC component current. Therefore, the DC current component monitoring method is used. However, since the DC component current is extremely small (generally in the nA range), it is easily affected by stray currents. Moreover, when the leakage resistance on the cable end surface decreases due to contamination or rain, the measurement error is large. Therefore, the end must be cleaned and the measurement must be performed in good weather, which greatly limits the use of this method. 2.2 DC Voltage Superposition Method The DC voltage superposition method was studied to investigate the relationship between the length of water trees and the insulation resistance in cables. The DC voltage superposition method suffers from significant measurement errors due to variations in stray current or decreased leakage resistance at the cable end. Furthermore, the DC voltage is applied to the cable via a neutral-grounded voltage transformer; if a DC power supply flows through the transformer for an extended period, magnetic saturation can occur, generating zero-sequence voltage, potentially causing malfunctions in substation relays. 2.3 Low-Frequency AC Superposition Method: This method investigates the relationship between water tree length and insulation resistance in cables. It is a relatively good method, as the monitored AC loss current theoretically increases with cable degradation. However, careful verification of the cable end's operating condition is crucial. For example, if stress rings are installed to adjust the electric field distribution at the end, even with good cable insulation, a large AC loss current may lead to a misjudgment of "poor insulation" based solely on online monitoring signals. 2.4 AC Voltage Superposition Method The measurement principle of the AC voltage superposition method is as follows: a 101 Hz AC voltage (i.e., twice the power frequency + 1 Hz) is superimposed on the cable's shielding layer, and a 1 Hz degradation signal is generated by monitoring tree dendrites. Since the degradation signal is strongest when a power frequency + approximately 1 Hz voltage is superimposed on a tree-degraded cable, this method can be used to detect the strength of the 1 Hz degradation signal to determine the degree of cable degradation. The advantages of this monitoring method are: ① Voltage can be superimposed from the cable grounding wire, making measurement simple and convenient. It can be used not only for online monitoring but also for live monitoring, allowing one set of equipment to monitor multiple cables; ② Because the superimposed voltage detects a known degradation signal, i.e., the 1 Hz signal, the detection accuracy is high and the anti-interference ability is strong; ③ It is less affected by factors such as armor insulation resistance and end contamination. 3 Conclusion Through the above analysis and comparison, we can find that among non-energized detection methods, the secondary pulse method is a relatively good method, while among energized detection methods, the AC voltage superposition method is currently a relatively good method. Although the methods for live-line testing are not yet mature, such as the judgment of the degree of insulation degradation, a lot of research is still needed. However, this is a development direction for power cable testing. References: [1] Xu Binggen, Li Shengxiang, Chen Zongjun. Power cable fault detection technology [M]. Beijing: Machinery Industry Press, 1999. [2] DEIS. The invention of chemically crosslinked polyethylene [J]. IEEE Electrical Insulation Magazine, 1999, 15 (1): 23-25. [3] Nakayama T. On line cable monitor developed in Japan [J]. IEEE on Power Delivery, 1991, 6 (4): 1359-1365. [4] Yuan Jinling. Power cable fault finding [J]. Hebei Electric Power Technology, 1995, 14 (3): 19-29