I. Overview In power systems, with the widespread use of inductive loads such as transformers and AC motors, a large amount of inductive reactive current frequently flows through the power supply and distribution equipment. This reactive current occupies a significant portion of the equipment's capacity and increases line transmission current, thus increasing feeder line losses and preventing the full utilization of power equipment. One solution is to use reactive power compensation devices to compensate for reactive power locally, minimizing or eliminating the need for additional equipment capacity and improving equipment utilization efficiency. The most common method is to use capacitor banks to provide capacitive current to compensate for inductive current, thereby improving the power factor. Currently, low-voltage capacitor centralized automatic compensation devices are widely used in distribution systems to compensate for low-voltage reactive power locally as needed. However, in high-voltage systems, traditional fixed capacitor compensation devices are still more commonly used, while centralized automatic compensation devices are not yet widely adopted. Because traditional compensation methods suffer from poor safety performance, low compensation accuracy, and high labor intensity, there is a strong demand for a more reliable, accurate, and automated compensation device for design selection. Since 1995, we have been piloting 6kV high-voltage capacitor automatic compensation devices in projects such as the second phase of the Tianjin Economic and Technological Development Area stormwater and sewage pumping station, the Donghai Road stormwater and sewage pumping station, the Taifeng Road stormwater pumping station, and the Yueyahe stormwater pumping station in Tianjin. After several years of use, it has been proven that the power factor after compensation reaches over 0.95, the level of automation is high, and the compensation effect is satisfactory. It has received unanimous praise from all users. This article, based on the engineering application experience, provides a brief introduction to the relevant technical issues of the high-voltage capacitor centralized automatic compensation device, hoping to stimulate further discussion. II. Compensation Implementation Plan and Determination of Compensation Capacity To achieve ideal compensation results, it is essential to first determine a reasonable compensation implementation plan and accurately calculate the capacity to be compensated. Currently, common compensation methods include the traditional fixed capacitor bank manual fuse insertion and removal control compensation capacity method; the single-unit random local capacitor compensation method; and the centralized capacitor automatic compensation method. Among these, the traditional compensation method is simple, but the compensation accuracy is low, the labor intensity is high, the risk is high, and it is too susceptible to human factors. The single-unit on-site compensation method compensates for individual equipment locally. Its advantage lies in compensating from the point of demand, reaching the primary location of compensation needs, and covering a wide range. Its disadvantage is the difficulty in determining the compensation capacity. Overcompensation must be avoided, and LC resonance and self-excitation must be prevented in the circuit. When calculating reactive current, the main components are the motor excitation current I0, the reactive current increment ID1 under full load, and the reactive current increment ID2 under underload. Because these parameters constantly change with the motor's operating state, there are too many dynamic variables, making it difficult to accurately determine the reactive current compensation requirement. Different production equipment has different starting capacity margins when selecting motors, resulting in varying motor saturation levels and different reactive current increments ID2 under underload. Furthermore, the actual operating state of the motor changes constantly; for example, the load rate of a water pump motor changes constantly with changes in the inlet and outlet water levels, making it impossible to determine the accurate operating conditions. The local compensation method for single-unit equipment, once the compensation capacity is determined, uses a fixed compensation capacity to balance the constantly fluctuating dynamic operating conditions, making it difficult to achieve satisfactory high-precision compensation results. Furthermore, in single-unit compensation capacitor devices, the compensation capacitor is fixedly installed one-to-one with the main unit. The compensation capacitor starts operating simultaneously with the main unit and is disconnected when the main unit stops operating. The capacitors between units are independent and cannot complement each other, resulting in underutilization of the capacitors and increased equipment investment. Moreover, municipal engineering projects are characterized by concentrated operating times, large equipment capacities, and even lower utilization rates of standby equipment. Furthermore, because the compensation capacitor starts operating simultaneously with the main unit, the main unit's starting current and the capacitor's inrush current are at their maximum values ​​simultaneously, and the sum of these two maximum currents amplifies the inrush current effect. If a centralized automatic compensation method for grouped equipment is adopted, the compensation capacity can be automatically activated according to the overall operating conditions at the time, achieving higher compensation accuracy. The shorter the step size of the compensation equipment, the higher the compensation accuracy. If the step size is infinitely variable, the power factor can theoretically be accurate to 1, laying the foundation for high-precision and accurate compensation. Furthermore, regardless of which motor is operating, the compensation capacitors can be put into operation according to the overall needs of the line, ensuring full utilization of each set of compensation capacitors. III. Compensation Equipment Step Length Division and Equipment Configuration Although theoretically, the compensation accuracy of a stepless automatic compensation device can reach 1, in practical applications of general municipal engineering, to rationally utilize limited funds, the theoretical maximum value is not required; meeting the engineering accuracy requirements is sufficient. Therefore, in most cases, multiple devices operate in parallel. When there are four or more devices, dividing the required maximum compensation capacity into 6-8 equal-step capacity inputs can basically meet the actual accuracy requirements of the project. Similar to the commonly used low-voltage capacitor automatic compensation devices, the use of an 8-step equal-capacity input scheme is very common, and its theory can be extended to high-voltage capacitor compensation devices. However, in high-voltage systems, if the low-voltage compensation approach is followed, equal-capacity configuration can still be used for schemes controlled by high-voltage vacuum contactors. For cases using vacuum circuit breakers, because vacuum circuit breakers are relatively expensive, minimizing the number of vacuum circuit breakers used while ensuring the same functionality has a significant effect on saving investment. In engineering projects, the number of vacuum circuit breakers can be reduced by selecting the appropriate controller. For example, for equipment groups using capacitor banks with equal-step capacity distribution, 7-step compensation requires 7 vacuum circuit breakers. However, if a 1+2+4 unequal-capacity controller configuration is used, only 3 vacuum circuit breakers are needed to achieve the effect of 7-step equal-step capacity compensation, in the form of 1, 2, 1+2, 4, 4+1, 4+2, 4+2+1. This ensures compensation accuracy while significantly saving on initial equipment investment. IV. Protection and Control In addition to conventional protection and control, the protection and control of high-voltage capacitor automatic compensation devices have some special considerations. Some issues that are easily overlooked during the design and commissioning of protection systems, encountered in actual engineering projects, are briefly introduced here. In practical engineering projects, 7-8 steps of control are generally used depending on the number of motors. In addition to conventional protection against overvoltage and overcurrent, the protection system must employ comprehensive three-phase protection to avoid protection failure and fault expansion caused by single-phase faults. Appropriate configuration of inrush current limiting reactors is crucial to strictly prevent damage caused by electromagnetic resonance. In addition, the protection system must pay attention to the voltage superposition problem on the capacitors when they are automatically switched on. When a group of capacitors is taken out of operation, it must be fully discharged before being put back into operation. This ensures that the residual voltage on them is reduced to within the allowable voltage range upon re-switching, preventing overvoltage damage caused by the superposition of residual voltage and rated voltage. Secondly, in the control system, special attention needs to be paid to the correct phase configuration of the working power supply, signal power supply, and other detected quantities. Correct vector configuration is a strong guarantee and minimum requirement for smooth equipment commissioning; otherwise, it will bring unnecessary trouble and increase the workload, sometimes even preventing the correct conclusions from being reached. The design of the control system varies slightly depending on the components used. For example, the contactor of the compensation device: if an electromagnetic vacuum contactor is used, the on/off state is a single 1-0 signal; if a mechanical contactor or a vacuum circuit breaker is used, the on/off state must be two independent signals. Both control methods have their advantages and disadvantages, leading to different conclusions from different perspectives such as energy saving and noise reduction. Opinions vary. The design can adopt economical, reasonable, practical, and technologically advanced equipment configurations according to the specific conditions of the project. The control principle when using mechanical contactors or vacuum circuit breakers is shown in the "Control Principle Diagram of Automatic Capacitor Compensation Device". V. Conclusion Currently, most widely used compensation methods are stepped compensation. To achieve more accurate compensation, the ideal compensation method is a stepless automatic compensation method with microcomputer control. This method can achieve the most ideal accuracy of the compensated power factor as needed. However, due to its high technical content, investment cost is relatively high, and it also requires a high level of management and maintenance. In summary, the capacitor compensation method can be selected according to the specific conditions of the project. Under the premise of meeting the compensation accuracy requirements, priority should be given to equipment with low project cost, reliable operation and management, and convenient maintenance. Therefore, adopting a stepped automatic compensation method with unequal capacity configuration is one of the economical, reasonable, and feasible solutions.