Compensation methods for fluid flow reactive power compensation capacitors

Abstract

This invention discloses a compensation method for flow-current reactive power compensation capacitors, relating to the field of reactive power compensation technology. The method includes the following steps: acquiring grid data, calculating the required compensation capacity, and determining the target group of capacitors to be switched; dividing the target group of capacitors into several subgroups according to capacity or voltage level, with each subgroup containing several capacitor units; based on a digital twin model, performing time-sharing switching sequence planning for the subgroups to obtain an optimized switching safety sequence; gradually establishing terminal voltages for each subgroup in the switching safety sequence to smoothly connect it to the bus; sequentially switching each subgroup and its internal capacitor units according to the switching safety sequence, while simultaneously monitoring the bus voltage and switching current in real time, until the reactive power compensation connection of the entire target group is completed. This method solves the problems of large transient impacts and complex control logic during the switching of flow-current reactive power compensation capacitors.

Description

Technical Field

[0001] This invention relates to the field of reactive power compensation technology, and more specifically, to a compensation method for a liquid flow reactive power compensation capacitor. Background Technology

[0002] In existing power systems, oil-immersed parallel capacitors are commonly used for reactive power compensation to improve the power factor and reduce reactive power losses. Oil-immersed parallel capacitors have advantages such as mature structure, stable capacity, and long lifespan, and are widely used in high-voltage, large-capacity distribution networks. Existing compensation methods typically connect capacitor units to the busbar through centralized or group switching to achieve reactive power compensation for the target load, thereby maintaining grid voltage stability and ensuring the power factor meets expected requirements. Under normal operating conditions, this method can effectively improve grid power quality and reduce line losses.

[0003] However, in practical applications, the switching of oil-immersed parallel capacitors can generate significant inrush current and transient voltage fluctuations. Especially under high voltage, large capacity, and group switching conditions, this impact can exert considerable stress on the capacitors and their control equipment, potentially leading to dielectric breakdown, switch damage, or excessive grid transients. Since traditional switching methods cannot effectively control these transient processes, safety and reliability risks exist.

[0004] Soft-start technology has been widely used in motor starting, power electronic equipment, and other high-voltage, high-power applications to limit starting current or control surge current. By adjusting the switching speed or introducing current limiting measures, peak current and voltage overshoot can be reduced during transient processes, thereby protecting equipment and improving system stability. These technologies provide valuable insights for transient control of capacitor switching, helping to mitigate the impact problems caused by high-voltage, high-capacity switching.

[0005] While soft starters offer good buffering performance in other fields, their application to the switching control of oil-immersed parallel capacitors faces significant obstacles. The main problem is that implementing soft starters increases the complexity of the switching control logic, requiring precise coordination of the switching sequence and intervals of multiple capacitor units and subgroups, while also considering real-time changes in bus voltage and load. This complex control logic is difficult to automate and ensure safety in traditional control systems.

[0006] To address the above problems, this invention proposes a solution. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a compensation method for a liquid flow reactive power compensation capacitor. This method utilizes time-sharing switching sequence planning and soft-start control based on digital twins to address the problems of large transient impacts and complex control logic during switching of the liquid flow reactive power compensation capacitor.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] The compensation method for liquid flow reactive power compensation capacitors includes the following steps: acquiring grid data, calculating the required compensation capacity, and determining the target group of capacitors to be switched; dividing the target group of capacitors into several subgroups according to capacity or voltage level, with each subgroup containing several capacitor units; based on a digital twin model, performing time-sharing switching sequence planning for the subgroups to obtain an optimized switching safety sequence; gradually establishing terminal voltages for each subgroup in the switching safety sequence to smoothly connect them to the bus; sequentially switching each subgroup and its internal capacitor units according to the switching safety sequence, while simultaneously monitoring the bus voltage and switching current in real time, until the reactive power compensation connection of the entire target group is completed.

[0010] In a preferred embodiment, the steps of acquiring grid data, calculating the required compensation capacity, and determining the target group of capacitors to be switched are as follows: The grid data includes real-time voltage, real-time current, and real-time power factor; a discrete Fourier transform is performed on the grid data to extract the complex spectrum of the fundamental frequency component, obtaining the fundamental effective value and fundamental phase difference. The fundamental effective value includes the fundamental voltage effective value and the fundamental current effective value, and the fundamental phase difference includes the fundamental phase angle difference between the voltage phasor and the current phasor; the fundamental active power is calculated based on the fundamental effective value and the fundamental phase difference using a three-term balance; the real-time phase angle is calculated inversely based on the real-time power factor; a target power factor is set, and the corresponding target phase angle is calculated inversely; the real-time power factor is adjusted to the target power factor, and the required reactive power baseline deficit is calculated by combining the target phase angle and the real-time phase angle, which serves as the target compensation capacity; based on the target compensation capacity, available capacitor units are searched in a preset capacitor bank library, and candidate capacitor target groups are selected according to the minimum overshoot priority capacity matching principle; the target group of capacitors to be switched is determined based on mixed integer optimization matching of the combination with the closest capacity to the target compensation capacity.

[0011] In a preferred embodiment, dividing the target capacitor group into several subgroups according to capacity or voltage level, with each subgroup containing several capacitor units, specifically involves: establishing a capacitor unit attribute matrix based on the rated capacity and voltage level of each capacitor unit within the target capacitor group; clustering the capacitor unit attribute matrix using a hierarchical clustering method, grouping capacitors with capacity and voltage level differences within a preset first range into the same subgroup to obtain a first division result; and optimizing the first division result based on the operating state parameters of the capacitor units, eliminating capacitor units that do not meet the operating conditions to obtain a second division result.

[0012] In a preferred embodiment, the subgroups are time-division switching sequence planning to obtain an optimized switching safety sequence. Specifically, this involves: establishing an initial model of the subgroup and bus-power source equivalent parameters, initializing the subgroup equivalent capacitance and front-end voltage state; updating and adjusting the initial model parameters based on grid measurement data, adjusting the initial state of the subgroups, and obtaining a transient prediction model; performing transient prediction on each subgroup gradually connected to the bus based on the transient prediction model, and predicting the subgroup switching response data; adjusting the subgroup switching order based on a constraint optimization method, adjusting the subgroup connection interval according to capacity and electrical distance, balancing the impact of each subgroup switching on the bus, and outputting a switching safety sequence; and sequentially switching each subgroup and its internal capacitor unit according to the switching safety sequence.

[0013] In a preferred embodiment, the transient prediction model is used to progressively predict the switching response data of each subgroup as it is connected to the bus. Specifically, this involves: equating each subgroup to a series impedance and equivalent capacitance, and using the bus and bus-power source as the subgroup's equivalent impedance; using the front-end voltage state after the last switching as the initial value of the subgroup's front-end voltage; establishing the transient network relationship between the bus and the subgroup based on the series impedance, equivalent capacitance, and subgroup equivalent impedance to obtain the initial state equation; adjusting the impedance and capacitance parameters of the transient model based on real-time grid data; updating the bus recovery time constant based on the small-amplitude response data of the last switching; obtaining the updated initial state of each subgroup before its connection; progressively connecting each subgroup to the bus according to a preset subgroup switching order; calculating the bus voltage change, switching current, and inrush current peak value based on the transient model in each step to generate subgroup switching response data; and using the prediction results of each step to update the initial conditions for the next subgroup connection, thus achieving continuous prediction.

[0014] In a preferred embodiment, the constraint optimization method adjusts the subgroup switching sequence, adjusts the subgroup connection interval according to capacity and electrical distance, balances the impact of each subgroup switching on the bus, and outputs a switching safety sequence. Specifically, it defines the subgroup risk characteristic value based on the voltage difference before the bus, the predicted inrush current peak value of the subgroup, and the weighted sum of the subgroup capacitance; arranges the subgroups in ascending order according to the subgroup risk characteristic value to obtain a preliminary switching sequence; for electrically adjacent subgroups, if the distance between them is less than a preset first lower limit, the first switching time is delayed in the sorting; after satisfying the electrical distance constraint, the capacity optimization is performed on the subgroup connection time in the sequence so that the bus connection capacity does not exceed a set second upper limit in any time period; the preliminary sequence is input into the transient prediction model to calculate the bus voltage change, switching current, and inrush current peak value; if the prediction result exceeds the safety constraint, the sequence order is adjusted; the subgroup order corresponding to the adjusted switching time forms the switching safety sequence.

[0015] In a preferred embodiment, the step of gradually establishing the terminal voltage of each subgroup in the switching safety sequence to smoothly connect it to the bus specifically involves: obtaining the target time and initial voltage state of the current subgroup to be connected in the switching safety sequence; establishing a terminal voltage adjustment model based on the target time and initial voltage state; gradually adjusting the subgroup terminal voltage through a controllable series impedance before connection to generate a voltage trajectory that gradually approximates the bus voltage; performing layered connection within the subgroup according to unit capacity and electrical distance to gradually establish terminal voltage consistency within the subgroup; introducing a closed-loop control law during the voltage adjustment process to correct the adjustment process based on real-time voltage difference and current feedback; closing the series impedance channel when the voltage difference is less than a preset connection threshold to smoothly connect the subgroup to the bus; and updating the bus state and subgroup state after connection as the initial conditions for subsequent subgroup terminal voltage adjustment.

[0016] In a preferred embodiment, before integration, the subgroup terminal voltage is gradually adjusted using a controllable series impedance to generate a voltage trajectory that gradually approximates the bus voltage. Specifically, this involves: obtaining the target time and initial terminal voltage state of the subgroup to be integrated in the switching safety sequence; establishing a terminal voltage adjustment model using the subgroup's equivalent capacitance and series impedance, wherein the terminal voltage adjustment model characterizes the dynamic mapping of the subgroup terminal voltage over time; generating a terminal voltage trajectory that gradually approximates the bus voltage based on the terminal voltage adjustment model, with each step corresponding to a target terminal voltage at a time node; after each time step of terminal voltage adjustment, comparing the actual terminal voltage with the bus voltage, calculating the corresponding switching current increment, and applying it to the subgroup; when the subgroup terminal voltage reaches the bus voltage, the subgroup integration operation is completed; and using the final terminal voltage state of the integrated subgroup as the initial condition for the next subgroup terminal voltage adjustment.

[0017] In a preferred embodiment, the step of stratifying and integrating the capacitors within a subgroup according to their unit capacity and electrical distance to gradually establish voltage consistency within the subgroup specifically involves: reading the initial terminal voltage, rated capacity, and electrical location of each capacitor unit within the subgroup, and using the overall terminal voltage trajectory of the subgroup as a reference target for internal integration; stratifying the capacitor units within the subgroup according to their capacity and electrical distance, starting with the unit with the largest terminal voltage and capacity, and gradually expanding to units with smaller capacity and farther locations; calculating the target terminal voltage for each unit in each layer based on the overall terminal voltage trajectory of the subgroup; gradually applying adjustments to make the unit's terminal voltage consistent with the target terminal voltage within the layer; and updating the voltage state of the unit after each adjustment for use in the calculation of the next layer of units.

[0018] In a preferred embodiment, the step of sequentially switching each subgroup and its internal capacitor units according to the switching safety sequence, while simultaneously monitoring the bus voltage and switching current in real time, until the reactive power compensation connection of the entire target group is completed, specifically involves: selecting the current subgroup and its internal capacitor units to be switched according to the generated switching safety sequence; sequentially operating each capacitor unit for switching, while simultaneously collecting bus voltage and switching current data in real time; comparing the real-time monitoring data with a preset safety threshold to determine whether the switching process is normal; until all subgroups and their internal capacitor units are switched, realizing the reactive power compensation connection of the entire target group; and updating the final state of the bus and each subgroup to provide initial conditions for subsequent operation and adjustment.

[0019] The technical effects and advantages of the compensation method of the fluid flow reactive power compensation capacitor of the present invention are as follows:

[0020] 1. This invention applies soft-start technology in the switching of liquid-immersed parallel capacitors to achieve gradual establishment and closed-loop regulation of subgroup terminal voltage, enabling each subgroup and its internal capacitor unit to smoothly connect to the bus. This effectively suppresses transient impacts during switching, ensures the smoothness of compensation operation and the stability of bus voltage, and improves the safety and reliability of system operation. It solves the problem that the compensation method of oil-immersed parallel capacitors may generate large current impacts and overvoltages at the moment of switching, leading to dielectric breakdown and the risk of switching element losses.

[0021] 2. This invention plans the switching sequence based on a digital twin model and generates a safe switching sequence by combining transient prediction and constraint optimization. This makes complex control logic quantifiable and automated, enabling each subgroup to be switched sequentially in an optimized order. At the same time, the bus voltage and switching current are monitored in real time. This solves the problem of complex control logic in the practical application of soft-start technology, especially in multi-subgroup and multi-unit switching scenarios, where it is necessary to coordinate the switching sequence, interval, and bus response, which is difficult to achieve efficiently using traditional methods. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the compensation method of the liquid flow reactive power compensation capacitor of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1, Figure 1 The present invention provides a compensation method for the fluid flow reactive power compensation capacitor, comprising the following steps:

[0025] S1, acquire power grid data, calculate the required compensation capacity, and determine the target group of capacitors to be switched on;

[0026] In this embodiment, the steps of acquiring power grid data, calculating the required compensation capacity, and determining the target group of capacitors to be switched on and off specifically include:

[0027] The power grid data includes real-time voltage, real-time current, and real-time power factor;

[0028] Perform a discrete Fourier transform on the power grid data to extract the complex spectrum of the fundamental frequency component, and obtain the fundamental effective value and fundamental phase difference. The fundamental effective value includes the fundamental voltage effective value and the fundamental current effective value, and the fundamental phase difference includes the fundamental phase angle difference between the voltage phasor and the current phasor.

[0029] The active power of the fundamental wave is calculated based on the fundamental wave effective value and the fundamental wave phase difference, using a three-term balance.

[0030] The real-time phase angle is calculated by inverse calculation of the real-time power factor; a target power factor is set, and the corresponding target phase angle is calculated by inverse calculation.

[0031] The real-time power factor is brought to the target power factor, and the required baseline reactive power deficit is calculated by combining the target phase angle and the real-time phase angle, which serves as the target compensation capacity.

[0032] Based on the target compensation capacity, available capacitor units are retrieved from the preset capacitor bank library, and candidate capacitor target banks are selected according to the capacity matching principle of minimum overshoot priority.

[0033] The target group of capacitors to be switched is determined by using a combination of mixed integers to optimize the matching capacity that is closest to the target compensation capacity.

[0034] It should be noted that the following is a feasible example of calculating the fundamental active power:

[0035] ,

[0036] In the formula For the fundamental frequency, This is the effective value of the fundamental voltage. This is the effective value of the fundamental current. This is the phase angle difference.

[0037] It should be noted that the following is a feasible example of calculating the real-time phase angle in reverse:

[0038] ,

[0039] In the formula, The power factor.

[0040] It should be noted that the following is a feasible example of calculating the target compensation capacity:

[0041] ,

[0042] In the formula For real-time phase angle, The target phase angle.

[0043] It should be noted that the purpose of extracting the fundamental component from the power grid data using discrete Fourier transform is to filter out higher harmonics and noise interference, ensuring that subsequent active and reactive power calculations are based on the fundamental component of the power grid.

[0044] It should be noted that the capacitor bank library is typically established based on the rated capacity, rated voltage level, and available unit storage of oil-immersed parallel capacitors, combined with the switching constraints of on-site operation and maintenance, forming a pre-set library. Each capacitor unit serves as the minimum switching unit for subsequent combination selection.

[0045] It should be noted that the capacity matching principle refers to selecting the scheme with the smallest error between the target compensation capacity and the candidate capacitor combination, while giving priority to meeting the constraint of "minimum overshoot". That is, under the premise of meeting the target compensation, try to avoid the combined capacity exceeding the target capacity by too much, so as to reduce the risk of grid voltage fluctuations.

[0046] It should be noted that mixed integer optimization is used to guarantee the global optimal solution in the combination selection of capacitor units. Its decision variable is the switching state of the capacitor unit (takes a value of 0 or 1), the constraint is that the total capacity is close to the target compensation capacity, and the objective function is to minimize the error.

[0047] It should be noted that Discrete Fourier Transform (DFT) extraction of complex spectra refers to converting real-time acquired voltage and current signals into the frequency domain to obtain a complex representation at the fundamental frequency (default 50Hz). The modulus of the complex spectrum corresponds to the amplitude information of the fundamental wave. Combined with the number of sampling points and the sampling frequency, it can be converted into the effective value of the voltage or current. The argument of the complex spectrum represents the phase information of the signal at the fundamental frequency. The following are feasible expressions for the effective values ​​of voltage and current:

[0048] , ,

[0049] In the formula, and These represent the complex spectral moduli of voltage and current at the fundamental frequency, respectively.

[0050] It should be noted that the following are examples of feasible arguments for complex spectra: , The phase angle difference between the voltage phasor and the current phasor is the fundamental phase angle difference, where... and These are the fundamental phase angles of the voltage and current, respectively.

[0051] This embodiment uses Discrete Fourier Transform (DFT) on grid voltage, current, and power factor data to extract the fundamental RMS value and phase angle difference. Combined with the target power factor, it accurately calculates the reactive power deficit, enabling real-time and precise calculation of the compensation capacity of the oil-immersed reactive power compensation capacitor. This method not only ensures that the selection of the target capacitor group to be switched is closer to the dynamic reactive power demand of the grid, avoiding frequent capacitor group operations due to over- or under-compensation, but also provides a quantifiable capacity benchmark for subsequent subgroup division and time-series switching. This fundamentally alleviates the current surge and overvoltage risks during the switching of high-voltage, large-capacity oil-immersed parallel capacitors, while also providing data support for the optimized planning of complex control logic.

[0052] S2, the target group of capacitors is divided into several subgroups according to capacity or voltage level, and each subgroup contains several capacitor units;

[0053] In this embodiment, dividing the target capacitor group into several subgroups according to its capacity or voltage level, with each subgroup containing several capacitor units, specifically involves:

[0054] Establish a capacitor cell attribute matrix based on the rated capacity and voltage level of each capacitor cell in the capacitor target group;

[0055] Based on the hierarchical clustering method, the attribute matrix of capacitor cells is clustered, and capacitors whose capacity and voltage level difference are within a preset first range are divided into the same subgroup to obtain the first division result.

[0056] Based on the operating state parameters of the capacitor cells, the first partitioning result is constrained and optimized, and capacitor cells that do not meet the operating conditions are eliminated to obtain the second partitioning result.

[0057] It should be noted that the operating status parameters include operating temperature. Insulation resistance and historical number of cuts A feasible example of the attribute vector for each capacitor unit is as follows: ,in For rated capacity, For the rated voltage level, combine the attribute vectors of all capacitor cells within the group into a capacitor cell attribute matrix.

[0058] It should be noted that hierarchical clustering uses Euclidean distance to calculate the similarity between units. The clustering criterion is that the difference in capacity and the difference in voltage level both fall within a preset first range. Capacitor units that meet the conditions are grouped into the same subgroup to obtain the first partitioning result.

[0059] It should be noted that the constraint optimization process involves further checking the operating status parameters of each subgroup in the first partitioning result. If the unit's operating temperature exceeds the set threshold, or the insulation resistance is lower than the set threshold, or the number of switching exceeds the upper limit, then the unit is determined to not meet the operating conditions, and such units are removed or their switching is postponed, resulting in the optimized second partitioning result.

[0060] It should be noted that the capacitor cell attribute matrix is ​​not a simple two-dimensional table of capacity and voltage, but rather includes multiple dimensions of operational attributes. For example, in addition to rated capacity and voltage level, it can also include parameters such as temperature monitoring values, insulation resistance, and historical switching counts. The purpose of this is to ensure that the divided subgroups are not only matched in terms of capacity and voltage, but also maintain consistency in operational health, avoiding the imbalance problem of "strong and weak mixed" within a single group.

[0061] It's important to note that the hierarchical clustering method used here determines similarity by calculating the capacitance and voltage differences between capacitor cells, automatically forming subgroups rather than relying on manual experience. This avoids human subjectivity and ensures the stability and repeatability of the partitioning results across different scenarios. The core idea is to "cluster homogeneous capacitors together," making subsequent time-sharing switching easier to achieve balanced electrical behavior.

[0062] It should be noted that "unmet operating conditions" in constraint optimization mainly refer to the risks that switching capacitor cells may pose under the current state. For example, if the cell temperature is too high, continued switching may lead to dielectric breakdown; if the insulation resistance decreases, there is a risk of leakage; if the number of switching operations is too high, the contactor lifespan will be insufficient. By eliminating these cells, the safety and reliability of capacitor bank switching can be significantly improved.

[0063] In this embodiment, by dividing the target capacitor bank into several subgroups according to capacity or voltage level and performing constraint optimization based on operating status parameters, not only can the strong inrush current problem caused by a large capacity being connected at once be avoided during the switching process of the flow-current reactive power compensation capacitor, but also capacitor units in poor health can be eliminated, thereby reducing the risk of dielectric breakdown and failure. Furthermore, this attribute matrix-based and clustering-based partitioning method allows for finer-grained adjustment of the compensation capacity, improving the dynamic accuracy and real-time performance of flow-current capacitor compensation, and effectively resolving the contradiction between safety and accuracy that is difficult to achieve with traditional capacitor banks.

[0064] S3. Based on the digital twin model, time-sharing sequence planning is performed on the subgroup to obtain the optimized safe sequence for switching.

[0065] In this embodiment, the step of planning the time-sharing distribution sequence for the subgroup based on the digital twin model to obtain the optimized distribution security sequence is specifically as follows:

[0066] Establish an initial model of the subgroup and bus-power source equivalent parameters, and initialize the subgroup equivalent capacitance and front-end voltage state;

[0067] The initial model parameters are updated and adjusted based on the power grid measurement data, and the initial state of the subgroup is adjusted to obtain the transient prediction model;

[0068] Based on the transient prediction model, transient predictions are made for each subgroup gradually connected to the bus, and the switching response data of the subgroup is predicted.

[0069] The subgroup switching sequence is adjusted based on the constraint optimization method. The subgroup integration interval is adjusted according to capacity and electrical distance to balance the impact of each subgroup switching on the bus and output a switching safety sequence.

[0070] Each subgroup and its internal capacitor unit are switched sequentially according to the switching safety sequence.

[0071] It should be noted that the transient prediction model is essentially a dynamic model composed of an equivalent circuit model and real-time updated parameters, and is implemented using a reduced-order RLC model.

[0072] It should be noted that the bus-power source equivalent parameters include the equivalent power source voltage and equivalent impedance, which are obtained through offline calculation, online identification, or fitting of historical operating data. The specific method used depends on the power grid scale and monitoring conditions.

[0073] Furthermore, based on the transient prediction model, transient prediction is performed on each subgroup gradually connected to the bus to predict the subgroup switching response data, specifically as follows:

[0074] Each subgroup is equivalent to a series impedance and an equivalent capacitance, and the bus and bus-power supply are the subgroup equivalent impedances.

[0075] Use the front-end voltage state after the last switching as the initial value of the subgroup front-end voltage;

[0076] The transient network relationship between the bus and the subgroup is established based on the series impedance, equivalent capacitance, and subgroup equivalent impedance, and the initial state equation is obtained.

[0077] Adjust the impedance and capacitance parameters of the transient model based on real-time power grid data;

[0078] Update the bus recovery time constant based on the small-amplitude response data from the previous switching operation;

[0079] Obtain the updated initial state of each subgroup before merging;

[0080] According to the preset subgroup switching order, each subgroup will be gradually connected to the busbar;

[0081] During each integration step, the bus voltage change, switching current and inrush current peak value are calculated based on the transient model to generate subgroup switching response data.

[0082] The prediction results of each step are used to update the initial conditions for the next subgroup to be incorporated, thus achieving continuous prediction.

[0083] It should be noted that the following are feasible examples of calculating the initial state equations:

[0084] ,

[0085] In the formula Let K be the switching current of subgroup k. This is the subgroup terminal voltage. Bus voltage This is the equivalent capacitance of the subgroup. This is the equivalent impedance of the subgroup.

[0086] It should be noted that the feasible expression for updating the initial state is as follows:

[0087] ,

[0088] In the formula For the small response data of the last throw, For time constant correction function, The updated initial voltage for the subgroup. This is the initial voltage of the subgroup.

[0089] It should be noted that the following are feasible calculation examples for bus voltage and inrush current peak value: , In the formula The peak inrush flow when subgroup k is switched on. Let be the mapping of the equivalent impedance of the i-th subgroup in the bus network. This is the power supply voltage.

[0090] It should be noted that the series impedance and equivalent capacitance are equivalent parameters calculated based on the rated capacity, voltage level and connection method of the actual capacitor unit within the subgroup. They are not separate formulas, but are aggregated through the equivalent network method. The bus-power supply equivalent impedance is identified by bus measurement data and historical operating data.

[0091] It should be noted that the initial value of the front-end voltage is the steady-state or transient transition state of the subgroup front-end port after the previous switching is completed. It is obtained through measurement or simulation estimation and is used as the initial condition for establishing the next transient model.

[0092] It should be noted that the initial state equation is a differential equation based on the RLC equivalent network, used to describe the dynamic relationship between the bus voltage and the sub-group switching current, but does not involve specific values ​​and is a publicly disclosed model structure.

[0093] It should be noted that real-time power grid data refers to bus voltage, branch current, and power factor data. The equivalent impedance and capacitance of the model are updated through recursive least squares or parameter identification algorithms to improve prediction accuracy.

[0094] It should be noted that small-amplitude response data refers to the step response of the bus voltage under slight switching or disconnection of the subgroup. The bus recovery time obtained by fitting is often used for subsequent transient prediction.

[0095] It should be noted that, according to the preset subgroup switching order, each subgroup is gradually connected to the bus. In each step of connection, the bus voltage change, switching current and inrush current peak value are calculated based on the transient model. The calculation process is a step-by-step iteration, and the output of each step is used as the initial condition for the next step. The bus voltage change, switching current and inrush current peak value are predicted quantities, which are calculated through the transient model, rather than directly measured.

[0096] It should be noted that continuous prediction refers to rolling iteration, which uses the previous step's subgroup switching response data as the input to the next step's model to achieve dynamic sequence planning. This method ensures the consistency of the switching response prediction, but the specific values ​​are obtained by subsequent model simulation or real-time measurement.

[0097] Furthermore, the method of adjusting the subgroup switching sequence based on constraint optimization, adjusting the subgroup integration interval according to capacity and electrical distance, balancing the impact of each subgroup switching on the bus, and outputting a safe switching sequence are as follows:

[0098] The risk characteristic value of a subgroup is defined by weighted summation of the voltage difference before the bus, the predicted peak inrush current of the subgroup, and the subgroup capacitance.

[0099] The subgroups are sorted in ascending order according to their risk characteristic values ​​to obtain a preliminary cutting sequence.

[0100] For electrically adjacent subgroups, if the distance between them is less than a preset first lower limit, the first switching time is delayed in the sorting process;

[0101] After satisfying the electrical distance constraints, capacity optimization is performed on the timing of subgroup integration in the sequence to ensure that the bus integration capacity does not exceed the set second upper limit in any time period.

[0102] The preliminary sequence is input into the transient prediction model to calculate the bus voltage change, switching current and inrush current peak. If the prediction results exceed the safety constraints, the sequence order is adjusted.

[0103] The adjusted order of the subgroups corresponding to the switching times forms the safe switching sequence.

[0104] It should be noted that the following is a feasible calculation example for the first switching moment: In the formula, The minimum interval time, for Subgroup's pitching timing, and These are electrically adjacent subgroups.

[0105] It should be noted that the following is a feasible example of a constraint expression to ensure that the bus's integrated capacity does not exceed a set second upper limit within any given time period: In the formula The second upper limit is P, where P is the total capacity of the busbar connected to the line.

[0106] It should be noted that the risk characteristic value of the subgroup is not a single parameter, but is obtained by weighted summation of multiple factors affecting the safety of switching. The weighting factor can be set empirically or obtained by simulation or fitting historical operating data. The selection of weight values ​​under different power grid operating conditions is a conventional design optimization problem.

[0107] It should be noted that the term "electrically adjacent subgroups" refers to subgroups in the bus topology that have physically close connection points and low equivalent impedance. The quantification of electrical distance can be achieved using the equivalent impedance method or network sensitivity analysis; those skilled in the art can select an appropriate method based on the characteristics of the power grid wiring.

[0108] It should be noted that the capacity optimization process employs linear constraint programming to ensure that the cumulative capacity within any given switching time period does not exceed a preset safety threshold. The specific optimization algorithm can be selected based on the system complexity, such as a fast greedy algorithm, branch and bound method, or convex optimization approximation.

[0109] It should be noted that the transient prediction model is derived from the aforementioned digital twin modeling and real-time parameter identification. Its prediction accuracy depends on the bus recovery time constant, subgroup equivalent parameters, and grid data sampling accuracy. The sequence adjustment after the prediction exceeds the limit is a model-based correction iteration, the basic idea of ​​which is rolling time-domain optimization.

[0110] This embodiment introduces a digital twin model to model and predict the switching process of the flow-current reactive power compensation capacitor subgroups. It enables transient calculations of bus voltage changes, switching current, and inrush current peak values ​​before switching, and combines constraint optimization methods to rationally arrange the switching sequence and connection intervals of the subgroups, thereby generating a safe switching sequence that meets safety constraints. This process effectively avoids the voltage surges and excessive inrush current problems caused by uncontrollable switching transients in traditional reactive power compensation methods, ensuring the smoothness and safety of the compensation action, and solving the switching safety problem of flow-current reactive power compensation capacitors during dynamic compensation.

[0111] S4, gradually establish the terminal voltage for each subgroup in the switching safety sequence, and smoothly connect it to the bus;

[0112] In this embodiment, the step-by-step establishment of terminal voltages for each subgroup in the switching safety sequence to smoothly integrate them into the bus specifically involves:

[0113] Obtain the target time and initial voltage state of the subgroup to be connected in the current switching safety sequence;

[0114] Establish a terminal voltage regulation model based on the target time and initial voltage state;

[0115] Before merging, the subgroup terminal voltage is gradually adjusted by a controllable series impedance to generate a voltage trajectory that gradually approximates the bus voltage.

[0116] Within each subgroup, units are integrated in layers based on unit capacity and electrical distance, gradually establishing terminal voltage consistency within the subgroup.

[0117] A closed-loop control law is introduced during the voltage regulation process to correct the regulation process based on real-time voltage difference and current feedback.

[0118] When the voltage difference is less than the preset connection threshold, the series impedance channel is closed, and the subgroup is smoothly connected to the bus.

[0119] The updated bus status and sub-group status are used as the initial conditions for subsequent sub-group terminal voltage regulation.

[0120] It should be noted that the terminal voltage regulation model is used to describe the dynamic mapping relationship of the sub-group terminal voltage over time. This model is established by taking the initial terminal voltage state of the sub-group to be integrated, the target integration time in the switching safety sequence, and the rated capacity and electrical position of the capacitor units within the sub-group as inputs, combined with parameters such as the sub-group equivalent capacitance, series impedance, and bus-source equivalent impedance. The terminal voltage regulation model is used to generate the trajectory of the sub-group terminal voltage gradually approaching the bus voltage, providing an accurate reference for subsequent smooth integration. The model parameters can be updated online through digital twin methods, simulation calculations, or historical switching response data to ensure that the dynamic mapping reflects the actual power grid characteristics.

[0121] It should be noted that a closed-loop control law is introduced during the terminal voltage regulation process. Based on the difference between the real-time bus voltage and the sub-group terminal voltage, as well as the feedback from the sub-group switching current, the series impedance is dynamically adjusted. The control process iterates in time steps: at each step, the error is calculated based on the actual measured difference between the terminal voltage and the bus voltage, and this error is converted into an adjustment signal for the series impedance, correcting the sub-group terminal voltage to gradually approach the bus voltage. Simultaneously, the increment of the switching current participates in the control as feedback, suppressing inrush current and voltage overshoot, thus achieving continuous closed-loop regulation.

[0122] It should be noted that within a subgroup, units are layered according to their rated capacity and electrical location. The order of integration within each layer begins with the unit with the largest terminal voltage deviation or the largest capacity, gradually expanding towards units with smaller capacities and more distant locations. Each layer calculates its target terminal voltage based on the overall terminal voltage trajectory of the subgroup and applies adjustment via series impedance or adjustable voltage devices to ensure that the terminal voltage of each unit matches the target voltage. After each layer is completed, the voltage state of the units in that layer is updated for use in the calculation of the next layer, achieving terminal voltage consistency within the subgroup.

[0123] It should be noted that the completion of sub-group integration is determined by the difference between the terminal voltage and the bus voltage being less than a preset threshold. The threshold can be set according to the bus capacity, sub-group capacity, or grid characteristics, and can be calculated using absolute voltage difference or weighted difference. When the threshold is reached, the series impedance channel is closed to allow the sub-group to smoothly integrate into the bus, ensuring that the integration process does not cause drastic fluctuations in bus voltage or exceed the inrush current peak limit.

[0124] Furthermore, before integration, the sub-group terminal voltage is gradually adjusted through a controllable series impedance to generate a voltage trajectory that gradually approximates the bus voltage, specifically as follows:

[0125] Obtain the target time and initial terminal voltage state of the subgroup to be connected in the current switching safety sequence;

[0126] A terminal voltage regulation model is established using the subgroup equivalent capacitance and series impedance. This model characterizes the dynamic mapping of the subgroup terminal voltage over time.

[0127] Based on the terminal voltage regulation model, a terminal voltage trajectory that gradually approximates the bus voltage is generated, with each step corresponding to a target terminal voltage at a time node;

[0128] After each time step of the terminal voltage adjustment is completed, the actual terminal voltage is compared with the bus voltage, the corresponding switching current increment is calculated, and applied to the subgroup.

[0129] When the voltage at the subgroup terminal reaches the bus voltage, the subgroup connection operation is completed.

[0130] The terminal voltage state of the subgroup that has been incorporated into the subgroup is used as the initial condition for adjusting the terminal voltage of the next subgroup.

[0131] It should be noted that the following is a calculation example of a feasible voltage regulation model:

[0132] ,

[0133] In the formula, This is the subgroup terminal voltage. Bus voltage This is the equivalent capacitance of the subgroup. For the subgroup series impedance. For controllable injection current, For time steps.

[0134] It should be noted that the following is a feasible example of calculating the terminal voltage trajectory:

[0135] ,

[0136] In the formula, k is the approximation coefficient, used to control the speed.

[0137] Here is a feasible example of calculating the switching current increment:

[0138] .

[0139] Here is a feasible example of calculating the updated terminal voltage:

[0140] .

[0141] It should be noted that when the subgroup terminal voltage reaches the bus voltage, it means that it only needs to meet a certain range, that is, the absolute value is less than the preset error, not that it has to be exactly the same.

[0142] It should be noted that the aforementioned terminal voltage regulation model is not simply a parameter listing, but rather a dynamic mapping established based on the subgroup's equivalent capacitance and series impedance, used to describe the variation of the subgroup's terminal voltage over time. The model includes the continuous relationship of voltage variation over time, used to calculate the target terminal voltage at each time step, ensuring that the voltage change is quantifiable and calculable.

[0143] It should be noted that using the terminal voltage trajectory as a control reference actually involves calculating the deviation between the current terminal voltage and the target terminal voltage at each time step, and calculating the switching current increment accordingly, which is then gradually applied to the subgroup. This process is closed-loop control; after each step, the subgroup terminal voltage state is updated to provide initial conditions for the next calculation, thus achieving continuous adjustment.

[0144] Furthermore, the process of stratifying and integrating subgroups according to unit capacity and electrical distance, and gradually establishing terminal voltage consistency within subgroups, specifically involves:

[0145] Read the initial terminal voltage, rated capacity and electrical position of each capacitor unit in the subgroup, and use the overall terminal voltage trajectory of the subgroup as the reference target for internal integration;

[0146] The capacitor units within the subgroup are layered according to their capacitance and electrical distance, with the layering order starting from the unit with the largest terminal voltage capacitance and gradually expanding to the units with smaller capacitance and farther location.

[0147] For each layer of cells, calculate the target terminal voltage of each cell based on the overall terminal voltage trajectory of the subgroup;

[0148] Gradually apply adjustments to make the cell terminal voltage consistent with the target terminal voltage within the layer;

[0149] After each adjustment, the voltage state of that unit is updated for calculation of the next layer of units.

[0150] It should be noted that the following is a feasible example of calculating the target terminal voltage for each unit:

[0151] ,

[0152] In the formula, Let J be the target terminal voltage of element j at time t. This represents the value of the overall terminal voltage trajectory of the subgroup at time t. For the capacity of cell j, This represents the total capacity of all cells in this layer.

[0153] It should be noted that the overall subgroup terminal voltage trajectory is not an arbitrary curve, but rather a reference for the dynamic changes in subgroup terminal voltage generated in the previous step. This reference guides the gradual integration of internal units, ensuring a smooth transition between the voltage of each unit and the overall terminal voltage. Specifically, the trajectory is quantified into the target voltage value at each time point, serving as the basis for calculating the target terminal voltage of each unit.

[0154] It should be noted that the stratification is not arbitrary, but rather ordered according to unit capacity and electrical distance. Units with larger capacity and closer electrical locations are prioritized for integration, gradually expanding to units with smaller capacity and farther electrical locations. This stratification principle ensures that the voltage regulation sequence of units within each layer is clear, allowing for iterative updates of the state layer by layer and preventing integration shocks.

[0155] It should be noted that the target terminal voltage of each unit is calculated using the overall terminal voltage trajectory of the subgroup, taking into account the actual voltage state and capacity distribution of the preceding units in that layer. The calculation result is a reference voltage value for each unit during the integration process, facilitating closed-loop regulation.

[0156] It should be noted that gradual adjustment means measuring the cell terminal voltage at each time step, calculating the deviation from the target terminal voltage, and then adjusting the switching current or control quantity to minimize the deviation. After each step, the cell voltage state is updated and used as the initial condition for the next layer's calculation, thereby ensuring continuity and consistency between layers and between cells.

[0157] In this embodiment, a time-sharing switching sequence is generated based on a simulation or digital twin model, and each subgroup is switched sequentially according to an optimized order. This makes the complex control logic operable and automated, effectively coordinating the switching order and interval of multiple oil-immersed parallel capacitors during soft start-up. This avoids operational conflicts or switching delays caused by the complexity of the control logic, while ensuring that the switching behavior of each subgroup on the bus is predictable and stable, thereby improving the overall reliability and safety of the hydraulic reactive power compensation system.

[0158] S5, according to the switching safety sequence, each subgroup and its internal capacitor unit are switched sequentially, while the bus voltage and switching current are monitored in real time until the reactive power compensation connection of the entire target group is completed.

[0159] In this embodiment, the step of sequentially switching each subgroup and its internal capacitor units according to the switching safety sequence, while simultaneously monitoring the bus voltage and switching current in real time, until the reactive power compensation connection of the entire target group is completed, specifically involves:

[0160] Select the current subgroup to be switched and its internal capacitor units according to the generated switching safety sequence;

[0161] Each capacitor unit is switched on and off sequentially, while the bus voltage and switching current data are collected in real time.

[0162] The real-time monitoring data is compared with the preset safety threshold to determine whether the switching process is normal.

[0163] Until all subgroups and their internal capacitor units are switched on and off, the reactive power compensation connection of the entire target group is realized.

[0164] Update the final state of the bus and each subgroup to provide initial conditions for subsequent operation and adjustment.

[0165] It should be noted that the real-time monitoring data mentioned above during the switching process mainly includes the bus terminal voltage and the switching current of each subgroup. These data are collected through existing power grid measuring devices and are used to determine whether the switching process exceeds the safety threshold and to trigger necessary protection or correction measures.

[0166] It should be noted that the preset safety threshold refers to the limit value set for the bus voltage fluctuation amplitude and the peak value of the switching current. The threshold can be determined according to the capacitor capacity, bus level and group switching characteristics. This threshold setting method is a mature existing technology.

[0167] It should be noted that the order and method of switching each capacitor unit as described above can be combined with the aforementioned switching safety sequence, or can refer to the control logic of traditional oil-immersed parallel capacitors. The specific operation includes gradually closing or opening the switch to achieve safe access within and between subgroups.

[0168] It should be noted that updating the final status of the bus and each sub-group includes recordings of bus voltage, sub-group terminal voltage, and the status of capacitor units that have been connected to the line. This information is used for reference in the next compensation action or adjustment strategy. In existing technologies, this status update mechanism is usually accomplished through a power monitoring system.

[0169] This embodiment applies soft-start technology during the switching of hydraulic reactive power compensation capacitors to achieve gradual establishment and closed-loop regulation of subgroup terminal voltage, enabling each subgroup and its internal capacitor units to smoothly connect to the bus. This effectively suppresses transient impacts during switching, ensures the smoothness of compensation operations and the stability of bus voltage, and improves the safety and reliability of system operation. It also solves the problem that the compensation method for oil-immersed parallel capacitors may generate large current surges and overvoltages at the moment of switching, which is particularly sensitive for high-voltage, large-capacity, group-switched oil-immersed parallel capacitors. Such surges increase the risk of dielectric breakdown and switching element losses.

[0170] This embodiment uses a digital twin model to plan the time-sharing switching sequence and combines transient prediction and constraint optimization to generate a safe switching sequence. This makes complex control logic quantifiable and automated, enabling each subgroup to be switched sequentially in an optimized order. At the same time, the bus voltage and switching current are monitored in real time. This solves the problem of complex control logic in soft-start technology in practical applications, especially in multi-subgroup and multi-unit switching scenarios, where it is necessary to coordinate the switching sequence, interval, and bus response, which is difficult to achieve efficiently using traditional methods.

Claims

1. A method of compensation for a liquid stream reactive compensation capacitor, characterized in that, The method comprises the following steps: obtaining power grid data, calculating required compensation capacity, and determining a target capacitor group to be switched in; dividing the target capacitor group into a plurality of subgroups according to capacity or voltage level, each subgroup containing a plurality of capacitor units; establishing an initial model and initializing the front-end voltage state of each subgroup; updating the parameters of the initial model and the initial state of each subgroup according to the measured data of the power grid to obtain a transient prediction model; based on the transient prediction model, performing transient prediction on the process of each subgroup being gradually integrated into the bus, and predicting the switching response data of each subgroup; based on the switching response data, adjusting the switching sequence of each subgroup by using a constraint optimization method, and adjusting the interval of each subgroup being integrated according to capacity and electrical distance to balance the influence of the switching of each subgroup on the bus, and outputting a safe switching sequence; based on the voltage difference, the inrush current peak value and the capacitor capacity, calculating the risk characteristic value of each subgroup, and generating a preliminary switching sequence in ascending order of the risk characteristic value; delaying the switching of adjacent subgroups with an electrical distance less than a threshold, and optimizing the switching time of each subgroup; through the transient prediction model, if the bus voltage change, the switching current or the inrush current peak value exceeds the safety constraint, the sequence is adjusted, and a safe switching sequence meeting all constraints is outputted; according to the safe switching sequence, each subgroup and the internal capacitor units are switched in turn; for each subgroup in the safe switching sequence, the end voltage is gradually established to make the subgroup smoothly integrated into the bus; according to the safe switching sequence, each subgroup and the internal capacitor units are switched in turn, and the bus voltage and the switching current are monitored in real time until the entire target group is connected for reactive power compensation.

2. The method of claim 1, wherein the liquid flow reactive compensation capacitor is a capacitor of a liquid flow reactive compensation capacitor bank. The calculation of the required compensation capacity and the determination of the target capacitor group to be switched in are specifically as follows: performing discrete Fourier transform on the power grid data, extracting the complex spectrum of the fundamental frequency component, obtaining the fundamental effective value and the fundamental phase difference, and calculating the fundamental active power; according to the real-time power factor, the real-time phase angle is back calculated, and the target power factor is set to back calculate the target phase angle; according to the fundamental active power, the target phase angle and the real-time phase angle, the target compensation capacity is calculated; according to the target compensation capacity, from the pre-set capacitor group library, the candidate capacitor target group is selected according to the capacity matching principle of minimum overshoot priority; based on the mixed integer optimization method, the combination of the candidate capacitor target group with the total capacity closest to the target compensation capacity is determined as the target capacitor group to be switched in.

3. The method of claim 2, wherein the liquid flow reactive compensation capacitor is a capacitor of a liquid flow reactive compensation capacitor bank. The target capacitor group is divided into a plurality of subgroups according to capacity or voltage level, and each subgroup contains a plurality of capacitor units, specifically as follows: according to the rated capacity and voltage level of each capacitor unit in the target capacitor group, a capacitor unit attribute matrix is established; based on the hierarchical clustering method, the capacitor unit attribute matrix is clustered, the capacitor units with a capacity and voltage level difference within a pre-set first range are divided into the same subgroup, and a first division result is obtained; the first division result is subjected to constraint optimization, and the capacitor units that do not meet the operating conditions are removed to obtain a second division result.

4. The method of claim 3, wherein the liquid flow reactive compensation capacitor is a capacitor of a liquid flow reactive compensation capacitor bank. Based on the transient prediction model, the transient prediction is performed on the process of each subgroup being gradually integrated into the bus to predict the switching response data of each subgroup, specifically as follows: the capacitor subgroup is equivalent to a series impedance and an equivalent capacitor, and together with the equivalent impedance of the bus and the power supply, forms a transient network model; The front-end voltage state after the last switching is taken as the initial front-end voltage value of the sub-pack; A transient network relationship between the bus and the sub-pack is established to obtain an initial state equation; Impedance and capacitance parameters of the initial state equation are adjusted according to real-time power grid data; A bus recovery time constant is updated according to small-amplitude response data of the last switching to obtain an updated initial state of each sub-pack before being connected to the bus; According to a preset sub-pack switching sequence, each sub-pack is gradually connected to the bus, and in each step of the connection, bus voltage variation, switching current and inrush current peak value are calculated based on the initial state equation to generate sub-pack switching response data; The prediction result of each step is used for initial condition updating of the next step of sub-pack connection to realize continuous prediction.

5. The method of claim 4, wherein the liquid flow reactive compensation capacitor is a capacitor of a liquid flow reactive compensation capacitor bank. Based on the switching response data, a constraint optimization method is used to adjust the sub-pack switching sequence and the interval of sub-pack connection according to capacity and electrical distance to balance the influence of sub-pack switching on the bus, and a safe switching sequence is output, specifically as follows: A sub-pack risk characteristic value is defined by weighted summation of voltage difference in front of the bus, sub-pack predicted inrush current peak value and sub-pack capacitance capacity; The sub-packs are arranged in ascending order according to the sub-pack risk characteristic value to obtain a preliminary switching sequence; For adjacent sub-packs with an electrical distance less than a preset first lower limit, the switching time of the latter sub-pack is delayed to meet the electrical distance constraint; On the sequence meeting the electrical distance constraint, the switching time of each sub-pack is optimized to make the total capacity connected to the bus in any time period not exceed a set second upper limit; The preliminary sequence is input into a transient prediction model for verification, and if bus voltage variation, switching current or inrush current peak value exceeds a safety constraint, the sequence order is adjusted; The final sub-pack sequence meeting all constraints and the switching time form a safe switching sequence.

6. The method of claim 5, wherein the liquid flow reactive compensation capacitor is a capacitor of the type shown in Fig.

1. The end voltage of each sub-pack in the safe switching sequence is gradually established to make the sub-pack smoothly connected to the bus, specifically as follows: The target time and initial voltage state of a current sub-pack to be connected in the safe switching sequence are obtained; An end voltage adjustment model is established based on the target time and initial voltage state; Before the connection, the end voltage of the sub-pack is gradually adjusted through a controllable series impedance to generate a voltage trajectory approaching the bus voltage; The sub-pack is hierarchically connected according to unit capacity and electrical distance to establish end voltage consistency in the sub-pack; Based on real-time voltage difference and current feedback, a closed-loop control law is used to correct the adjustment process; When the voltage difference is less than a connection threshold, the series impedance channel is closed to smoothly connect the sub-pack to the bus; The system state after the connection is updated as an initial condition for subsequent end voltage adjustment of the sub-pack.

7. The compensation method for the fluid flow reactive power compensation capacitor according to claim 6, characterized in that, The end voltage of the sub-pack is gradually adjusted through a controllable series impedance before the connection to generate a voltage trajectory gradually approaching the bus voltage, specifically as follows: The target time and initial end voltage state of a current sub-pack to be connected in the safe switching sequence are obtained; An end voltage adjustment model is established by using the equivalent capacitance and series impedance of the sub-pack, and the end voltage adjustment model represents the dynamic mapping of the change of the end voltage of the sub-pack with time; An end voltage trajectory gradually approaching the bus voltage is generated according to the end voltage adjustment model; After adjustment at each time step, the actual end voltage is compared with the bus voltage, the corresponding switching current increment is calculated and applied to the sub-pack; When the end voltage of the sub-pack is consistent with the bus voltage, the sub-pack connection operation is completed. The final terminal voltage state of the incorporated sub-group is taken as the initial condition for the next sub-group terminal voltage adjustment.

8. The method of claim 7, wherein the liquid flow reactive compensation capacitor is a capacitor of a type selected from the group consisting of: The internal sub-group is layered and incorporated according to the unit capacity and electrical distance, establishing the consistency of the internal sub-group terminal voltage, specifically: ​ Reading the initial terminal voltage, rated capacity and electrical position of each capacitor unit in the sub-group, taking the overall terminal voltage trajectory of the sub-group as the reference target for internal incorporation; Layering the internal capacitor units of the sub-group according to the capacity size and electrical distance; For each layer of units, calculate the target terminal voltage of each unit according to the overall terminal voltage trajectory of the sub-group; Gradually apply adjustments to make the unit terminal voltage consistent with the target terminal voltage within the layer; After each adjustment, update the unit voltage state for the calculation of the next layer of units.

9. The method of claim 8, wherein the liquid flow reactive compensation capacitor is a capacitor of a liquid flow reactive compensation capacitor bank. According to the switching safety sequence, each sub-group and its internal capacitor units are switched in turn, while the bus voltage and switching current are monitored in real time until the entire target group is connected for reactive power compensation, specifically: According to the switching safety sequence, select the current sub-group and its internal capacitor units to be switched; Switch each capacitor unit in turn and collect bus voltage and switching current data in real time; Compare the real-time monitoring data with the preset safety threshold to determine whether the switching process is normal; Complete the switching of all sub-groups and their internal capacitor units to achieve the entire target group reactive power compensation connection; Update the final running state of the bus and each sub-group to provide initial conditions for subsequent operation and adjustment.

Images