A quick response control method of a magnetic pressure hybrid control type dynamic reactive power compensation device

Abstract

The present application relates to the technical field of reactive power compensation, and particularly relates to a quick response control method of a magnetic pressure mixed control type dynamic reactive power compensation device. The method predicts the predicted compensation amount at each future moment; according to the predicted active power and power factor, the compensation amount value interval required to improve the power factor to the preset target power factor range is determined; according to the overall fluctuation characteristics of the predicted active and reactive power sequences and the quantitative deviation degree between each predicted compensation amount and the compensation amount value interval, the basic compensation amount is determined; according to the deviation characteristics of the active and reactive power sequences and the instantaneous change rate of the power change curve constructed by fusing the active and reactive power sequences, the magnetic control compensation weight at each moment is determined; the basic compensation amount is divided into the magnetic control compensation amount and the electric control compensation amount according to the magnetic control compensation weight, and the magnetic control compensation device and the electric control compensation device are cooperatively controlled to output the compensation amount at the corresponding moment.

Description

Technical Field

[0001] This invention relates to the field of reactive power compensation technology, specifically to a fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device. Background Technology

[0002] In industries such as smelting, power grids, and chemicals, reactive power demand is enormous. Traditional electrical control compensation devices use a stepped adjustment method, resulting in slow response times and difficulty in achieving ideal compensation effects. Furthermore, frequent tap changer operation can cause transformer oil to crack due to arcing, leading to high maintenance requirements. While magnetically controlled dynamic compensation devices offer high accuracy, they also require significant investment. In this context, a magnetically controlled hybrid dynamic reactive power compensation device can greatly meet reactive power demands, combining the advantages of both electrical control and magnetically controlled dynamic compensation devices, offering a higher cost-performance ratio.

[0003] Currently, when performing reactive power compensation using a magnetic pressure hybrid control type dynamic reactive power compensation device, a single control method is often used in related technologies. However, when the two compensation methods are mixed, this method often uses real-time data feedback control, which lacks the ability to predict load changes and does not take into account the different applicable situations between the electric control compensation device and the magnetic control compensation device. If the control interval is too long when adjusting the compensation amount, it will also affect the compensation effect. Summary of the Invention

[0004] To address the technical problems of single-control technology lacking the ability to predict load changes and failing to consider the applicability between electronic and magnetic compensation devices, thus affecting the compensation effect, this invention provides a fast-response control method for a magnetic-pressure hybrid control type dynamic reactive power compensation device. The specific technical solution adopted is as follows:

[0005] This invention proposes a fast response control method for a magnetic-voltage hybrid control type dynamic reactive power compensation device. The magnetic-voltage hybrid control type dynamic reactive power compensation device includes a magnetic control compensation device and an electrical control compensation device. The method includes:

[0006] Real-time acquisition of power data from grid loads generates a power data sequence, which includes at least power factor, active power, and reactive power under uncompensated conditions.

[0007] Based on historical power data sequences, predict the power data sequence of the load within a preset time period and the predicted compensation amount at each future moment;

[0008] For each future time point, based on the predicted active power and power factor, determine the range of compensation values ​​required to raise the power factor to the preset target power factor range; based on the overall fluctuation characteristics of the predicted active power sequence and reactive power sequence and the degree of quantitative deviation between each predicted compensation value and its compensation value range at the same time, determine the basic compensation value.

[0009] A sliding preset time window is set within a future preset period. Based on the deviation characteristics of the active power sequence and the reactive power sequence within the time window and the instantaneous rate of change of the power change curve constructed by fusing the active and reactive power sequences, the magnetic control compensation weight at each moment is determined.

[0010] Based on the magnetic compensation weight, the basic compensation amount is divided into magnetic compensation amount and electrical compensation amount, and the magnetic compensation device and electrical compensation device are coordinated to output the compensation amount at the corresponding time.

[0011] Furthermore, the power factor includes the original power factor in the uncompensated state and the expected target power factor after compensation. The process for determining the predicted compensation amount includes:

[0012] For each moment in a future preset time period, based on the predicted active power, the original power factor, and the target power factor, the predicted compensation amount is obtained through a preset compensation amount calculation function.

[0013] Furthermore, the process of determining the range of compensation values ​​includes:

[0014] For each moment within a preset future time period, based on the predicted active power at that moment and the upper limit of the preset target power factor range, the corresponding first target reactive power is derived in reverse; based on the predicted active power at that moment and the lower limit of the preset target power factor range, the corresponding second target reactive power is derived in reverse.

[0015] Calculate the difference between the first target reactive power and the predicted uncompensated reactive power at this moment, and use it as the upper boundary value of the compensation range.

[0016] The difference between the second target reactive power and the predicted uncompensated reactive power at that moment is calculated and used as the lower boundary value of the compensation range.

[0017] Furthermore, the power factor is calculated based on reactive power and active power.

[0018] Furthermore, the process for determining the overall fluctuation characteristics of the predicted active power sequence and reactive power sequence includes:

[0019] Within a predetermined time period in the future, STL decomposition processing is performed on the predicted reactive power series and active power series respectively to obtain the decomposition data corresponding to each series. The decomposition data includes trend terms, seasonality terms and residual terms.

[0020] The residual terms corresponding to each future moment in the decomposed data are quantified and used as instantaneous fluctuation quantities to characterize the degree of fluctuation in power data at each moment.

[0021] Furthermore, the process for determining the degree of quantization deviation between each predicted compensation amount and its corresponding value range includes:

[0022] For each moment in a future preset time period, the deviation between the predicted compensation amount and the compensation amount range at the same moment is calculated. The deviation is used to quantify the degree of deviation of each predicted compensation amount from its corresponding compensation amount range.

[0023] Furthermore, the method for calculating the deviation includes:

[0024] If the predicted compensation amount is within the range of compensation amount values ​​at the same time, the deviation is 0;

[0025] If the predicted compensation amount is less than the lower boundary value of the compensation amount range at the same time, the deviation is the difference between the lower boundary value and the predicted compensation amount.

[0026] If the predicted compensation amount is greater than the upper boundary value of the compensation amount range at the same time, the deviation is the difference between the predicted compensation amount and the upper boundary value.

[0027] Furthermore, the process of determining the basic compensation amount includes:

[0028] For each future moment, calculate the product between the reciprocal of the instantaneous fluctuation at the same moment and the deviation.

[0029] Summing all the products within a preset future time period yields a first summation value.

[0030] The exponential function value is obtained by performing exponential operations with the natural constant as the base and the opposite of the first summation value as the exponent.

[0031] Calculate the product between the exponential function value and the predicted compensation amount at each future time point, and use it as the first compensation amount at each time point; calculate the weighted average of the first compensation amounts at each time point in the future preset time period, and use it as the basic compensation amount.

[0032] Furthermore, the process of determining the magnetically controlled compensation weights includes:

[0033] For each time window, calculate the first difference between the active power at each moment within the time window and the mean of all active power; calculate the second difference between the reactive power at each moment within the time window and the mean of all reactive power.

[0034] The sum of the first difference and the second difference is calculated as the total difference; the product of the total difference and the reciprocal of the instantaneous fluctuation is calculated; all the products within the time window are summed to obtain the power data fluctuation amount used to characterize the stability of the power data within the time window; and the power data fluctuation amount is used as the fluctuation index at each moment within the corresponding time window.

[0035] Calculate the slope of the power change curve constructed by fusing active power series and reactive power series at each time step;

[0036] Based on the product of the reciprocal of the fluctuation index and the reciprocal of the absolute value of the slope at the same moment, the magnetic control compensation weight at each moment within the time window is obtained through a preset magnetic control compensation weight calculation function.

[0037] Furthermore, a fast-response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device also includes:

[0038] Real-time acquisition of the actual required compensation amount after compensation; calculation of the normalized deviation between the actual required compensation amount and the basic compensation amount;

[0039] The magnetically controlled compensation weight decreases as the deviation increases, and the compensation amount reduced by the magnetically controlled compensation weight is automatically transferred to the electrically controlled compensation branch.

[0040] The present invention has the following beneficial effects:

[0041] This invention first predicts the load's power data sequence and the predicted compensation amount at each future moment based on historical power data sequences. This helps to address potential sudden changes in the load and avoid voltage flicker caused by delayed compensation decisions in the magnetic voltage hybrid dynamic reactive power compensation device. Next, by analyzing the compensation amount range, it predicts the possible changes in reactive power compensation at each moment and then analyzes the most reasonable base compensation amount. This avoids excessive compensation output from the subsequent electronically controlled compensation device, which could affect grid stability, and also suppresses harmonic resonance in the magnetically controlled compensation device. Finally, by using appropriate magnetically controlled compensation weights, it scientifically allocates the compensation outputs of the electronically controlled compensation device and the magnetically controlled compensation device based on the base compensation amount, significantly improving the reactive power compensation effect of the magnetic voltage hybrid dynamic reactive power compensation device. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of a fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to an embodiment of the present invention;

[0044] Figure 2 This is a structural example diagram of the data acquisition component of a magnetic pressure hybrid control type dynamic reactive power compensation device provided in an embodiment of the present invention;

[0045] Figure 3 This is an example diagram illustrating the process of determining the basic compensation amount according to an embodiment of the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0048] The following description, in conjunction with the accompanying drawings, details the specific scheme of the fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device provided by the present invention.

[0049] Please see Figure 1 The diagram illustrates a fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to an embodiment of the present invention. The method includes:

[0050] S101: Real-time acquisition of power data of grid load and generation of power data sequence, wherein the power data includes at least power factor, active power and reactive power under uncompensated state.

[0051] It is understood that the magnetic pressure hybrid control type dynamic reactive power compensation device includes a magnetic control compensation device and an electrical control compensation device.

[0052] For example, a structural example diagram of the data acquisition component of a magnetic pressure hybrid control type dynamic reactive power compensation device is shown below. Figure 2 As shown, the acquisition component is used to collect power data of the grid load. The device includes a preset current sensor and a voltage sensor to collect the real-time voltage and current of the grid load. The magnetic voltage hybrid control type dynamic reactive power compensation device also includes other common acquisition components, such as amplifiers. The specific functions and connection methods are common technical means and will not be described in detail in this embodiment. Under normal circumstances, the voltage and current are collected synchronously through the current sensor and voltage sensor at a preset sampling frequency. Then, the collected current and voltage data are denoised and smoothed. Existing algorithms are used to calculate reactive power and power factor. Then, based on the power factor, current and voltage, active power is determined through existing calculation formulas.

[0053] It should be noted that the methods for determining power factor, active power, and reactive power under uncompensated conditions are all well-known technical means in the art, and will not be described in detail in this embodiment.

[0054] It should be noted that the specific value of the preset sampling frequency is determined according to actual needs, and this embodiment does not impose a specific limitation. For example, the preset sampling frequency can be 1kHz (i.e., 1000 samples per second).

[0055] It should be noted that since the compensated reactive power has been regulated, it cannot reflect the reactive power characteristics of the load itself. Only the uncompensated reactive power can truly reflect the original reactive power demand of the load. Therefore, when predicting the reactive power compensation amount at various future times, the reactive power in the uncompensated state should be selected.

[0056] S102: Based on historical power data sequences, predict the power data sequence of the load within a preset time period and the predicted compensation amount at each future time.

[0057] It should be noted that existing time series forecasting algorithms can be used to predict the power data sequence of the load within a preset period based on historical power data sequences. For example, ARIMA (Autoregressive Integrated Moving Average) can be used to capture the dynamic patterns of power data and thus predict the power data sequence. The specific methods for predicting the power data sequence are well known to those skilled in the art and will not be described in detail in this embodiment.

[0058] The specific value of the future preset time period is determined according to actual needs. This embodiment does not make specific limitations. For example, it can predict the power data sequence of the load within the next 5 minutes.

[0059] It should be noted that the power factor includes the original power factor in the uncompensated state and the expected target power factor after compensation.

[0060] The original power factor, the power factor before compensation, can usually be read from an electricity meter or a power factor meter.

[0061] The target power factor is the power factor that is desired to be achieved after compensation. Based on prior knowledge, the power factor stipulated by the state is known, and its value is usually in the range of 0.9 to 0.95.

[0062] In this embodiment, the predicted compensation amount can be obtained based on the predicted power data sequence. As one possible implementation method, for each moment in a future preset time period, the predicted compensation amount is obtained through a preset compensation amount calculation function based on the predicted active power, the original power factor, and the target power factor.

[0063] It should be noted that the preset compensation amount calculation function is a common existing technology, and will not be described in detail in this embodiment. For example, the preset compensation amount calculation function can be expressed as: ,in, Indicates the predicted compensation amount; Indicates active power; This represents the tangent value corresponding to the original power factor; This represents the tangent value corresponding to the target power factor.

[0064] The power data sequence includes a power factor sequence, an active power sequence, and a reactive power sequence under uncompensated conditions. The power factor sequence consists of power factors arranged in chronological order; the active power sequence consists of active power arranged in chronological order; and the reactive power sequence consists of reactive power under uncompensated conditions arranged in chronological order.

[0065] S103: For each future time point, based on the predicted active power and power factor, determine the range of compensation values ​​required to raise the power factor to the preset target power factor range; based on the overall fluctuation characteristics of the predicted active power sequence and reactive power sequence and the degree of quantitative deviation between each predicted compensation value and its compensation value range at the same time, determine the basic compensation value.

[0066] It's important to understand that the fundamental purpose of a magnetic pressure hybrid control type dynamic reactive power compensation device during dynamic reactive power compensation is to improve the power factor. By analyzing the deviation in the power factor, it determines whether to increase or decrease reactive power compensation. The goal is to maintain the power factor within a reasonable range. Reactive power is the most direct compensation target. By analyzing changes in reactive power in real time, the reactive power compensation amount of the magnetic pressure hybrid control type dynamic reactive power compensation device is adjusted.

[0067] It should be noted that the power factor can be obtained based on the ratio of apparent power to active power. According to existing knowledge, apparent power, active power, and reactive power satisfy the Pythagorean theorem. Therefore, based on active and reactive power, and using the Pythagorean theorem, we can obtain apparent power, and thus deduce that the power factor can be calculated based on reactive and active power. The power factor can be expressed by the following formula: , ,in, Indicates the power factor; S represents apparent power; Q represents active power; Q represents reactive power.

[0068] It should be noted that, under actual operating conditions, the active power cannot be zero during normal operation of the power grid load.

[0069] It should be noted that, as can be seen from the formula for calculating the power factor, given the active power and the power factor, the reactive power can be derived in reverse.

[0070] It should be noted that, in order to keep the power factor within a reasonable range in the future, the reactive power compensation amount can be derived by analyzing the range of changes required for the power factor, i.e., the range of changes required for the compensation amount.

[0071] In this embodiment, for each moment within a preset future time period, based on the predicted active power at that moment and the upper limit of the preset target power factor range, the corresponding first target reactive power is derived in reverse; based on the predicted active power at that moment and the lower limit of the preset target power factor range, the corresponding second target reactive power is derived in reverse; the difference between the first target reactive power and the predicted uncompensated reactive power at that moment is calculated as the upper boundary value of the compensation range; the difference between the second target reactive power and the predicted uncompensated reactive power at that moment is calculated as the lower boundary value of the compensation range.

[0072] It should be noted that the specific range of the preset target power factor range is determined according to actual needs, and this embodiment does not impose a specific limitation. For example, the range of the preset target power factor range can be 0.9 to 0.95, which conforms to the national standard. Here, 0.9 is the lower limit of the preset target power factor range, and 0.95 is the upper limit of the preset target power factor range.

[0073] It's important to understand that when the grid voltage fluctuates, the magnetic voltage hybrid control type dynamic reactive power compensation device dynamically adjusts the reactive power to ensure voltage stability. Both excessively low and excessively high voltages can affect equipment operation and power quality. It's known that voltage changes in the grid are influenced by the load. Therefore, changes in load power data can cause grid voltage fluctuations. Grid-connected loads are generally quite large, such as factory equipment or grid paralleling. Therefore, loads with large power fluctuations have a significant impact on the grid. Generally, when the voltage is too high, reactive power compensation is reduced to maintain the power factor within a reasonable range, and the voltage is lowered to a stable range. When the current is too high, reactive power compensation is usually increased to maintain the power factor within a reasonable range.

[0074] It should be noted that when the connected load current increases, the decision to increase or decrease reactive power compensation should be based on the specific cause of the current increase (whether it is dominated by inductive reactive power or active power). The core logic is to optimize the current by adjusting the reactive power, reduce line losses, and maintain system stability. If the current increase is caused by excessive inductive reactive power, the reactive power compensation should be increased. If the current increase is caused by an increase in active power, there is no need to adjust the reactive power compensation (or it can be finely adjusted to maintain voltage stability). Therefore, when the current increases, it is important to first determine whether it is caused by excessive inductive reactive power. The specific determination method is a well-known technical means in the art and will not be described in this embodiment.

[0075] The process of determining the basic compensation amount is as follows: Figure 3 As shown, it includes:

[0076] S103-1: Within a predetermined time period in the future, perform STL decomposition processing on the predicted reactive power series and active power series respectively to obtain the decomposition data corresponding to each series. The decomposition data includes trend terms, seasonality terms and residual terms.

[0077] It should be noted that the STL decomposition process is a seasonal-trend decomposition method based on local weighted regression. The trend term reflects the long-term change pattern of the corresponding sequence, the seasonal term reflects the periodic fluctuation characteristics of the corresponding sequence, and the residual term reflects the random fluctuation component in the corresponding sequence after removing the trend and seasonal terms.

[0078] Seasonal-Trend decomposition using LOESS (STL) is an existing decomposition technique for time series data. Therefore, the specific methods for obtaining the decomposed data are well-known to those skilled in the art, and will not be described in detail in this embodiment.

[0079] S103-2: The residual terms corresponding to each future moment in the quantified decomposition data are used as instantaneous fluctuation quantities to characterize the degree of fluctuation in power data at each moment.

[0080] It should be noted that the specific method for quantifying and decomposing the residual terms corresponding to each future time in the data is based on existing technical means. For example, firstly, in order to eliminate the difference in magnitude of the residual terms at different times, the residual terms are normalized. Then, considering that the residual terms at different times have different degrees of influence on the stability assessment of power data, weight coefficients are assigned to the normalized residual terms. The weight coefficients can be dynamically adjusted according to factors such as data reliability and historical fluctuations, and finally, the quantified residual terms are determined.

[0081] It's important to understand that in the STL process decomposition, decomposing the active power series and reactive power series separately yields two residual terms (reflecting the random fluctuations in active and reactive power that cannot be explained by the trend and periodic terms, respectively). The core logic for quantifying these two residual terms into a unified "instantaneous fluctuation quantity" is to integrate their fluctuation characteristics to comprehensively reflect the overall stability of the power system.

[0082] As one possible implementation method, the two residual terms are first quantified separately, and then the quantified values ​​of the two residuals are integrated into a comprehensive index through weighted or combined operations. For example, the instantaneous fluctuation can be obtained by taking the mean after normalization.

[0083] It's important to understand that the larger the instantaneous fluctuation, i.e., the larger the quantified residual term, the more drastic the fluctuation in the power data at that moment, and the lower the data stability. The quantified residual term provides an accurate basis for subsequently determining parameters such as the basic compensation amount and the magnetic control compensation weight. The value of the instantaneous fluctuation is negatively correlated with system stability: the smaller the instantaneous fluctuation, the weaker the random fluctuations in active and reactive power, and the more stable the system operation; the larger the instantaneous fluctuation, the stronger the random disturbance reflected by the residual term, and the worse the system stability, requiring compensation strategies (such as adjusting the magnetic control / electric control weights) to suppress the fluctuations.

[0084] S103-3: For each moment in a future preset time period, calculate the deviation between the predicted compensation amount and the compensation amount range at the same moment. The deviation is used to quantify the degree of deviation of each predicted compensation amount from its corresponding compensation amount range.

[0085] In this embodiment, if the predicted compensation amount is within the range of compensation amount values ​​at the same time, the deviation is 0; if the predicted compensation amount is less than the lower boundary value of the range of compensation amount values ​​at the same time, the deviation is the difference between the lower boundary value and the predicted compensation amount; if the predicted compensation amount is greater than the upper boundary value of the range of compensation amount values ​​at the same time, the deviation is the difference between the predicted compensation amount and the upper boundary value.

[0086] S103-4: For each future moment, calculate the product between the reciprocal of the instantaneous fluctuation at the same moment and the deviation; sum all products within the preset future time period to obtain the first sum value; perform exponential operation with the natural constant as the base and the opposite of the first sum value as the exponent to obtain the exponential function value.

[0087] It should be noted that, from the perspective of equipment safety, the compensation amount should be prevented from exceeding the rated capacity of the device (such as the maximum adjustment range of the magnetic / electric control device); from the perspective of power grid stability, the compensation should be kept within a safe range to avoid excessive power factor or voltage fluctuations caused by over-compensation. When determining a reasonable basic compensation amount, it is necessary to consider not only that the adjustment of the magnetic control should be kept within a certain stable range, but also that the electric control compensation device can respond quickly when there are large power changes. Therefore, the long-term variation patterns of reactive power sequence and active power sequence in the future preset period can be analyzed to explore the appropriate reactive power compensation amount, i.e., the basic compensation amount.

[0088] The exponential function value is a dimensionless safety adjustment factor that is used to perform global attenuation control on the predicted compensation amount based on the real-time risk level of the power grid, so as to obtain a more appropriate compensation amount at each time. It integrates two types of key parameters: one is the instantaneous fluctuation, which is used to take into account the transient stability of the power grid; the other is the deviation, which is used to solve the limitations of single index control.

[0089] Since the exponential function establishes a nonlinear attenuation relationship, if there are large instantaneous fluctuations in a future time period, the quantification result of the residual term (instantaneous fluctuation) reflects the randomness and severity of power fluctuations. When the instantaneous fluctuation is large (severe fluctuation), it indicates that the system faces strong random disturbances. In this case, it is necessary to increase the dynamic response weight of the basic compensation amount, and offset the impact of fluctuations on the power grid by increasing the adjustment amplitude and speed of the compensation amount. Conversely, when the instantaneous fluctuation is small (smooth fluctuation), the adjustment intensity of the compensation amount can be appropriately reduced to reduce the number of device actions and reduce energy consumption. Furthermore, if the deviation is large, it indicates that the predicted compensation amount deviates further from the target range. In this case, it is necessary to adjust the basic compensation amount to make the compensation amount closer to a reasonable range. That is, when the fluctuation is smooth, the influence of the deviation is amplified (the compensation amount needs to be corrected to a reasonable range first), and when the fluctuation is severe, the influence of the deviation is weakened (avoiding over-correction that aggravates fluctuations). In summary, the larger the first summation value, the smaller the exponential function value, which plays a role in "suppressing extreme value interference". Therefore, the exponential function value can be expressed by the following formula:

[0090]

[0091] Where 'a' represents the value of the exponential function; M represents the instantaneous fluctuation at time i within a future preset time period; M represents the number of times within the future preset time period. This represents the deviation at time i within a preset future time period; This represents an exponential function with the natural constant as its base.

[0092] It should be noted that, under actual operating conditions, the quantized residual term, i.e., the instantaneous fluctuation, is almost impossible to be zero in practical applications due to factors such as the inherent fluctuations of the power system, the limitations of the prediction model, the approximation of the quantization method, and measurement errors.

[0093] For example, if the future time period is preset to be within 30 seconds, and 1 second is one moment, then there are 30 moments within the future 30 seconds, that is, M is 30.

[0094] S103-5: Calculate the product between the exponential function value and the predicted compensation amount at each future time, and use it as the first compensation amount at each time; calculate the weighted average of the first compensation amounts at each time in the future preset time period, and use it as the basic compensation amount.

[0095] To ensure that the compensation strategy can cover the worst operating conditions (such as short-term strong fluctuations and large deviations from the range), avoid grid instability caused by insufficient compensation, and comply with the principle of "safety first" in the power system, the weighted average of the first compensation amount at each moment in the future preset time period is calculated as the basic compensation amount.

[0096] It should be noted that the weights can be dynamically allocated based on instantaneous fluctuations. The core logic of weight allocation is: the more severe the grid fluctuations, the higher the weight of the corresponding first compensation amount. Specifically, the weights are dynamically adjusted based on the two key indicators "instantaneous fluctuations" and "deviation" in the document. For example, instantaneous fluctuations are used as the core weight factor. Instantaneous fluctuations are quantified by the residual terms obtained through STL decomposition, reflecting the degree of random fluctuation in active / reactive power at that moment (the larger the value, the more severe the fluctuation). The weight is positively correlated with the instantaneous fluctuations: the larger the instantaneous fluctuations at a certain moment, the worse the grid stability, and the higher the compensation accuracy at that moment should be prioritized, thus giving it a higher weight. Deviation is used as a correction factor. Deviation reflects the degree of deviation between the predicted compensation amount and the reasonable range (upper / lower boundary) at that moment (the larger the value, the more severe the deviation). If the deviation is large at a certain moment, even if the instantaneous fluctuations are small, its weight should be appropriately increased to ensure that the compensation amount is corrected to the reasonable range. The weights for all moments are normalized to ensure that the sum of the weights for all moments is 1, avoiding distortion of the total compensation amount due to weight allocation.

[0097] It should be noted that the specific method for calculating the weights based on the two weighting factors is a well-known technique in the art, and will not be described in detail in this embodiment. For example, the weights can be obtained through factor analysis.

[0098] S104: Set a sliding preset time window within a future preset period, and determine the magnetic control compensation weight at each moment based on the deviation characteristics of the active power sequence and reactive power sequence within the time window and the instantaneous change rate of the power change curve constructed by fusing the active and reactive power sequences.

[0099] It's important to understand that the reactive power compensation output of a magnetic-voltage hybrid dynamic reactive power compensation device is equal to the sum of the magnetic control compensation output of the magnetic control device and the electrical control compensation output of the electrical control device. Since magnetic control and electrical control each perform different compensation functions—magnetic control typically adjusts the inductive reactance or magnetic field strength of the inductor, suitable for relatively stable load changes or long-term compensation; and electrical control, by rapidly switching the operating states of capacitors or reactors, can more accurately respond to load fluctuations and rapid voltage changes, suitable for short-term dynamic compensation—magnetic control provides stable, long-term reactive power regulation, while electrical control is responsible for handling instantaneous fluctuations and providing rapid response. The combination of these two ensures that when the load changes slowly, fine adjustment is achieved through magnetic control, while rapid compensation is achieved through electrical control when the load changes drastically. The electrical control compensation device provides extremely fast response and quickly adjusts reactive power, while magnetic control provides relatively stable compensation without generating excessive fluctuations. The combination of the two can effectively address various dynamic and steady-state demands of the power grid. Therefore, after determining a reasonable basic compensation amount, further subdivision of the basic compensation amount is necessary.

[0100] In this embodiment, for each time window, a first difference is calculated between the active power at each moment within the time window and the mean of all active power; a second difference is calculated between the reactive power at each moment within the time window and the mean of all reactive power; the sum of the first and second differences is calculated as the total difference; the product of the total difference and the reciprocal of the instantaneous fluctuation is calculated; all products within the time window are summed to obtain the power data fluctuation, which characterizes the stability of the power data within the time window; and the power data fluctuation is used as the fluctuation index at each moment within the corresponding time window; the slope of the power change curve constructed by fusing the active power sequence and the reactive power sequence is calculated at each moment; based on the product of the reciprocal of the fluctuation index at the same moment and the reciprocal of the absolute value of the slope, the magnetic control compensation weight at each moment within the time window is obtained through a preset magnetic control compensation weight calculation function.

[0101] It should be noted that the specific information of the preset time window can be determined according to actual needs. This embodiment does not impose specific limitations. For example, if the future preset time period is 30 seconds, then in order to fully understand the fluctuation of the active power sequence and reactive power sequence within the future preset time period, the preset time window can be a sliding time window with a fixed length of 3 seconds.

[0102] It should be noted that the specific construction method of the power change curve constructed by integrating the active power series and reactive power series is a well-known technique in the art, and will not be elaborated in this embodiment. For example, firstly, the active power series and reactive power series under uncompensated conditions are calibrated on the time axis to ensure that each data point of the two series corresponds one-to-one in time, eliminating time deviation; based on existing power system formulas, such as... , By integrating the two characteristic quantities of active power and reactive power through the power system formula, the comprehensive relationship between active and reactive power is reflected. Finally, a single curve is plotted with time as the horizontal axis and apparent power or power factor as the vertical axis; or active power curves and reactive power curves are plotted separately in the same coordinate system, and the coordinated change law of the two is shown by superposition.

[0103] Since a large difference between reactive power and the mean of all reactive power at a given moment, and a large difference between active power and the mean of all active power, indicates a significant deviation of the load power from the average level within the window, it signifies more severe power fluctuations at that moment. By multiplying the power deviation with the instantaneous fluctuations, a larger sum of the first and second differences at a given moment (i.e., the total difference) coupled with smaller instantaneous fluctuations (a larger reciprocal) results in a larger product and a higher contribution to the total fluctuation. The summation operation then aggregates the fluctuation characteristics of a single moment into the overall fluctuation level of the window. The larger the fluctuation, the more unstable the power data within the window, requiring rapid electronic control compensation. In this case, the compensation output of the electronic control compensation device needs to be increased beyond the base compensation, and the magnetic control compensation weight should be relatively small. Conversely, a larger absolute value of the slope reflects the changing trend of the power data, indicating a faster rate of change and requiring a faster subsequent compensation response. In this case, the compensation output of the electronic control compensation device can be increased, and the magnetic control compensation weight should be relatively small. Therefore, the magnetic control compensation weight can be represented by the following preset magnetic control compensation weight calculation function:

[0104]

[0105] in, This represents the magnetic control compensation weight at time i within a preset future time period; ;in, This represents the fluctuation index at time i within a preset future time period; It represents the instantaneous fluctuation at the Z-th moment within any time window; This represents the active power at the Z-th moment within any time window; express The average of all active power within the corresponding time window; This represents the reactive power at the Z-th moment within any time window; express The average of all reactive power within the corresponding time window; The slope represents the slope at time i within a preset time period; | represents the absolute value; N represents the number of times within any time window.

[0106] It should be noted that in the denominator as well as Add a very small positive offset to Mathematically, this ensures the denominator as well as Strictly greater than zero, this completely eliminates the possibility of a zero denominator in division operations, ensuring the robustness and executability of the calculation formula. It is preset to a sufficiently small positive value, and the specific value is chosen based on the principle that: as well as Under normal operating conditions where the value is significantly greater than zero, For the calculation results The impact is negligible, for example, The specific value is determined based on actual needs. It can be 0.01, but this embodiment does not impose a specific limitation.

[0107] It should be noted that, under actual operating conditions, the quantized residual term, i.e., the instantaneous fluctuation, is almost never zero in practical applications due to factors such as the inherent fluctuations of the power system, the limitations of the prediction model, the approximation of the quantization method, and measurement errors. It cannot be zero.

[0108] It should be noted that the sum of the magnetic compensation weight and the electronic compensation weight is 1 (i.e., electronic compensation weight = 1 - magnetic compensation weight).

[0109] S105: Based on the magnetic compensation weight, the basic compensation amount is divided into magnetic compensation amount and electrical compensation amount, and the magnetic compensation device and electrical compensation device are coordinated to output the compensation amount at the corresponding time.

[0110] In this embodiment, the product of the basic compensation amount and the magnetic control compensation weight is calculated as the magnetic control compensation amount; the difference between the basic compensation amount and the magnetic control compensation amount is calculated as the electronic control compensation amount.

[0111] It is important to understand that the above analysis is based on the quantification of the collected predicted power data. However, when reactive power compensation is performed, there may be some deviation between the actual collected power data and the predicted value. When performing power control compensation, the response is faster, so the compensation amount can be adjusted directly based on the actual collected power data.

[0112] In this embodiment, the actual required compensation amount after compensation is collected in real time; the normalized deviation between the actual required compensation amount and the basic compensation amount is calculated; the magnetically controlled compensation weight is controlled to decrease as the deviation increases; and the compensation amount reduced by the magnetically controlled compensation weight is automatically transferred to the electrically controlled compensation branch.

[0113] The normalized deviation is the deviation between the actual required compensation amount and the basic compensation amount at the same time. The specific calculation method of the deviation can be referred to the explanation of step S103 in this embodiment, which will not be repeated in this embodiment.

[0114] It should be noted that the sum of the corrected magnetic control compensation and electrical control compensation remains consistent with the basic compensation, and the corrected electrical control compensation does not exceed its maximum adjustment range, while the corrected magnetic control compensation does not exceed its rated compensation capacity.

[0115] It should be noted that the specific information on the maximum adjustment range and the rated compensation capacity is determined based on prior knowledge, and this embodiment does not impose specific limitations. For example, the maximum adjustment range can be 999 kvar, and the rated compensation capacity can be 2400 kvar.

[0116] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0117] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device, characterized in that, The device includes a magnetically controlled compensation device and an electrically controlled compensation device, and the method includes: Real-time acquisition of power data from grid loads generates a power data sequence, which includes at least power factor, active power, and reactive power under uncompensated conditions. Based on historical power data sequences, predict the power data sequence of the load within a preset time period and the predicted compensation amount at each future moment; For each future time point, based on the predicted active power and power factor, determine the range of compensation values ​​required to raise the power factor to the preset target power factor range; based on the overall fluctuation characteristics of the predicted active power sequence and reactive power sequence and the degree of quantitative deviation between each predicted compensation value and its compensation value range at the same time, determine the basic compensation value. A sliding preset time window is set within a future preset period. Based on the deviation characteristics of the active power sequence and the reactive power sequence within the time window and the instantaneous rate of change of the power change curve constructed by fusing the active and reactive power sequences, the magnetic control compensation weight at each moment is determined. Based on the magnetically controlled compensation weight, the basic compensation amount is divided into magnetically controlled compensation amount and electrically controlled compensation amount, and the magnetically controlled compensation device and the electrically controlled compensation device are coordinated to output the compensation amount at the corresponding time. The process for determining the overall fluctuation characteristics of the predicted active power sequence and reactive power sequence includes: Within a predetermined time period in the future, STL decomposition processing is performed on the predicted reactive power series and active power series respectively to obtain the decomposition data corresponding to each series. The decomposition data includes trend terms, seasonality terms and residual terms. The residual terms corresponding to each future time in the decomposed data are quantified and used as instantaneous fluctuation quantities to characterize the degree of fluctuation of power data at each time. The process of determining the magnetically controlled compensation weights includes: For each time window, calculate the first difference between the active power at each moment within the time window and the mean of all active power; calculate the second difference between the reactive power at each moment within the time window and the mean of all reactive power. The sum of the first difference and the second difference is calculated as the total difference; the product of the total difference and the reciprocal of the instantaneous fluctuation is calculated; all the products within the time window are summed to obtain the power data fluctuation amount used to characterize the stability of the power data within the time window; and the power data fluctuation amount is used as the fluctuation index at each moment within the corresponding time window. Calculate the slope of the power change curve constructed by fusing active power series and reactive power series at each time step; Based on the product of the reciprocal of the fluctuation index and the reciprocal of the absolute value of the slope at the same moment, the magnetic control compensation weight at each moment within the time window is obtained through a preset magnetic control compensation weight calculation function.

2. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 1, characterized in that, The power factor includes the original power factor in the uncompensated state and the expected target power factor after compensation. The process of determining the predicted compensation amount includes: For each moment in a future preset time period, based on the predicted active power, the original power factor, and the target power factor, the predicted compensation amount is obtained through a preset compensation amount calculation function.

3. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 2, characterized in that, The process of determining the range of compensation values ​​includes: For each moment within a preset future time period, based on the predicted active power at that moment and the upper limit of the preset target power factor range, the corresponding first target reactive power is derived in reverse; based on the predicted active power at that moment and the lower limit of the preset target power factor range, the corresponding second target reactive power is derived in reverse. Calculate the difference between the first target reactive power and the predicted uncompensated reactive power at this moment, and use it as the upper boundary value of the compensation range. The difference between the second target reactive power and the predicted uncompensated reactive power at that moment is calculated and used as the lower boundary value of the compensation range.

4. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 3, characterized in that, The power factor is calculated based on reactive power and active power.

5. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 1, characterized in that, The process for determining the degree of quantification deviation between each predicted compensation amount and its corresponding compensation amount range includes: For each moment in a future preset time period, the deviation between the predicted compensation amount and the compensation amount range at the same moment is calculated. The deviation is used to quantify the degree of deviation of each predicted compensation amount from its corresponding compensation amount range.

6. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 5, characterized in that, The method for calculating the deviation includes: If the predicted compensation amount is within the range of compensation amount values ​​at the same time, the deviation is 0; If the predicted compensation amount is less than the lower boundary value of the compensation amount range at the same time, the deviation is the difference between the lower boundary value and the predicted compensation amount. If the predicted compensation amount is greater than the upper boundary value of the compensation amount range at the same time, the deviation is the difference between the predicted compensation amount and the upper boundary value.

7. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 6, characterized in that, The process of determining the basic compensation amount includes: For each future moment, calculate the product between the reciprocal of the instantaneous fluctuation at the same moment and the deviation. Summing all the products within a future preset time period yields a first summation value; The exponential function value is obtained by performing exponential operations with the natural constant as the base and the opposite of the first summation value as the exponent. Calculate the product between the exponential function value and the predicted compensation amount at each future time point, and use it as the first compensation amount at each time point; calculate the weighted average of the first compensation amounts at each time point in the future preset time period, and use it as the basic compensation amount.

8. The fast response control method for a magnetic pressure hybrid control type dynamic reactive power compensation device according to claim 1, characterized in that, The method further includes: Real-time acquisition of the actual required compensation amount after compensation; calculation of the normalized deviation between the actual required compensation amount and the basic compensation amount. The magnetically controlled compensation weight decreases as the deviation increases, and the compensation amount reduced by the magnetically controlled compensation weight is automatically transferred to the electrically controlled compensation branch.

Images