Active-reactive collaborative optimization scheduling method for power distribution network

Abstract

The invention relates to the technical field of power grid dispatching, in particular to an active-reactive collaborative optimization dispatching method for a power distribution network. Acquiring voltage, active power and power grid line data at the uploading moment of the user side and various monitoring data of the transformer substation in a preset time period; because the user side data and the transformer substation side data are in space-time asymmetry and have measurement blind areas, various data numerical value characteristics are fused and analyzed and compared with the user side voltage, a voltage-power transmission coefficient is determined, an accurate voltage prediction model is constructed, and the voltage prediction value of the user side at each preset moment is obtained; as the time goes on or the photovoltaic fluctuates greatly, the error increases, so that the active power change in the preset time period and the active power difference of the user side at the to-be-measured moment are corrected based on the difference characteristics of the transformer station monitoring data, and the corrected available capacity of the inverter is obtained; and determining an adjustment burden index by combining the voltage predicted value, and finally adjusting the transformer according to an index accumulation condition and the voltage predicted value.

Description

A method for coordinated optimization scheduling of active and reactive power in power distribution networks Technical Field

[0001] This invention relates to the field of power grid dispatching technology, specifically to a method for coordinated optimization dispatching of active and reactive power in distribution networks. Background Technology

[0002] In today's power systems, the distribution network, as a crucial link connecting the generation end and the user end, is vital for ensuring the quality of power supply through its operational efficiency and stability. With the large-scale integration of distributed energy sources (such as solar photovoltaic and wind power) into the distribution network, and the diversification and dynamic changes in user-side electricity demand, the distribution of active and reactive power in the distribution network has become more complex and difficult to predict.

[0003] In modern power distribution network operation, with the increasing penetration rate of distributed photovoltaic (PV), the random fluctuations in its output pose a severe challenge to feeder voltage control. Existing power distribution network voltage regulation mainly relies on on-load tap changers (OLTCs) and reactive power support from PV inverters, and is based on control strategies with fixed times and thresholds. However, in actual engineering projects, the power distribution network measurement system often suffers from severe spatiotemporal asymmetry: substations typically possess high-frequency real-time monitoring capabilities at the second level (such as SCADA systems), while the vast user side (distributed PV access points) is often limited by communication bandwidth or centralized data collection strategies, providing only low-frequency periodic data at the minute level (e.g., 15 minutes). This measurement blind spot prevents the dispatch system from sensing the voltage fluctuations at the end of the line caused by sudden changes in PV output in real time. Therefore, existing control strategies based on fixed thresholds struggle to identify this critical risk state where "the voltage appears normal but the regulation resources are exhausted," easily leading to missed opportunities for OLTC activation and triggering voltage over-limit accidents. Summary of the Invention

[0004] To address the severe spatiotemporal asymmetry in distribution network measurement systems during practical engineering projects, leading to measurement blind spots and the inability of dispatch systems to detect voltage fluctuations at the end of the grid caused by sudden changes in photovoltaic output, and the difficulty of identifying critical risk states where "voltage appears normal but regulation resources are exhausted" using existing fixed-threshold-based control strategies, which can easily lead to missed opportunities for OLTC (Optimal Voltage Control Center) activation and voltage over-limit accidents, this invention aims to provide a method for coordinated optimization and dispatching of active and reactive power in distribution networks. The specific technical solution is as follows: acquiring grid line data, user-side voltage and active power uploaded at the user-side upload time, and various monitoring data from substations at each preset time within a preset time period. The preset time period is from the upload time to the current time, and the monitoring data includes substation voltage, reactive power, and active power. Power; The system integrates and analyzes the numerical characteristics of various monitoring data at the upload time and grid line data, compares them with the user-side voltage, and determines the voltage-power transmission coefficient; based on the numerical characteristics of various monitoring data at each preset time and grid line data, and combined with the voltage-power transmission coefficient, the predicted voltage value on the user side is obtained; based on the difference characteristics of the monitoring data at each preset time and the upload time, the changes in active power in the monitoring data within the preset time period and the difference between them and the active power on the user side at the time to be measured are corrected to obtain the corrected available capacity of the inverter at each preset time; based on the predicted voltage value on the user side and the corrected available capacity of the inverter at each preset time, the regulation burden index is determined; based on the cumulative situation of the regulation burden index at the preset time, combined with the predicted voltage value on the user side, the transformer is regulated.

[0005] Furthermore, the method for obtaining the voltage-power transmission coefficient includes: the power grid line data includes nominal resistance, nominal reactance, and rated voltage of the substation side bus; the theoretical voltage drop component at the upload time is determined based on the line voltage drop longitudinal component formula, the numerical characteristics of various monitoring data at the upload time, and the power grid line data; the difference between the substation voltage and the user-side voltage at the upload time is used as the measured voltage difference at the upload time, and the ratio of the measured voltage difference to the theoretical voltage drop component is used as the voltage-power transmission coefficient.

[0006] Furthermore, the formula model for the theoretical voltage drop component at the upload moment includes: in, This represents the theoretical voltage drop component at the upload moment; Indicates the nominal resistance; Indicates the active power at the time of upload; Indicates the nominal reactance; Indicates the reactive power at the time of upload; This indicates the rated voltage of the busbar on the substation side.

[0007] Furthermore, the method for obtaining the voltage prediction value includes: determining the theoretical voltage drop component at each preset time based on the line voltage drop longitudinal component formula, the numerical characteristics of various monitoring data at each preset time, and the power grid line data; using the product of the voltage-power transmission coefficient and the theoretical voltage drop component at each preset time as the derived voltage drop value; and using the difference between the substation voltage at each preset time and the derived voltage drop value at each preset time as the voltage prediction value on the user side at each preset time.

[0008] Furthermore, the formula model for the theoretical voltage drop component at each preset time includes: in, This represents the theoretical voltage drop component at each preset time point; Indicates the nominal resistance; This represents the active power at each preset time point; Indicates the nominal reactance; This represents the reactive power at each preset time point; This indicates the rated voltage of the busbar on the substation side.

[0009] Furthermore, the method for obtaining the corrected available capacity includes: determining the error correction coefficient for each preset time based on the difference characteristics of the monitoring data at each preset time and the upload time at each preset time; analyzing the difference between the active power at each preset time and the upload time within a preset time period, and comparing it with the active power on the user side at the upload time, thereby calculating the theoretical maximum reactive power capacity of the inverter at each preset time; and using the ratio of the theoretical maximum reactive power capacity of the inverter at each preset time to the error correction coefficient as the corrected available capacity of the inverter at each preset time.

[0010] Furthermore, the method for obtaining the error correction coefficient includes: calculating the normalized value of the Euclidean distance between the active power and reactive power of the substation at each preset time and the active power and reactive power of the substation at the upload time, using it as the operating condition deviation index, and using the sum of the operating condition deviation index and a preset constant as the error correction coefficient.

[0011] Furthermore, the method for obtaining the theoretical maximum reactive power capacity includes: taking the difference between the active power of the substation at each preset time and the active power at the upload time as the power increment at each preset time relative to the upload time; taking the product of the power increment corresponding to each preset time and a preset proportional coefficient as the theoretical change; taking the difference between the active power on the user side at the upload time and the theoretical change corresponding to each preset time as the nominal active power output of the inverter at each preset time; if the square of the rated capacity of the inverter is less than the square of the nominal active power output of the inverter at each preset time, then the theoretical maximum reactive power capacity of the inverter at each preset time is recorded as the preset value; otherwise, the difference between the square of the rated capacity of the inverter and the square of the nominal active power output of the inverter at each preset time is calculated, and the square root of the obtained difference is taken as the theoretical maximum reactive power capacity of the inverter at each preset time.

[0012] Furthermore, the method for obtaining the regulation burden index includes: calculating the absolute value of the difference between the predicted voltage value on the user side and the preset nominal voltage at each preset time, and the ratio of this difference to the preset voltage-reactive power sensitivity coefficient, as the theoretical demand at each preset time; taking the difference between the theoretical demand at each preset time and the corrected available capacity of the inverter at each preset time as the reactive power regulation deficit at each preset time; when the reactive power regulation deficit is less than or equal to 0, setting the regulation burden index at each preset time to a preset value; otherwise, taking the product of the reactive power regulation deficit at each preset time and the preset voltage-reactive power sensitivity coefficient as the regulation burden index at each preset time.

[0013] Furthermore, the adjustment of the transformer based on the cumulative adjustment burden index at preset times, combined with the voltage prediction value on the user side, includes: determining the time window at the current time in terms of timing; within the time window at the current time, integrating and summing the adjustment burden indices at all preset times to obtain the cumulative adjustment demand index; when the cumulative adjustment demand index is less than the preset cumulative trigger threshold, keeping the OLTC tap position unchanged; when the cumulative adjustment demand index is greater than or equal to the preset cumulative trigger threshold, if the voltage prediction value on the user side at the current time is less than the preset nominal voltage, then raising the OLTC tap position by one level to increase the voltage; if the voltage prediction value on the user side at the current time is greater than the preset nominal voltage, then lowering the OLTC tap position by one level to decrease the voltage.

[0014] This invention offers the following advantages: it acquires user-side voltage and active power at the time of upload, along with grid line data and various monitoring data from the substation at each preset time within a preset time period. Due to the spatiotemporal asymmetry between user-side and substation-side data, measurement blind spots can occur. Therefore, this invention integrates and analyzes the numerical characteristics of various monitoring data at the upload time, along with grid line data, and compares them with user-side voltage to determine the voltage-power transmission coefficient. An accurate transmission coefficient is crucial for constructing an accurate voltage prediction model, improving the agreement between predicted and actual voltage values. This coefficient is then combined with the numerical characteristics of various monitoring data at each preset time and grid line data to obtain the predicted voltage value for the user side at each preset time. Since the voltage prediction is based on the linear extrapolation assumption that "the operating conditions at the preset time are similar to those at the upload time," however, with the passage of time or significant fluctuations in photovoltaic power, the actual operating conditions may deviate significantly from the calibration benchmark, leading to increased linear extrapolation errors. Therefore, based on the differences in monitoring data between the substation at each preset time and the upload time, the changes in active power in the monitoring data within the preset time period and the differences in active power on the user side with the time to be measured are corrected to obtain the corrected available capacity of the inverter at each preset time. Accurate assessment of the corrected available capacity of the inverter helps to rationally allocate the output of distributed energy resources. Furthermore, based on the voltage prediction value on the user side and the corrected available capacity of the inverter at each preset time, a regulation burden index is determined. This index comprehensively considers the voltage situation and the regulation capability of the inverter. Finally, based on the cumulative regulation burden index at the preset time, combined with the voltage prediction value on the user side, the transformer is regulated. This coordinated dispatching method can better balance the distribution of active and reactive power in the power grid, improve voltage quality, and reduce line losses. Attached Figure Description

[0015] 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.

[0016] Figure 1 is a flowchart of a method for coordinated optimization scheduling of active and reactive power in a power distribution network according to an embodiment of the present invention; Figure 2 is a flowchart of a method for obtaining modified available capacity according to an embodiment of the present invention. Detailed Implementation

[0017] 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 power distribution network active-reactive power coordinated optimization scheduling method 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.

[0018] 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.

[0019] The following description, in conjunction with the accompanying drawings, details the specific scheme of the active-reactive power coordinated optimization scheduling method for power distribution networks provided by this invention.

[0020] Please refer to Figure 1, which shows a flowchart of a method for coordinated optimization scheduling of active and reactive power in a distribution network according to an embodiment of the present invention. The method includes the following steps: Step S1: Obtain power grid line data, user-side voltage and user-side active power uploaded at the upload time, and various monitoring data of the substation at each preset time in a preset time period, wherein the preset time period is the period from the upload time to the current time, and the monitoring data includes substation voltage, reactive power and active power.

[0021] This solution is designed for scenarios where distributed photovoltaic (PV) power is integrated into the distribution network. Distributed PV is typically installed on the user side, and the inverter is the core component of PV power generation, used to convert the DC power generated by the PV panels into AC power and feed it into the grid.

[0022] In the initial stage of power-on startup of the distribution network system, or in the event of a system reset due to prolonged lack of user-side data, a default state configuration is required to ensure basic operational safety. Specifically, static ledger data from the local database of the distribution network is read to obtain grid line data. The grid line data includes the nominal resistance (unit: ohms Ω, e.g., 0.13Ω / km; if the line is 5km long, the nominal resistance is 0.65Ω), nominal reactance (unit: ohms Ω, e.g., 0.35Ω / km; if the line is 5km long, the nominal reactance is 1.75Ω), and the rated voltage (unit: kilovolts kV, e.g., 10kV) of the substation-side busbars.

[0023] Before receiving the first valid user-side data uploaded at the upload time, the system initializes the voltage-power transmission coefficient to the default value of 1.0. The data uploaded by the user side at the upload time includes the user-side voltage and the user-side active power, which can be obtained through the smart meter.

[0024] Furthermore, since smart meters or inverters on the user side are typically limited by communication bandwidth and can only upload data at longer intervals (e.g., 15 minutes), while data from the substation side is usually real-time data at the second level, direct calculation would lead to timing misalignment. Therefore, an event-triggered data alignment mechanism must be established. That is, when the system receives uploaded data from the user side, it uses the time period between the upload time and the current time as a preset time period and acquires various monitoring data of the substation at each preset time within this preset time period. The types of monitoring data include active power, reactive power, and substation voltage. This monitoring data can be acquired through voltage transformers, current transformers, power meters, etc. In this embodiment of the invention, the monitoring data acquisition frequency is set to once per second.

[0025] Step S2: Integrate and analyze the numerical characteristics of various monitoring data at the upload time and the power grid line data, and compare them with the user-side voltage to determine the voltage-power transmission coefficient; Based on the numerical characteristics of various monitoring data at each preset time and the power grid line data, and combined with the voltage-power transmission coefficient, obtain the predicted voltage value on the user side.

[0026] The power distribution network is a complex system, and the transmission relationship between voltage and power is affected by various factors, such as the resistance and reactance of the power grid lines and the operating status of substations. Theoretical models or empirical values ​​alone cannot accurately describe the actual transmission situation. Therefore, by integrating and analyzing the numerical characteristics of various monitoring data and power grid line data at the upload time, the voltage-power transmission coefficient can be determined. This approach fully considers these practical factors, making the coefficient closer to the actual operating characteristics of the power grid.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the voltage-power transmission coefficient includes: based on the data acquisition step in step S1, it is known that the power grid line data includes nominal resistance, nominal reactance, and rated voltage of the substation side bus.

[0028] In power systems, resistance and reactance determine the voltage drop and power loss generated when current flows through a line. Nominal resistance reflects the heat-related portion of the line's impediment to current, while nominal reactance is primarily related to the line's inductive characteristics, affecting the voltage phase and amplitude. The rated voltage of the substation bus is an important reference value for grid operation, providing a benchmark voltage level for calculating the theoretical voltage drop component, ensuring the calculation results accurately reflect voltage changes in the actual grid. The longitudinal component formula for line voltage drop is derived based on circuit theory and electromagnetic principles. It accurately describes the relationship between the voltage drop generated by active and reactive power across resistance and reactance and line parameters and power, quantifying voltage changes during power transmission. Therefore, the theoretical voltage drop component at the upload time is first determined based on the longitudinal component formula for line voltage drop, the numerical characteristics of various monitoring data at the upload time, and grid line data. The formula model for the theoretical voltage drop component at the upload time includes: in, This represents the theoretical voltage drop component at the upload moment; Indicates the nominal resistance; Indicates the active power at the time of upload; Indicates the nominal reactance; Indicates the reactive power at the time of upload; This indicates the rated voltage of the busbar on the substation side.

[0029] In the above formula model, This reflects the contribution of active power to the voltage drop across the resistor. This reflects the voltage drop contribution of reactive power to the reactance. Therefore, the sum in the numerator reflects the total voltage loss of electrical energy in the transmission line caused by transmitting this power. By comparing this with the rated voltage of the substation side bus in the denominator, the numerator is normalized to the voltage dimension, thus obtaining the theoretical voltage drop component at the time of transmission.

[0030] Next, the difference between the substation voltage and the user-side voltage at the upload time is taken as the measured voltage difference at the upload time. The measured voltage difference reflects the actual physical voltage drop between the substation voltage and the user-side voltage uploaded by the user at the upload time.

[0031] Finally, the ratio of the measured voltage difference at the upload time to the theoretical voltage drop component is used as the voltage-power transfer coefficient. The voltage-power transfer coefficient characterizes the correction ratio for line parameter drift, topology errors, and measurement errors under actual conditions. The closer the value is to 1, the closer the actual voltage difference is to the theoretical voltage difference, and thus the higher the model accuracy. When calculating the ratio, if the theoretical voltage drop component at the upload time is 0 (a special case), the denominator is modified to the sum of the theoretical voltage drop component and a preset constant. The preset constant is used to prevent the denominator from being 0; here, a value of 0.001 can be taken.

[0032] It should be noted that in actual implementation scenarios, a reasonable range can be set for the voltage-power transmission coefficient, such as [0.5, 2.0]. When the calculated voltage-power transmission coefficient is not within this reasonable range, it means that the calculation result deviates significantly from common sense, which may be due to bad data or transient interference. In this case, the voltage-power transmission coefficient can be set to the voltage-power transmission coefficient at the previous upload time.

[0033] After completing the above data benchmark construction, for the non-measuring interval, that is, the time interval during which the user-side data is no longer updated within the preset time period, it is necessary to use the known substation-side data within the preset time period, combined with the voltage-power transmission coefficient obtained above, to dynamically estimate the end state (user side). Therefore, based on the numerical characteristics of various monitoring data at each preset time and the power grid line data, and combined with the voltage-power transmission coefficient obtained above, the voltage prediction value of the user side can be obtained.

[0034] Preferably, in one embodiment of the present invention, the method for obtaining the voltage prediction value includes: firstly, similarly, determining the theoretical voltage drop component at each preset time based on the line voltage drop longitudinal component formula, the numerical characteristics of various monitoring data at each preset time, and the power grid line data; the formula model for the theoretical voltage drop component at each preset time includes: in, This represents the theoretical voltage drop component at each preset time point; Indicates the nominal resistance; This represents the active power at each preset time point; Indicates the nominal reactance; This represents the reactive power at each preset time point; This indicates the rated voltage of the busbar on the substation side.

[0035] After obtaining the theoretical voltage drop component at each preset time, since the voltage-power transfer coefficient reflects the degree to which the theoretical voltage drop needs to be corrected, the product of the voltage-power transfer coefficient and the theoretical voltage drop component at each preset time is directly used as the derived voltage drop value. The derived voltage drop value can be used to subsequently compensate for line parameter drift and topology errors.

[0036] Finally, the substation voltage at each preset time is the measured voltage on the substation side, which is used as the reference term for voltage extrapolation. Since the user-side voltage is equal to the substation voltage minus the voltage drop on the line, the difference between the substation voltage at each preset time and the extrapolated voltage drop value at each preset time is used as the predicted voltage value on the user side at each preset time.

[0037] At this point, the predicted voltage value of the user side at each preset time within the preset time period can be obtained (the preset time does not include the time to be measured, because the voltage of the user side at the time to be measured already has a real value).

[0038] Step S3: Based on the differences between the monitoring data at each preset time and the upload time of the substation, correct the changes in active power in the monitoring data within the preset time period and the differences between the active power on the user side and the active power at the time to be measured, and obtain the corrected available capacity of the inverter at each preset time; determine the regulation burden index based on the voltage prediction value on the user side and the corrected available capacity of the inverter at each preset time.

[0039] After obtaining the predicted voltage value of the user side at each preset time within the preset time period, the predicted voltage value reflects the voltage demand of the user side (demand side). Therefore, in this step, the photovoltaic output power of the user side can be estimated to reflect whether the demand can be met.

[0040] In distributed photovoltaic (PV) high-penetration distribution network feeders, power fluctuations within short time scales (seconds to minutes) are mainly dominated by the randomness of PV output (such as cloud cover), while the traditional electricity load of users usually changes slowly and can be regarded as a quasi-steady-state component. Therefore, analyzing the changes in active power in the monitoring data within a preset time period and the difference between the active power on the user side and the active power at the time of measurement can be used to determine the theoretical maximum reactive power capacity of the inverter at each preset time. Since the actual operating conditions may deviate significantly from the calibration benchmark as time goes by or PV fluctuates greatly, the aforementioned linear extrapolation error may increase. In order to quantify this extrapolation risk, this step also calculates the error correction coefficient at each preset time by using the difference characteristics of the monitoring data at each preset time of the substation and the upload time. This is used to dynamically shrink the theoretical maximum reactive power capacity of the PV inverter, thereby obtaining the theoretically corrected usable capacity of the inverter at each preset time.

[0041] Preferably, in one embodiment of the present invention, the method for obtaining corrected available capacity includes: Please refer to Figure 2, which shows a flowchart of the method for obtaining corrected available capacity in one embodiment of the present invention. The method includes the following steps: Step S301: Determine the error correction coefficient for each preset time based on the difference characteristics of the monitoring data at each preset time and the upload time of the substation.

[0042] Substation monitoring data dynamically changes with time, grid load, equipment operating status, and other factors. The data at different preset times and the uploaded data points can differ significantly. If these differences are not considered, directly using the raw data to calculate inverter capacity will lead to a large deviation from the actual situation. Therefore, the Euclidean distance between the active and reactive power of the substation at each preset time and the active and reactive power at the uploaded time is calculated and divided by the transformer's rated capacity for normalization. This yields the operating condition deviation index. The magnitude of the operating condition deviation index directly reflects the degree of topological deviation of the grid power flow state at each preset time relative to the uploaded time. When the value is approximately 0, it indicates that the operating conditions at that preset time are almost identical to those at the uploaded time, and the aforementioned extrapolation model has extremely high reliability. When the value increases significantly, it indicates that the operating conditions may be fluctuating greatly, and the error of the aforementioned extrapolation model will increase, reducing its reliability.

[0043] Finally, the deviation index of the working condition is mapped to the error correction coefficient. In this embodiment of the invention, the sum of the deviation index and the preset constant 1 can be used as the error correction coefficient. The larger the value, the larger the error and the greater the degree of correction required.

[0044] Step S302: Within a preset time period, analyze the difference between the active power at each preset time and the upload time, and compare it with the active power on the user side at the upload time, thereby calculating the theoretical maximum reactive power capacity of the inverter at each preset time.

[0045] The difference between the active power at each preset time and the active power at the upload time is taken as the power increment at each preset time relative to the upload time. The power increment reflects the increase in active power at the substation side at each preset time relative to the upload time.

[0046] Then, using a preset proportional coefficient at the end (user side), the power increment at the beginning (substation side) at each preset time is allocated to the user side: the product of the power increment corresponding to each preset time and the preset proportional coefficient (usually set as the ratio of the photovoltaic installed capacity on the user side to the total photovoltaic installed capacity of the feeder, with a value between 0 and 1) is taken as the theoretical change, and the difference between the active power on the user side at the upload time and the theoretical change corresponding to each preset time is taken as the nominal active power output of the inverter at each preset time. If the substation side observes a power decrease (i.e., power increment < 0, assuming positive flow to the user), then ignoring load fluctuations, this usually means an increase in the photovoltaic output at the end (the photovoltaic reverse power offsets the power sent down from the beginning). Therefore, a minus sign is used in the above to reflect this reverse change relationship. Through this calculation, the trend of photovoltaic output change can be tracked in real time even in the absence of real-time communication.

[0047] Since the rated capacity of the inverter is fixed, the reactive power regulation capability of the photovoltaic inverter is strictly constrained by its capacity circle (PQ curve), meaning that its active power output and reactive power output satisfy a certain mathematical relationship, expressed as: Where S is the rated capacity, Q is the reactive power output, and P is the active power output. Therefore, if the square of the inverter's rated capacity is less than the square of the inverter's nominal active power output at each preset time, the theoretical maximum reactive power capacity of the inverter at each preset time is recorded as the preset value of 0 to ensure the inverter's safe operation and avoid overload. Otherwise, the difference between the square of the inverter's rated capacity and the square of the inverter's nominal active power output at each preset time is calculated, and the square root of this difference is taken as the theoretical maximum reactive power capacity of the inverter at each preset time.

[0048] Step S303: Use the error correction coefficient at each preset time to correct the theoretical maximum reactive capacity of the inverter, thereby obtaining the corrected usable capacity of the inverter at each preset time.

[0049] Based on the analysis in step S301, it can be seen that the larger the error correction coefficient at the preset time, the lower the reliability. That is, the inverter cannot be relied on to exert its full theoretical capacity to support the voltage. Instead, it is necessary to conservatively assume that its safe and usable capacity should be smaller. Therefore, the ratio of the inverter's theoretical maximum reactive capacity to the error correction coefficient at each preset time is used as the corrected usable capacity of the inverter at each preset time.

[0050] After determining the available capacity of the inverter at each preset time, it can be combined with the voltage prediction value on the user side to calculate the actual voltage regulation deficit, which is used to synthesize the regulation burden index.

[0051] Preferably, in one embodiment of the present invention, the method for obtaining the adjustment load index includes: firstly, calculating the theoretical reactive power deficit required to restore the user-side voltage to the reference value; calculating the absolute value of the difference between the predicted voltage value and the preset nominal voltage at each preset time, where the absolute value of the difference reflects the degree of voltage deviation; and the ratio of this absolute value of the difference to the preset voltage-reactive power sensitivity coefficient, which converts the voltage into reactive power, thereby obtaining the theoretical reactive power demand at each preset time. It should be noted that in this embodiment of the present invention, the preset nominal voltage can be set to 10KV; the preset voltage-reactive power sensitivity coefficient is defined as the voltage change caused by a unit of reactive power, and is usually taken as a positive real number, here set to 0.0005kV / kVar.

[0052] Then, the difference between the theoretical demand at each preset time and the corrected available capacity of the inverter at each preset time is calculated as the reactive power regulation deficit at each preset time. The larger the reactive power regulation deficit, the greater the degree to which the corrected available capacity at that preset time cannot meet the theoretical demand for reactive power at that preset time; if it is negative or 0, it means that it can meet the demand.

[0053] Therefore, when the reactive power regulation deficit is less than or equal to 0, the regulation burden index at each preset time is set to the preset value of 0; otherwise, the reactive power regulation deficit at each preset time is mapped back to the voltage dimension, that is, the product of the reactive power regulation deficit and the preset voltage-reactive power sensitivity coefficient is used as the regulation burden index at each preset time. The larger the regulation burden index, the more serious the problem and the higher the degree of regulation difficulty.

[0054] Step S4: Based on the cumulative adjustment load index at a preset time and combined with the voltage prediction value on the user side, adjust the transformer.

[0055] Based on the aforementioned steps, the regulation load index at each preset time point can be obtained. Since photovoltaic power output is often affected by cloud cover, resulting in drastic fluctuations on the order of seconds, the regulation load index calculated in the aforementioned steps may contain numerous transient spikes. To prevent the on-load tap-changing transformer (OLTC) from frequently operating in response to these short-term noises, this step employs a sliding window integration method to analyze the cumulative regulation load index at the preset time points, and combines this with the voltage prediction value from the user side to regulate the transformer.

[0056] Preferably, in one embodiment of the present invention, the transformer is adjusted based on the cumulative adjustment burden index at preset times and the voltage prediction value on the user side. This includes: determining the time window at the current time in terms of timing. In this embodiment of the present invention, the size of the time window is set to 60 seconds, which can be adjusted according to the implementation scenario. Within the time window at the current time, the cumulative adjustment demand index is obtained by integrating and summing the adjustment burden indices at all preset times. The larger the cumulative adjustment demand index, the greater the accumulated risk, and the more adjustment is needed; conversely, the smaller the cumulative adjustment demand index, the smaller the accumulated risk, and no adjustment is needed.

[0057] Therefore, when the cumulative adjustment demand is less than the preset cumulative trigger threshold, the OLTC tap position remains unchanged; when the cumulative adjustment demand is greater than or equal to the preset cumulative trigger threshold, the adjustment direction needs to be determined: if the current voltage prediction value on the user side is less than the preset nominal voltage, it means that the current voltage prediction value on the user side is too low, and the voltage should be increased. Therefore, the OLTC tap position is adjusted up one level to increase the voltage; if the current voltage prediction value on the user side is greater than the preset nominal voltage, it means that the current voltage prediction value on the user side is too high, and the voltage should be decreased. Therefore, the OLTC tap position is adjusted down one level to decrease the voltage.

[0058] It should be noted that in this embodiment of the present invention, the preset cumulative trigger threshold is set to 0.5 kV·s.

[0059] In summary, this method acquires user-side voltage and active power at the upload time, along with grid line data and various monitoring data from the substation at each preset time within a preset time period. Due to the spatiotemporal asymmetry between user-side and substation-side data, measurement blind spots can occur. Therefore, it integrates and analyzes the numerical characteristics of various monitoring data at the upload time, along with grid line data, and compares them with user-side voltage to determine the voltage-power transmission coefficient. An accurate transmission coefficient is crucial for constructing an accurate voltage prediction model, improving the agreement between predicted and actual voltage values. This coefficient is then combined with the numerical characteristics of various monitoring data at each preset time and grid line data to obtain the predicted voltage value for the user side at each preset time. Since the voltage prediction is based on the linear extrapolation assumption that "the operating conditions at the preset time are similar to those at the upload time," however, with the passage of time or significant fluctuations in photovoltaic power, the actual operating conditions may deviate significantly from the calibration benchmark, leading to increased linear extrapolation errors. Therefore, based on the differences in monitoring data between the substation at each preset time and the upload time, the changes in active power in the monitoring data within the preset time period and the differences in active power on the user side with the time to be measured are corrected to obtain the corrected available capacity of the inverter at each preset time. Accurate assessment of the corrected available capacity of the inverter helps to rationally allocate the output of distributed energy resources. Furthermore, based on the voltage prediction value on the user side and the corrected available capacity of the inverter at each preset time, a regulation burden index is determined. This index comprehensively considers the voltage situation and the regulation capability of the inverter. Finally, based on the cumulative regulation burden index at the preset time, combined with the voltage prediction value on the user side, the transformer is regulated. This coordinated dispatching method can better balance the distribution of active and reactive power in the power grid, improve voltage quality, and reduce line losses.

[0060] 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.

[0061] 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 method for coordinated active and reactive power scheduling in a power distribution network, characterized in that, The method includes: acquiring power grid line data, user-side voltage and active power uploaded by the user side at the upload time, and various monitoring data of the substation at each preset time within a preset time period, wherein the preset time period is the period from the upload time to the current time, and the monitoring data includes substation voltage, reactive power, and active power; fusing and analyzing the numerical characteristics of various monitoring data at the upload time and power grid line data, and comparing them with the user-side voltage to determine the voltage-power transmission coefficient; obtaining the predicted voltage value of the user side based on the numerical characteristics of various monitoring data at each preset time and power grid line data, and combining them with the voltage-power transmission coefficient; correcting the changes in active power in the monitoring data within the preset time period and its differences from the user-side active power at the time to be measured based on the differences in monitoring data at each preset time and the upload time, to obtain the corrected available capacity of the inverter at each preset time; determining the regulation burden index based on the predicted voltage value of the user side and the corrected available capacity of the inverter at each preset time; and adjusting the transformer based on the cumulative regulation burden index at the preset time and the predicted voltage value of the user side.

2. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 1, characterized in that, The method for obtaining the voltage-power transmission coefficient includes: the power grid line data including nominal resistance, nominal reactance, and rated voltage of the substation side bus; determining the theoretical voltage drop component at the upload time based on the line voltage drop longitudinal component formula, the numerical characteristics of various monitoring data at the upload time, and the power grid line data; using the difference between the substation voltage and the user-side voltage at the upload time as the measured voltage difference at the upload time, and using the ratio of the measured voltage difference to the theoretical voltage drop component as the voltage-power transmission coefficient.

3. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 2, characterized in that, The formula model for the theoretical voltage drop component at the upload moment includes: in, This represents the theoretical voltage drop component at the upload moment; Indicates the nominal resistance; Indicates the active power at the time of upload; Indicates the nominal reactance; Indicates the reactive power at the time of upload; This indicates the rated voltage of the busbar on the substation side.

4. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 1, characterized in that, The method for obtaining the voltage prediction value includes: determining the theoretical voltage drop component at each preset time based on the line voltage drop longitudinal component formula, the numerical characteristics of various monitoring data at each preset time, and the power grid line data; using the product of the voltage-power transmission coefficient and the theoretical voltage drop component at each preset time as the derived voltage drop value; and using the difference between the substation voltage at each preset time and the derived voltage drop value at each preset time as the voltage prediction value on the user side at each preset time.

5. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 4, characterized in that, The formula model for the theoretical pressure drop component at each preset time includes: in, This represents the theoretical voltage drop component at each preset time point; Indicates the nominal resistance; This represents the active power at each preset time point; Indicates the nominal reactance; This represents the reactive power at each preset time point; This indicates the rated voltage of the busbar on the substation side.

6. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 1, characterized in that, The method for obtaining the corrected available capacity includes: determining the error correction coefficient for each preset time based on the difference characteristics of the monitoring data at each preset time and the upload time at each preset time; analyzing the difference between the active power at each preset time and the upload time within a preset time period, and comparing it with the active power on the user side at the upload time, thereby calculating the theoretical maximum reactive power capacity of the inverter at each preset time; and taking the ratio of the theoretical maximum reactive power capacity of the inverter at each preset time to the error correction coefficient as the corrected available capacity of the inverter at each preset time.

7. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 6, characterized in that, The method for obtaining the error correction coefficient includes: calculating the normalized value of the Euclidean distance between the active power and reactive power of the substation at each preset time and the active power and reactive power of the substation at the upload time, which is used as the operating condition deviation index, and using the sum of the operating condition deviation index and the preset constant as the error correction coefficient.

8. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 6, characterized in that, The method for obtaining the theoretical maximum reactive power capacity includes: taking the difference between the active power of the substation at each preset time and the active power at the upload time as the power increment at each preset time relative to the upload time; taking the product of the power increment at each preset time and a preset proportional coefficient as the theoretical change; taking the difference between the active power on the user side at the upload time and the theoretical change at each preset time as the nominal active power output of the inverter at each preset time; if the square of the rated capacity of the inverter is less than the square of the nominal active power output of the inverter at each preset time, then the theoretical maximum reactive power capacity of the inverter at each preset time is recorded as the preset value; otherwise, the difference between the square of the rated capacity of the inverter and the square of the nominal active power output of the inverter at each preset time is calculated, and the square root of the difference is taken as the theoretical maximum reactive power capacity of the inverter at each preset time.

9. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 1, characterized in that, The method for obtaining the regulation burden index includes: calculating the absolute value of the difference between the predicted voltage value on the user side and the preset nominal voltage at each preset time, and the ratio of this difference to the preset voltage-reactive power sensitivity coefficient, as the theoretical demand at each preset time; taking the difference between the theoretical demand at each preset time and the corrected available capacity of the inverter at each preset time as the reactive power regulation deficit at each preset time; when the reactive power regulation deficit is less than or equal to 0, setting the regulation burden index at each preset time to a preset value; otherwise, taking the product of the reactive power regulation deficit at each preset time and the preset voltage-reactive power sensitivity coefficient as the regulation burden index at each preset time.

10. The active and reactive power coordinated optimization scheduling method for a distribution network according to claim 1, characterized in that, Based on the cumulative adjustment burden index at preset times, and combined with the voltage prediction value on the user side, the transformer is adjusted, including: in terms of timing, determining the time window of the current time; within the time window of the current time, integrating and summing the adjustment burden indices of all preset times to obtain the cumulative adjustment demand index; when the cumulative adjustment demand index is less than the preset cumulative trigger threshold, keeping the OLTC tap position unchanged; when the cumulative adjustment demand index is greater than or equal to the preset cumulative trigger threshold, if the voltage prediction value on the user side at the current time is less than the preset nominal voltage, then the OLTC tap position is increased by one level to boost the voltage; if the voltage prediction value on the user side at the current time is greater than the preset nominal voltage, then the OLTC tap position is decreased by one level to reduce the voltage.

Images