Active power distribution network operation risk assessment method and system

Abstract

The invention is suitable for the technical field of distribution network risk assessment, and provides an active distribution network operation risk assessment method and system, and the method comprises the steps: constructing a probability distribution model of wind power output, photovoltaic output, conventional load and electric vehicles; the method comprises the following steps: randomly sampling wind speed, illumination intensity, load demand and electric vehicle behavior parameters by adopting a Monte Carlo simulation method, calculating wind power output power, photovoltaic output power, conventional load power and electric vehicle load power based on a sampling result, and generating a source-load power data set under multiple space-time scenes; constructing a risk index system including a node voltage out-of-limit risk, a branch overload risk and a load loss risk, and performing power distribution network time sequence load flow calculation based on the power data set to obtain risk values of all indexes in the risk index system; calculating a comprehensive operation risk index based on each index risk value; and the comprehensive operation risk index is combined with a grading early warning rule to determine a power distribution network operation risk assessment result, and the requirement of risk assessment of a complex scene can be met.

Description

Technical Field

[0001] This invention relates to the field of distribution network risk assessment technology, and in particular to an active distribution network operation risk assessment method and system. Background Technology

[0002] With the increasing proportion of renewable energy output and the rapid growth in the number of electric vehicles, the intermittency and volatility of power output caused by natural factors and user habits are making the structure, operation, and control of distribution networks increasingly complex. To enhance the stability, reliability, and cost-effectiveness of distribution network operation and promote the construction of a digital and intelligent robust power grid, it is urgent to study the quantitative assessment and early warning management of distribution network operation risks under uncertain environments, thereby promoting accurate decision-making and efficient formulation of distribution network planning and operation. Currently, the main active distribution network operation risk assessment methods are analytical methods. Based on the distribution network architecture and the logical connections between its equipment, a system reliability probability model is constructed, and analytical techniques are used to calculate and evaluate the system's risk indicators. However, this method is only suitable for simple, small-scale systems. The computational load is too large when dealing with complex networks, making it difficult to meet the assessment needs of complex operating scenarios. Summary of the Invention

[0003] This invention provides an active distribution network operation risk assessment method to address the problem of difficulty in meeting the assessment requirements of complex operation scenarios.

[0004] The first aspect of this invention provides a method for active distribution network operation risk assessment, comprising: Construct probability distribution models for wind power output, solar irradiance output, power demand of conventional loads, and behavioral parameters of electric vehicle loads; The Monte Carlo simulation method was used to randomly sample the wind speed, light intensity, load demand and electric vehicle behavior parameters. Based on the sampling results, the wind power output, photovoltaic power output, conventional load power and electric vehicle load power were calculated to generate a source-load power dataset in multiple time and space scenarios. A risk index system is constructed that includes node voltage over-limit risk, branch overload risk and load loss risk. Based on the source-load power dataset, the distribution network time-series power flow is calculated to obtain the risk value of each index in the risk index system. Calculate the comprehensive operational risk index based on the risk values ​​of each indicator in the aforementioned risk indicator system; The comprehensive operational risk indicators are combined with the hierarchical early warning rules to determine the operational risk assessment results of the distribution network.

[0005] Furthermore, the construction of the wind speed probability distribution model for wind power output, the irradiance probability distribution model for photovoltaic power output, the power demand probability distribution model for conventional loads, and the behavioral parameter probability distribution model for electric vehicle loads includes: The wind speed probability distribution model is modeled using a Weibull distribution, and the model's shape and scale parameters are determined based on meteorological data; the light intensity probability distribution model is modeled using a Beta distribution, and the model's shape parameters are determined based on meteorological data; the power demand probability distribution model is modeled using a normal distribution, and the model's expected value and standard deviation are determined based on historical load data statistics; the electric vehicle behavior parameter probability distribution model includes daily mileage modeled using a log-normal distribution, charging start time and charging end time modeled using a normal distribution, and charging power calculated using a constant power model.

[0006] Furthermore, the Monte Carlo simulation method is used to randomly sample the wind speed, solar irradiance, load demand, and electric vehicle behavior parameters. Based on the sampling results, wind power output, photovoltaic power output, conventional load power, and electric vehicle load power are calculated to generate a source-load power dataset under multiple spatiotemporal scenarios, including: Wind speed is randomly sampled using a Weibull distribution, and wind power output is calculated based on the wind power-wind speed conversion relationship. The photovoltaic output power is calculated based on the photovoltaic power-light intensity conversion relationship by random sampling using a Beta distribution. The load demand is randomly sampled from a normal distribution to generate the conventional load power. The behavior parameters of electric vehicles are jointly sampled. The daily driving mileage is sampled based on the log-normal distribution and converted into the charging demand. The charging start time and end time are sampled based on the normal distribution. Combined with the rated power of the charging pile and the charging time, the load power of the electric vehicle is calculated. Power data from preset time scales and distribution network node spatial dimensions are integrated to form a multi-temporal and spatial scenario source-load power dataset.

[0007] Furthermore, the risk value for node voltage exceeding the limit risk in the calculated risk index system includes: Calculation of risk value for voltage exceeding the upper limit: in: for time The magnitude of the risk when node voltage exceeds the upper limit. for The probability of a risk scenario at any given moment. The total number of scenes, for time node A flag indicating whether a voltage exceeding the upper limit risk event has occurred in a given scenario. for time node Voltage values ​​in the scenario, The severity of exceeding the voltage limit is determined by the magnitude of the voltage. , This is the maximum voltage value; Calculation of risk value for voltage exceeding the lower limit: in: for time The risk level of node voltage falling below the lower limit for time node A marker indicating whether a voltage drop below the lower limit risk event has occurred in a given scenario. The severity of the voltage exceeding the lower limit is determined by the magnitude of the voltage. , This is the minimum voltage value.

[0008] Furthermore, the risk value of branch overload risk in the calculated risk index system includes: Calculation of active power overload risk value for branch circuits: in: for time The extent of risk associated with the branch road. for time branch road Line overload indicator value in the scenario for time branch road Active power in the scenario This is a function to determine the severity of line overload. , for The maximum active power of the branch circuit.

[0009] Furthermore, the risk value of load loss risk in the calculated risk index system includes: Risk of load failure due to voltage exceeding the upper limit : Risk of load failure due to voltage falling below the lower limit : Risk of loss of load due to branch overload : System failure risk : in: This represents the total number of nodes in the distribution network. This represents the total number of branches in the distribution network. , , These are the load loss ratios for the risks of voltage exceeding the upper limit, voltage exceeding the lower limit, and branch overload, respectively, which are proportional to the corresponding severity function values.

[0010] Furthermore, the calculation of the comprehensive operational risk index based on the risk values ​​of each indicator in the risk indicator system includes: Calculate the daily distribution network operation risk value: in: , The daily voltage exceeds the upper and lower limits risk value. This represents the daily branch overload risk value. This represents the daily risk value for load loss. The total number of time periods. These are the weighting coefficients.

[0011] Furthermore, the calculation of the comprehensive operational risk index includes: in: This serves as the system's daily comprehensive operational risk indicator; These are the weighting coefficients for various risk values.

[0012] Furthermore, the step of combining the comprehensive operational risk indicators with tiered early warning rules to determine the distribution network operational risk assessment results includes: Based on the proportion of unloaded risk of node voltage exceeding limits and branch overload risk, the risk levels are divided into four categories: red, orange, yellow, and green; based on the line load rate, the risk levels are also divided into four categories: red, orange, yellow, and green.

[0013] A second aspect of the present invention provides an active distribution network operation risk assessment system, comprising: The probability distribution model building unit is used to build probability distribution models for wind power output, solar irradiance output, power demand of conventional loads, and behavioral parameters of electric vehicle loads. The source-load power dataset generation unit is used to randomly sample the wind speed, light intensity, load demand and electric vehicle behavior parameters using the Monte Carlo simulation method, and calculate the wind power output, photovoltaic power output, conventional load power and electric vehicle load power based on the sampling results to generate source-load power datasets in multiple time and space scenarios. The risk value determination unit for each indicator is used to construct a risk indicator system that includes node voltage over-limit risk, branch overload risk and load loss risk. Based on the source-load power dataset, the distribution network time-series power flow calculation is performed to obtain the risk value of each indicator in the risk indicator system. The comprehensive operational risk index calculation unit is used to calculate the comprehensive operational risk index based on the risk values ​​of each index in the risk index system. The distribution network operation risk assessment result determination unit is used to determine the distribution network operation risk assessment result by combining the comprehensive operation risk indicators with the hierarchical early warning rules.

[0014] As can be seen from the above technical solutions, the present invention has the following advantages: This invention constructs a probability distribution model for wind power output, photovoltaic power output, conventional load, and electric vehicle load. Using Monte Carlo simulation for random sampling, it calculates the power output of wind power, photovoltaic power, conventional load, and electric vehicle load, generating a source-load power dataset across multiple spatiotemporal scenarios. Based on a constructed risk index system, it calculates the risk value of each index. A comprehensive operational risk index is then calculated based on these risk values. Finally, the calculated comprehensive risk index is combined with tiered early warning rules to determine the distribution network operation risk assessment result. This invention establishes a source-load probability distribution model and, combined with statistical data, uses an appropriate probability distribution function to realistically simulate natural factors such as wind speed and sunlight, as well as behavioral factors related to load and electric vehicle loads, thereby better simulating the uncertain environment of distribution network operation. It quantifies risk indicators related to system safety and stability, helping operation and maintenance personnel to manage risks more intuitively and conveniently. Finally, it solves for the power flow and risk indicators of the uncertain system, thus obtaining the system's operational status. This effectively improves assessment efficiency and accuracy when facing risk assessment needs in complex scenarios. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of an embodiment of an active distribution network operation risk assessment method according to the present invention; Figure 2 This is a diagram of an IEEE 33-node power distribution system considering DG and electric vehicle access in this invention. Figure 3 (a) is a distribution diagram of the power ratio coefficients of wind power output, photovoltaic power output, and load in this invention; Figure 3(b) is a daily power distribution diagram of the electric vehicle charging station in this invention; Figure 4 (a) is a distribution diagram of the system timing operation risk indicators in this invention; Figure 4 (b) is a distribution diagram of the risk of voltage exceeding the upper limit at each node in this invention; Figure 5 (a) is a distribution diagram of the risk of voltage exceeding the upper limit at the DG access node in this invention; Figure 5 (b) is a distribution diagram of the risk of daily voltage exceeding the upper limit at each node in this invention; Figure 6 (a) is a diagram showing the time distribution of the node load loss ratio in this invention; Figure 6 (b) is a diagram showing the distribution of branch load rate indicators at different times in this invention; Figure 7 This is a timing risk warning diagram for the DG access system in this invention; Figure 8 This is a time distribution early warning diagram of the overall load loss when the DG access system is in this invention; Figure 9 This is a distribution diagram of the operational risk indicators of the system under DG access in this invention; Figure 10 (a) is a diagram showing the time distribution of the node load loss ratio in this invention; Figure 10 (b) is a diagram showing the distribution of branch load rate indicators at different times in this invention; Figure 11 This is a timing risk warning diagram for the DG and electric vehicle access system in this invention; Figure 12 This invention provides an early warning of the time distribution of comprehensive load loss when DG and electric vehicles are connected to the system. Detailed Implementation

[0016] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding to,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] Example 1 The implementation method in this embodiment can be implemented in a system, on a server, or on a terminal; no specific limitation is made. The following section will describe the active distribution network operation risk assessment method in this application from the perspective of system implementation. Please refer to... Figure 1 The method provided in this application includes: 101. Construct a probability distribution model for wind speed of wind power output, a probability distribution model for irradiance of photovoltaic power output, a probability distribution model for power demand of conventional loads, and a probability distribution model for behavioral parameters of electric vehicle loads; In this embodiment, the wind speed probability distribution model adopts the Weibull distribution model, and the shape and scale parameters of the model are determined based on meteorological data, as follows: In the formula: For wind speed, and The shape and scale parameters of this distribution are used to reflect the shape and average magnitude of the wind speed.

[0018] The relationship between wind turbine output and wind speed is modeled as follows: In the formula: This refers to the output value of the wind turbine generator; This refers to the rated (maximum) output of the wind turbine generator set; , , These are the cut-in, rated, and cut-out wind speeds, respectively.

[0019] Wind turbines often use an asynchronous generator model, and their connection to the system node can be simplified to a PQ node. Through the automatic switching compensator in the unit, the power factor can be kept constant, with a value of 0.85-0.95.

[0020] The relationship between active and reactive power in a wind turbine can be represented as follows: In the formula: This refers to the reactive power output of the wind turbine generator. This refers to the power factor of the wind turbine.

[0021] Wind power output varies depending on wind speed at different times of the day. According to the formula for calculating wind power output, the daily characteristic model of wind power output shows a certain proportional relationship with the temporal distribution of wind speed. However, wind speed data comes from weather forecasts. Due to errors in the weather forecast model and the actual wind speed collected, the predicted daily wind power output is not entirely accurate and differs from the measured values. To consider the practical significance of the simulation model, the daily characteristic model of wind power output consists of predicted values ​​and an error function. The error function can be considered as a non-standard normal distribution; therefore, the daily characteristic function of wind power output is shown below: In the formula: and The actual and predicted values ​​of wind power output; The standard deviation of this function is related to the length of the wind power forecast period, the scale, and the standard capacity.

[0022] In this embodiment, the light intensity probability distribution model uses a Beta distribution, and the model's shape parameters are determined based on meteorological data; specifically as follows: In the formula: and These represent the maximum and real-time light intensity during that time period, respectively. and These are the two shape parameters of the distribution.

[0023] The relationship between photovoltaic unit output and solar irradiance is shown in the following model: In the formula: The power output of photovoltaics; This represents the maximum output of distributed photovoltaic power. ; The size of the area to be covered by photovoltaic panels; This refers to the efficiency of converting light energy into electrical energy.

[0024] Photovoltaic output can also be approximated as a PQ node, because the power factor of a photovoltaic power generation system can be maintained at a constant value by the automatic switching of capacitor banks.

[0025] The relationship between the active power and reactive power of a photovoltaic power unit is shown below: In the formula: The reactive power component that powers the photovoltaic unit; This is the power factor of the output.

[0026] Photovoltaic power output efficiency varies with the intensity of sunlight throughout the day, and output intensity directly reflects the strength of sunlight. The error in the daily characteristics of photovoltaic power output mainly comes from the discrepancy between weather forecast data and actual data. This error function can be considered a non-standard normal distribution, as shown below: In the formula: The actual value of photovoltaic power output; Forecast values ​​for photovoltaic power output; The standard deviation of the error function is related to the rated capacity of the photovoltaic unit.

[0027] In this embodiment, the power demand probability distribution model adopts a normal distribution model, and the expected value and standard deviation of the model are determined based on historical load data statistics; specifically as follows: In the formula: and For load power, and This represents the mathematical expectation of the load power. and This represents the standard deviation of the load power.

[0028] Daily power fluctuations in load demand are a statistical pattern revealed through in-depth analysis of rich historical load data. While this pattern is accurate when judged over a long period, its fluctuation over a single day is influenced by varying daily conditions and potential unforeseen events, requiring consideration of the impact of daily uncertainties on the statistical pattern. Therefore, the daily load characteristic model also consists of a statistical regularity function and an error function. The error function can be considered a non-standard normal distribution, and the overall function is shown below: In the formula: and The actual and predicted values ​​of the load power; is the standard deviation of the error function, which is related to the load forecast value.

[0029] In this embodiment, the probability distribution model for electric vehicle behavior parameters includes modeling daily mileage using a log-normal distribution, modeling charging start and end times using a normal distribution, and calculating charging power using a constant power model; specifically as follows: The mileage traveled by an electric vehicle in a day follows a log-normal distribution, with the following function: In the formula: This represents the expected daily mileage. ; The variance of daily mileage. .

[0030] Whenever an electric vehicle is charged, the start time of charging is actually the time when the vehicle arrives at the charging station. This time point follows a normal distribution, and its function is as follows: In the formula: The expected value for the charging start time. ; This represents the variance of the charging start time. .

[0031] The time an electric vehicle leaves a charging station is the time it stops charging, and its function is as follows: In the formula: The expected value for the charging start time. ; This represents the variance of the charging start time. .

[0032] Each charging station in the charging station is considered to charge the electric vehicle using a constant power, as shown in the following function: In the formula: Start time for charging electric vehicles; The end time of charging the electric vehicle; This represents the power value of the charging station.

[0033] To align with the daily characteristic curves of new energy DG output and conventional load, it is assumed that each electric vehicle leaves the charging station only after it is fully charged. At the same time, electric vehicles entering the charging station during a certain period are considered to start charging from the next hour, thus facilitating the construction of the daily characteristic curve of electric vehicle load.

[0034] 102. The Monte Carlo simulation method was used to randomly sample wind speed, light intensity, load demand and electric vehicle behavior parameters. Based on the sampling results, wind power output, photovoltaic power output, conventional load power and electric vehicle load power were calculated to generate source-load power datasets in multiple time and space scenarios. In this embodiment, the process of generating source-side and load-side power datasets is as follows: 1. Randomly sample wind speeds using a Weibull distribution and calculate wind power output based on the wind power-wind speed conversion relationship; 2. Randomly sample the light intensity using a Beta distribution, and calculate the photovoltaic output power based on the photovoltaic power-light intensity conversion relationship; 3. Perform normal distribution random sampling on the load demand to generate conventional load power; 4. Jointly sample the behavioral parameters of electric vehicles, extract daily mileage based on log-normal distribution, convert it into charging demand, extract charging start time and end time based on normal distribution, and calculate the load power of electric vehicles by combining the rated power of charging piles and charging time. 5. Integrate power data from preset time scales and distribution network node spatial dimensions to form a multi-temporal and spatial scenario source-load power dataset.

[0035] 103. Construct a risk index system that includes node voltage over-limit risk, branch overload risk and load loss risk. Perform distribution network time-series power flow calculation based on source-load power dataset to obtain the risk value of each index in the risk index system. Distribution network time-series operational risk assessment is a process of evaluating the system's operational status at different times, using a 24-hour timescale. It primarily focuses on the time perspective, assessing the risk values ​​in real-time during actual system operation. The time-series risk assessment index comprises four standards for measuring operational status: voltage exceeding limits, line overload, load shedding, and overall risk. The relevant indicator function for a specific node at a given time is as follows: Calculation of risk value for voltage exceeding the upper limit: in: for time The magnitude of the risk when node voltage exceeds the upper limit. for The probability of a risk scenario at any given moment. The total number of scenes, for time node A flag indicating whether a voltage exceeding the upper limit risk event has occurred in a given scenario. for time node Voltage values ​​in the scenario, The severity of exceeding the voltage limit is determined by the magnitude of the voltage. , This is the maximum voltage value; Calculation of risk value for voltage exceeding the lower limit: in: for time The risk level of node voltage falling below the lower limit for time node A marker indicating whether a voltage drop below the lower limit risk event has occurred in a given scenario. The severity of the voltage exceeding the lower limit is determined by the magnitude of the voltage. , This is the minimum voltage value.

[0036] Calculation of active power overload risk value of branch circuit: in: for time The extent of risk associated with the branch road. for time branch road Line overload indicator value in the scenario for time branch road Active power in the scenario This is a function to determine the severity of line overload. , for The maximum active power of the branch circuit.

[0037] 104. Calculate the comprehensive operational risk index based on the risk values ​​of each indicator in the risk indicator system; The system's load shedding risk is a more severe risk caused by node voltage exceeding limits and line active power exceeding limits. That is, when the system node voltage exceeding limits or branch power exceeding limits exceeds a certain severity, the system will disconnect part of the load to maintain stable operation. The specific calculation is as follows: Risk of load failure due to voltage exceeding the upper limit : Risk of load failure due to voltage falling below the lower limit : Risk of loss of load due to branch overload : System failure risk : in: The total number of distribution network nodes. This represents the total number of branches in the distribution network. , , These are the load loss ratios for risks of voltage exceeding the upper limit, voltage exceeding the lower limit, and branch overload, respectively, which are proportional to the corresponding severity function values. , , This is the ratio of the load loss ratio to the severity. This refers to the voltage value exceeding the maximum severity limit. This is the voltage value that represents the maximum severity level when the voltage exceeds the lower limit. This represents the power value when the branch circuit exceeds its limit most severely.

[0038] Calculate the daily distribution network operation risk value: in: , The daily voltage exceeds the upper and lower limits risk value. This represents the daily branch overload risk value. This represents the daily risk value for load loss. The total number of time periods. These are the weighting coefficients.

[0039] The comprehensive operational risk indicators are: in: This serves as the system's daily comprehensive operational risk indicator; These are the weighting coefficients for various risk values.

[0040] 105. Combine comprehensive operational risk indicators with tiered early warning rules to determine the results of distribution network operation risk assessment.

[0041] In this embodiment, the risk levels are divided into four categories: red, orange, yellow, and green, based on the proportion of load loss of node voltage over-limit risk and branch overload risk; and based on the line load rate, the risk levels are also divided into four categories: red, orange, yellow, and green.

[0042] Specifically, according to the relevant regulations of the power grid company, the risk level of the distribution network can be determined based on its load shedding ratio. For example, it is meticulously divided into four levels, and the color system used for weather warnings is adopted for identification. The specific classification criteria and color levels are shown in Table 1: Table 1 Classification of Risk Levels Based on Load Loss Ratio The system's load underload ratio is determined by the risk of node voltage exceeding upper and lower limits and branch overload. Spatially, the load underload ratio risk level early warning can include two parts: node level and system level. Specifically, the node's load underload ratio is determined by a function of the severity of the node's voltage exceeding the limit, while the system's load underload ratio is determined by the load underload status of each node plus a weighted average of the node's power proportion.

[0043] When designing and planning a distribution network, the rated load capacity of the lines is usually set relatively high to account for the load growth brought about by future system development. This reduces the probability of line overload. Therefore, to reflect the actual power situation of the distribution network branches during system operation, the load factor is used. Risk level indicators serve as early warning indicators. Specific classification criteria and color-coded levels are shown in Table 2. Table 2 Classification of Line Load Rate Risk Level Indicators The following simulation analysis uses an improved IEEE 33-node power distribution system. The specific nodes, types, and capacities are shown in Table 1. The improved IEEE node power distribution system is as follows: Figure 2 As shown: Table 3 Access Capacity and Location Distribution Figure 3 This describes the daily characteristic parameter distribution of the source-load environment model. Figure 3 (a) is based on the daily characteristic model of wind power output, photovoltaic power output, and load demand. Without considering the error function, it uses the peak value (rated value) of each component as the benchmark to obtain the power ratio coefficient at each moment under the daily characteristic environment. Figure 3 (b) is a simulation sampling based on the daily characteristic model of electric vehicle distribution. The area has a total of 200 electric vehicles, and the daily power distribution of electric vehicle charging stations is obtained by fitting.

[0044] 1. The experimental process for risk assessment and early warning of distribution network operation considering DG is as follows: like Figure 4As shown in (a), considering the daily risks related to DG connection and load demand, the main daily risks are voltage exceeding the upper limit and load shedding caused by it. The voltage exceeding the upper limit risk occurs between 11:00 and 18:00, with no risk occurring at other times. Based on the daily characteristic ratio coefficient changes, the wind power output coefficient is greater than 0.5 between 11:00 and 20:00, and the photovoltaic output coefficient is greater than 0.5 between 8:00 and 17:00. The risk occurs when the total DG output is greater than 50%. During this period, the distribution network cannot fully absorb the power input from the renewable DG, leading to voltage increases and the corresponding risk. In the early morning and nighttime hours, the photovoltaic DG does not receive sunlight, and the wind speed remains low, resulting in low DG output and a risk-free system.

[0045] Figure 4 As shown in (b), the risk of voltage exceeding the upper limit mainly occurs at nodes 10-18 and 31-33. Based on the DG access location and access point risk, it can be seen that the regional voltage rise is more severe due to the input of new energy DG in this node range, resulting in the risk of voltage exceeding the upper limit. The overvoltage risk value is highest at node 14.

[0046] Combination Figure 5 (a) Analysis shows that the timing of the voltage exceedance risk at nodes 10-18 coincides with the peak wind power output period, as nodes 14 and 18 are connected to wind-generated distributed generation (DG). Nodes 31-33 are primarily connected to photovoltaic (PV) power, with the occurrences occurring around noon when the proportional-to-voltage ratio is 1. The reason why no voltage exceedance risk occurred at node 8, which was connected to wind-generated DG, is that there is a relatively long load upstream to absorb the power, effectively eliminating the risk. Observation Figure 5 (b) The risk of exceeding the daily voltage limit is highest at node 14.

[0047] By classifying the intermediate variables related to the above risk assessment indicators into different levels, a three-dimensional distribution of the node load failure ratio at different times can be obtained. Figure 6 (a) 3D time distribution of branch load rate ( Figure 6 (b) Distribution of risk warnings for distribution network nodes and branches in different time periods of the system ( Figure 7 ) and the time distribution of the overall system load loss ( Figure 8 ).

[0048] 2. The experimental process for risk assessment and early warning of power distribution network operation considering DG and electric vehicles is as follows: like Figure 9As shown, the risk of voltage exceeding the upper limit occurs during the daytime (12:00-17:00), while the risk of voltage exceeding the lower limit occurs at 21:00 at night, accompanied by the risk of load shedding. No related risks were observed during other times. Combining the daily characteristic power distribution, it can be concluded that the system experiences a certain power supply deficit due to the overlap between the peak load of charging stations and the evening peak load of regular loads. Furthermore, the reduced output of photovoltaic power and decreased wind speed at night lead to a decrease in DG output, resulting in a certain power supply deficit in the distribution network, which in turn causes the system to experience low voltage risks.

[0049] Considering the temporal risk distribution in the scenario of risk assessment and early warning for distributed generation (DG) in the distribution network, significant changes in the overall daily load characteristics of the system were observed after charging stations were connected to the distribution network system. During the daytime, the load of electric vehicle charging stations successfully absorbed the surplus power injected into the distribution network by DG, effectively reducing the risk of voltage exceeding the upper limit. However, at night, the charging demand of electric vehicle charging stations exacerbated the evening peak load, posing a certain challenge to the distribution network regarding the risk of voltage exceeding the lower limit.

[0050] By classifying the intermediate variables related to the above risk assessment indicators into different levels, a three-dimensional distribution of the node load failure ratio at different times can be obtained. Figure 10 (a) 3D time distribution of branch load rate ( Figure 10 (b) Distribution of risk warnings for distribution network nodes and branches in different time periods of the system ( Figure 11 ) and the time distribution of the overall system load loss ( Figure 12 ).

[0051] Comparing the risk warning situations under two different scenarios can more intuitively demonstrate the impact of electric vehicle charging station loads on the distribution network operation risk, and also reflect the reliability of the assessment method.

[0052] First, after the electric vehicle load was connected, no high-risk nodes were identified in the distribution network system during the daily assessment cycle. Simultaneously, compared to the 12-16 time period when distributed generation (DG) was connected alone, the number of medium-risk and higher-risk warning nodes in the system was effectively reduced. The number of medium-risk warning nodes with DG connection was 12, and the number of medium-risk warning nodes with DG and electric vehicle connection was 11, a reduction of 8.33%. The number of higher-risk warning nodes with DG connection was 20, and the number of higher-risk warning nodes with DG and electric vehicle connection was 13, a reduction of 35.00%. However, the load from electric vehicles also puts some pressure on the line load. During 8:00-9:00, 17:00-18:00, and 23:00-24:00, branches 1-2 are under medium load warning; from 18:00 to 23:00, branches 1-2 and 2-3 are under medium load warning. Compared to DG's independent access, there are two more medium load warning lines.

[0053] The overall system load shedding ratio diagram shows that the connection of electric vehicle loads effectively reduces the system's load shedding ratio. After the connection of electric vehicle charging stations, the period during which the overall load shedding ratio occurs is shortened from 12:00-19:00 to 12:00-18:00, a reduction of one hour. At the same time, the load shedding ratio during the period when the overall load shedding risk occurs is effectively reduced.

[0054] Example 2 An embodiment of an active distribution network operation risk assessment system of the present invention includes the following steps: The probability distribution model building unit is used to build probability distribution models for wind power output, solar irradiance output, power demand of conventional loads, and behavioral parameters of electric vehicle loads. The source-load power dataset generation unit is used to randomly sample wind speed, light intensity, load demand and electric vehicle behavior parameters using the Monte Carlo simulation method, and calculate wind power output, photovoltaic power output, conventional load power and electric vehicle load power based on the sampling results, generating source-load power datasets in multiple time and space scenarios. The risk value determination unit for each indicator is used to construct a risk indicator system that includes node voltage over-limit risk, branch overload risk and load loss risk. Based on the source-load power dataset, the distribution network time-series power flow calculation is performed to obtain the risk value of each indicator in the risk indicator system. The comprehensive operational risk index calculation unit is used to calculate the comprehensive operational risk index based on the risk values ​​of each indicator in the risk index system. The distribution network operation risk assessment result determination unit is used to determine the distribution network operation risk assessment result by combining comprehensive operation risk indicators with hierarchical early warning rules.

[0055] For specific limitations regarding the system, please refer to the method limitations described above, which will not be repeated here. Each module in the above system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0056] It is understood that those skilled in the art can combine various implementation methods in the above embodiments under the guidance of the above examples to obtain technical solutions with multiple implementation methods.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for assessing operational risks in an active distribution network, characterized in that, include: Construct probability distribution models for wind power output, solar irradiance output, power demand of conventional loads, and behavioral parameters of electric vehicle loads; The Monte Carlo simulation method was used to randomly sample the wind speed, light intensity, load demand and electric vehicle behavior parameters. Based on the sampling results, the wind power output, photovoltaic power output, conventional load power and electric vehicle load power were calculated to generate a source-load power dataset in multiple time and space scenarios. A risk index system is constructed that includes node voltage over-limit risk, branch overload risk and load loss risk. Based on the source-load power dataset, the distribution network time-series power flow is calculated to obtain the risk value of each index in the risk index system. Calculate the comprehensive operational risk index based on the risk values ​​of each indicator in the aforementioned risk indicator system; The comprehensive operational risk indicators are combined with the hierarchical early warning rules to determine the operational risk assessment results of the distribution network.

2. The active distribution network operation risk assessment method according to claim 1, characterized in that, The construction of the wind speed probability distribution model for wind power output, the irradiance probability distribution model for photovoltaic output, the power demand probability distribution model for conventional loads, and the behavioral parameter probability distribution model for electric vehicle loads includes: The wind speed probability distribution model is modeled using a Weibull distribution, and the model's shape and scale parameters are determined based on meteorological data; the light intensity probability distribution model is modeled using a Beta distribution, and the model's shape parameters are determined based on meteorological data; the power demand probability distribution model is modeled using a normal distribution, and the model's expected value and standard deviation are determined based on historical load data statistics; the electric vehicle behavior parameter probability distribution model includes daily mileage modeled using a log-normal distribution, charging start time and charging end time modeled using a normal distribution, and charging power calculated using a constant power model.

3. The active distribution network operation risk assessment method according to claim 1, characterized in that, The Monte Carlo simulation method is used to randomly sample the wind speed, solar irradiance, load demand, and electric vehicle behavior parameters. Based on the sampling results, wind power output, photovoltaic power output, conventional load power, and electric vehicle load power are calculated to generate a source-load power dataset under multiple spatiotemporal scenarios, including: Wind speed is randomly sampled using a Weibull distribution, and wind power output is calculated based on the wind power-wind speed conversion relationship. The photovoltaic output power is calculated based on the photovoltaic power-light intensity conversion relationship by random sampling using a Beta distribution. The load demand is randomly sampled from a normal distribution to generate the conventional load power. The behavior parameters of electric vehicles are jointly sampled. The daily driving mileage is sampled based on the log-normal distribution and converted into the charging demand. The charging start time and end time are sampled based on the normal distribution. Combined with the rated power of the charging pile and the charging time, the load power of the electric vehicle is calculated. Power data from preset time scales and distribution network node spatial dimensions are integrated to form a multi-temporal and spatial scenario source-load power dataset.

4. The active distribution network operation risk assessment method according to claim 1, characterized in that, The risk values ​​for node voltage exceedance risk in the calculated risk index system include: Calculation of risk value for voltage exceeding the upper limit: in: for time The magnitude of the risk when node voltage exceeds the upper limit. for The probability of a risk scenario at any given moment. The total number of scenes, for time node A flag indicating whether a voltage exceeding the upper limit risk event has occurred in a given scenario. for time node Voltage values ​​in the scenario The severity of exceeding the voltage limit is determined by the magnitude of the voltage. , This is the maximum voltage value; Calculation of risk value for voltage exceeding the lower limit: in: for time The magnitude of risk when the node voltage falls below the lower limit for time node A marker indicating whether a voltage drop below the lower limit risk event has occurred in a given scenario. The severity of the voltage exceeding the lower limit is determined by the magnitude of the voltage. , This is the minimum voltage value.

5. The active distribution network operation risk assessment method according to claim 4, characterized in that, The risk values ​​for branch overload risk in the calculated risk index system include: Calculation of active power overload risk value for branch circuits: in: for time The extent of risk associated with the branch road. for time branch road Line overload indicator value in the scenario for time branch road Active power in the scenario This is a function to determine the severity of line overload. , for The maximum active power of the branch circuit.

6. The active distribution network operation risk assessment method according to any one of claims 4 or 5, characterized in that, The risk values ​​for load decompression risk in the calculated risk index system include: Risk of load failure due to voltage exceeding the upper limit : Risk of load failure due to voltage falling below the lower limit : Risk of loss of load due to branch overload : System failure risk : in: This represents the total number of nodes in the distribution network. This represents the total number of branches in the distribution network. , , These are the load loss ratios for the risks of voltage exceeding the upper limit, voltage exceeding the lower limit, and branch overload, respectively, which are proportional to the corresponding severity function values.

7. The active distribution network operation risk assessment method according to claim 6, characterized in that, The calculation of the comprehensive operational risk index based on the risk values ​​of each indicator in the risk indicator system includes: Calculate the daily distribution network operation risk value: in: , The daily voltage exceeds the upper and lower limits risk value. This represents the daily branch overload risk value. This represents the daily risk value for load loss. The total number of time periods. These are the weighting coefficients.

8. The active distribution network operation risk assessment method according to claim 7, characterized in that, The comprehensive operational risk index includes: in: This serves as the system's daily comprehensive operational risk indicator; These are the weighting coefficients for various risk values.

9. The active distribution network operation risk assessment method according to claim 1, characterized in that, The step of determining the distribution network operation risk assessment result by combining the comprehensive operation risk indicators with the hierarchical early warning rules includes: Based on the proportion of load loss for node voltage over-limit risk and branch overload risk, the risk levels are divided into four categories: red, orange, yellow, and green; based on the line load rate, the risk levels are also divided into four categories: red, orange, yellow, and green.

10. An active distribution network operation risk assessment system, characterized in that, The method described by any one of claims 1-9 comprises: The probability distribution model building unit is used to build probability distribution models for wind power output, solar irradiance output, power demand of conventional loads, and behavioral parameters of electric vehicle loads. The source-load power dataset generation unit is used to randomly sample the wind speed, light intensity, load demand and electric vehicle behavior parameters using the Monte Carlo simulation method, and calculate the wind power output, photovoltaic power output, conventional load power and electric vehicle load power based on the sampling results to generate source-load power datasets in multiple time and space scenarios. The risk value determination unit for each indicator is used to construct a risk indicator system that includes node voltage over-limit risk, branch overload risk and load loss risk. Based on the source-load power dataset, the distribution network time-series power flow calculation is performed to obtain the risk value of each indicator in the risk indicator system. The comprehensive operational risk index calculation unit is used to calculate the comprehensive operational risk index based on the risk values ​​of each index in the risk index system. The distribution network operation risk assessment result determination unit is used to determine the distribution network operation risk assessment result by combining the comprehensive operation risk indicators with the hierarchical early warning rules.

Images