Distribution network risk assessment method

Abstract

The present invention proposes a distribution network risk assessment method, which relates to the technical field of distribution network risk assessment. To address the risk factors faced by existing distributed renewable energy power supply methods, such as unstable power supply, reverse flow, and islanding effects, this method uses a machine learning model to analyze big data information on various renewable energy sources to derive power supply structure information consisting of power generation information, power generation share information, and power supply stability information. Based on the analyzed power supply structure information, information on influencing factors corresponding to various renewable energy sources is then obtained, thereby accurately predicting the power generation of the distribution network within a preset timeframe. Furthermore, based on the remaining power value, the method assesses the risk of the distribution network from three aspects: power supply stability, reverse flow, and islanding effects.

Description

Technical Field

[0001] The present invention belongs to the technical field of risk assessment, and in particular relates to a technical method for risk assessment of a distribution network. Background Art

[0002] The large-scale integration of new grid-connected entities, such as distributed renewable energy, may introduce additional security risks and management challenges to the distribution network. For example, the output stability of renewable energy sources like solar and wind power is closely related to factors such as weather. Furthermore, when there is a mismatch between power supply and demand, the reverse flow of excess power can also introduce risks such as reverse power flow and islanding effects. Summary of the Invention

[0003] The purpose of the present invention is to provide a distribution network risk assessment method to solve the technical problems in the prior art such as low power supply stability, power flow reverse transmission and islanding effect caused by the grid connection of distributed renewable energy.

[0004] The present invention proposes a distribution network risk assessment method, which includes:

[0005] S1: Determine first power supply structure information according to first power supply information of a first distribution network in a first preset time interval;

[0006] The first power supply structure information includes the total power generation information of the first distribution network in the first preset time interval, the power generation proportion information of various types of renewable energy, and the power supply stability information of various types of renewable energy;

[0007] S2: determining a plurality of first influencing factor type information according to the first power supply structure information, and acquiring first influencing factor information within a second preset time interval according to the plurality of first influencing factor type information;

[0008] S3: Obtaining first power supply quantity prediction information according to the first power supply structure information and the first influencing factor information;

[0009] S4: Acquire first remaining power information according to the first power supply prediction information and the first power demand information;

[0010] S5: Obtain a first risk assessment result according to the first power supply structure information and the first remaining power information.

[0011] Preferably, the S1 includes the following sub-steps:

[0012] S11: Acquire first power supply information of the first distribution network within a first preset time interval; wherein the first power supply information is in the form of a data sequence corresponding to a time sequence;

[0013] S12: Determine first total power generation information of the first distribution network within the first preset time interval and first sub-power generation information corresponding to each type of new energy source based on the first power supply information;

[0014] S13: Acquire first power supply structure information according to the first power supply information, the first total power generation information, and the plurality of first sub-power generation information.

[0015] Preferably, the S13 includes the following sub-steps:

[0016] S131: Utilize the ratio of each first sub-power generation information to the first total power generation information as first power generation proportion information;

[0017] S132: Acquire a plurality of first power supply amount change curves corresponding to various types of new energy sources based on the first power supply information;

[0018] S133: Inputting the plurality of first power supply amount variation curves into a power supply stability determination model respectively to determine a plurality of first power supply stability information;

[0019] S134: Obtain the first power supply structure information according to the first total power generation information, the first power generation proportion information, and the plurality of first power supply stability information.

[0020] Preferably, the S2 includes the following sub-steps:

[0021] S21: Determine multiple first new energy power supply types according to the first power supply structure information;

[0022] S22: Determine a plurality of first influencing factor type information according to the plurality of first new energy power supply types;

[0023] S23: Acquire first influencing factor information corresponding to a plurality of first influencing factor type information within a second preset time interval.

[0024] Preferably, S3 includes the following sub-steps:

[0025] S31: Acquire multiple first historical power supply information from the first power supply structure information;

[0026] S32: Acquire a plurality of second influencing factor information corresponding to each piece of the first historical power supply information from the first influencing factor information, thereby obtaining a plurality of first power supply information sets;

[0027] S33: Inputting a plurality of first power supply information sets into the power supply amount information determination model to obtain the first power supply amount prediction information.

[0028] Preferably, the S5 includes the following sub-steps:

[0029] S51: Determine power supply stability risk information based on the first power supply structure information;

[0030] S52: Determine power flow reverse risk information based on the first surplus power information;

[0031] S53: Determine islanding risk information based on the first remaining power information and the first maintenance plan;

[0032] S54: Obtain the first risk assessment result according to the power supply stability risk information, the power flow reverse risk information, and the islanding risk information.

[0033] The distribution network risk assessment method proposed in the present invention aims to address the risk factors such as unstable power supply, reverse flow and islanding effect faced by the distributed renewable energy power supply mode in the existing technology. Based on the big data information of various renewable energy power supplies, the method uses a machine learning model to analyze and obtain power supply structure information consisting of power generation information, power generation ratio information and power supply stability information. Then, based on the power supply structure information obtained from the analysis, the influencing factor information corresponding to each type of renewable energy is obtained, so as to accurately predict the power generation of the distribution network within a preset time in the future. Then, based on the residual power value, the risk of the distribution network is assessed from three aspects: power supply stability, reverse flow and islanding effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0035] Figure 1 It is an execution flow chart of the distribution network risk assessment method of the present invention;

[0036] Figure 2 It is an execution flow chart of step S1 in the distribution network risk assessment method of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The exemplary embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0039] The distribution network risk assessment method of the present invention is described in detail below.

[0040] This embodiment proposes a distribution network risk assessment method, the method flow is as follows: Figure 1 As shown, the specific steps include:

[0041] S1: Determine first power supply structure information based on multiple first new energy historical information of a first distribution network.

[0042] With the diversification of power generation energy types, more and more new energy types are used to generate electricity. In this step, it is necessary to determine the first power supply structure information corresponding to the first distribution network grid based on the new energy historical information of the first distribution network grid within the first preset time interval.

[0043] The first power supply structure information includes information such as the new energy power supply type and power supply proportion corresponding to the first distribution network grid.

[0044] Said S1 comprises the following steps, the specific steps are as follows: Figure 2 As shown:

[0045] S11: Obtain first power supply information of the first distribution network within a first preset time interval.

[0046] The first power supply information is in the form of a data sequence corresponding to a time sequence. The data sequence is power supply information corresponding to various types of new energy sources within the first preset time interval and with a preset time interval as a step size.

[0047] S12: Determine first total power generation information of the first distribution network within the first preset time interval and first sub-power generation information corresponding to each type of new energy according to the first power supply information.

[0048] The calculation method of the first total power generation information is specifically to add the result of the data sequence of the first distribution network grid in the first preset time interval, which represents the total power generation of the first distribution network grid in the first preset time interval.

[0049] The first sub-generation capacity information is specifically calculated by summing the data sequences of each type of renewable energy source in the first distribution network within the first preset time interval. That is, each first sub-generation capacity information represents the total generation of each type of renewable energy source within the first preset time interval. The sum of each first sub-generation capacity information should be equal to the first total generation capacity information.

[0050] S13: Acquire first power supply structure information according to the first power supply information, the first total power generation information, and the plurality of first sub-power generation information.

[0051] The first power supply structure information includes the total power generation information of the first distribution network in the first preset time interval, the power generation proportion information of various types of new energy sources, and the power supply stability information of various types of new energy sources.

[0052] The S13 includes the following sub-steps:

[0053] S131: Utilize the ratio of each first sub-power generation information to the first total power generation information as first power generation proportion information.

[0054] Since the first total power generation information is the sum of multiple first sub-power generation information, multiple first power generation proportion information can be obtained by calculating the ratio information of the two.

[0055] S132: Acquire first power supply amount change curves corresponding to various types of new energy sources according to the first power supply information.

[0056] Since the first power supply information includes a power supply data change sequence corresponding to each type of new energy, the first power supply change curve can be obtained by fitting the power supply data corresponding to each time point in the change sequence.

[0057] S133: Inputting the plurality of first power supply amount variation curves into a power supply stability determination model respectively to determine a plurality of first power supply stability information.

[0058] The power supply stability determination model is obtained by training a convolutional neural network model. The specific training process is as follows:

[0059] First, power supply sample data is obtained, wherein each piece of the power supply sample data includes a plurality of standard power supply amount change curves corresponding to the specific type of new energy under a plurality of power supply stability information.

[0060] For example, for solar energy, a power supply sample data includes a standard power supply change curve corresponding to the solar power supply stability of the distribution network at values ​​such as 50%, 60%, and 90% within a preset time interval.

[0061] Then, the power supply sample data is used to train a convolutional neural network model to obtain the power supply stability determination model. In the power supply sample data, the historical power supply change curve of the distribution network is used as input and the power supply stability is used as output.

[0062] Through this step, a plurality of first power supply stability information corresponding to various types of new energy sources can be obtained.

[0063] S134: Obtain the first power supply structure information according to the first total power generation information, the first power generation proportion information, and the plurality of first power supply stability information.

[0064] The first total power generation information and the first power generation proportion information can display the total power generation of the first distribution network grid and each type of new energy within the first preset time interval, and can also obtain the power supply stability of each type of new energy within the first preset time interval.

[0065] S2: Determine a plurality of first influencing factor type information according to the first power supply structure information, and obtain first influencing factor information within a second preset time interval according to the plurality of first influencing factor type information.

[0066] The first influencing factor type information includes the main influencing factors corresponding to each new energy power supply type, including meteorological factors, hydrological factors, etc.

[0067] The S2 includes the following sub-steps:

[0068] S21: Determine multiple first new energy power supply types according to the first power supply structure information.

[0069] The first power supply structure information includes first power generation proportion information and first power supply stability information corresponding to various types of new energy.

[0070] In order to assess the risk of the distribution network, it is necessary to identify the types of renewable energy that have a greater impact on power generation and power generation stability. Therefore, in this step, it is mainly necessary to determine the types of renewable energy that have a large proportion of power generation and poor power supply stability corresponding to the first distribution network using the first power supply structure information. The main reason for using the above standards to determine the types of renewable energy is that: for renewable energy types with large power generation and good power supply stability, the probability of risk is often small; for renewable energy types with small power generation, usually there will not be too much risk due to unstable power supply.

[0071] Taking the above factors into consideration, we can identify several first-generation new energy power supply types that have a greater impact on the distribution network structure risk.

[0072] Preferably, the multiple new energy power supply types corresponding to the first power supply structure information can be sorted according to the probability of causing power supply risk, and the new energy power supply type with a probability greater than a preset value can be used as the first new energy power supply type.

[0073] S22: Determine multiple first influencing factor type information according to multiple first new energy power supply types.

[0074] Among them, each of the first new energy power supply types has a corresponding relationship with several influencing factors, for example, solar energy has a direct influencing relationship with sunlight exposure.

[0075] Preferably, after determining multiple influencing factor type information based on multiple first new energy power supply types, the multiple influencing factor type information can be weighted according to the probability that each first new energy power supply type can cause distribution network grid risk, thereby obtaining multiple first influencing factor type information.

[0076] S23: Acquire first influencing factor information corresponding to a plurality of first influencing factor type information within a second preset time interval.

[0077] In order to predict the power generation of the first distribution network in the future second preset time interval, in this step, it is necessary to obtain data corresponding to the determined multiple first influencing factor type information, and then form the first influencing factor information.

[0078] S3: Obtain first power supply quantity prediction information according to the first power supply structure information and the first influencing factor information.

[0079] In order to calculate the remaining power information that will cause power flow reversal and islanding effects in subsequent steps, in this step, it is necessary to determine the first power supply forecast information corresponding to the first distribution network grid in the future period based on the first power supply structure information and the first influencing factor information.

[0080] The S3 includes the following sub-steps:

[0081] S31: Acquire multiple pieces of first historical power supply information from the first power supply structure information.

[0082] Each of the first historical power supply information corresponds to a certain new energy type. For example, the plurality of the first historical power supply information may include solar power generation, wind power generation, etc. within the first historical time interval.

[0083] S32: Acquire a plurality of second influencing factor information corresponding to each piece of the first historical power supply information from the first influencing factor information, and thereby obtain a plurality of first power supply information sets.

[0084] Since each of the first historical power supply information corresponds to a certain new energy type, the corresponding multiple second influencing factor information can be determined in the first influencing factor information according to the new energy type, and each of the first historical power supply information and the corresponding multiple second influencing factor information can be combined into the first power supply information set, thereby obtaining multiple first power supply information sets, and each of the first power supply information sets corresponds to a certain type of new energy type.

[0085] For example, the first influencing factor information may include sunlight intensity, wind conditions, etc. For a new energy type such as solar energy, sunlight intensity may be determined from the first influencing information as the second influencing factor information.

[0086] S33: Inputting a plurality of first power supply information sets into the power supply amount information determination model to obtain the first power supply amount prediction information.

[0087] The power supply information determination model is obtained by training a convolutional neural network model, and the training process is as follows:

[0088] First, sample data is obtained. Each piece of sample data consists of multiple sets of historical power supply information and corresponding power supply amounts. For example, if a piece of sample data includes solar, wind, and nuclear power generation types, then the sample data includes three power supply information sets corresponding to solar, wind, and nuclear energy, respectively. Each power supply information set includes the power supply amount corresponding to that type of renewable energy within a specified historical time range, as well as information on the corresponding influencing factors, and includes the total power generation amount within a preset future time range.

[0089] Secondly, the sample data is used to train a convolutional neural network model to obtain the power supply information determination model. The training process is completed by taking multiple power supply information sets in the sample data as input and the total power generation within a future preset time range as output.

[0090] S4: Acquire first remaining power information according to the first power supply prediction information and the first power demand information.

[0091] In this step, it is necessary to predict the first power demand information corresponding to the first distribution network framework within a preset time range in the future, and use the difference between the first power supply prediction information and the first power demand information as the first remaining power information.

[0092] Preferably, the ratio of the first remaining power information to the first power supply amount prediction information may be used as additional information of the first remaining power information.

[0093] S5: Obtain a first risk assessment result according to the first power supply structure information and the first remaining power information.

[0094] In this step, the first power supply structure information and the first remaining power information obtained in the above steps are used to evaluate and obtain the first risk assessment result from three aspects: power supply stability, power flow reverse transmission and islanding effect.

[0095] The S5 comprises the following sub-steps:

[0096] S51: Determine power supply stability risk information according to the first power supply structure information.

[0097] Since the first power supply structure information includes a plurality of first power supply stability information corresponding to each new energy type, appropriate calculations can be performed on the plurality of first power supply stability information to obtain the power supply stability information. The power supply stability information is a probability value of a risk.

[0098] S52: Determine power flow reverse risk information based on the first surplus power information.

[0099] In this step, the probability of power flow reverse occurrence is determined based on the relationship between the first remaining power information and a first threshold value, and this probability is used as power flow reverse risk information. The first threshold value represents the critical remaining power value that will cause a power flow reverse risk. The probability of power flow reverse occurrence can be determined based on the relationship between the first remaining power information and the first threshold value.

[0100] S53: Determine island risk information according to the first remaining power information and the first maintenance plan.

[0101] In this step, the degree of matching between the first remaining power information and the first maintenance plan within a preset time interval is used as islanding risk information.

[0102] Among them, the first maintenance plan refers to a plan to arrange maintenance personnel to carry out distribution network grid maintenance within a preset time interval. If the maintenance plan and the remaining power information have a large overlap within the same time period, the probability of islanding risk will be greater.

[0103] The reason for obtaining the above-mentioned flow backflow risk information and island risk information is that when the surplus power value is large, the backflow will cause an adverse impact on the distribution network grid, thereby causing the distribution network to overload the threshold or bring safety hazards to maintenance personnel.

[0104] S54: Obtain the first risk assessment result according to the power supply stability risk information, the power flow reverse risk information, and the islanding risk information.

[0105] The distribution network risk assessment method proposed in the present invention aims to address the risk factors such as unstable power supply, reverse flow and islanding effect faced by the distributed renewable energy power supply mode in the existing technology. Based on the big data information of various renewable energy power supplies, the method uses a machine learning model to analyze and obtain power supply structure information consisting of power generation information, power generation ratio information and power supply stability information. Then, based on the power supply structure information obtained from the analysis, the influencing factor information corresponding to each type of renewable energy is obtained, so as to accurately predict the power generation of the distribution network within a preset time in the future. Then, based on the residual power value, the risk of the distribution network is assessed from three aspects: power supply stability, reverse flow and islanding effect.

[0106] The above description is only a preferred embodiment of the present invention. Therefore, any equivalent changes or modifications made according to the structure, characteristics and principles described in the scope of the patent application of the present invention are included in the scope of the patent application of the present invention.

Claims

1. A method for risk assessment of a distribution network structure, wherein the distribution network uses multiple types of renewable energy sources for power supply, characterized in that: The method includes: S1: Determine first power supply structure information according to first power supply information of a first distribution network in a first preset time interval; The first power supply structure information includes the total power generation information of the first distribution network in the first preset time interval, the power generation proportion information of each type of renewable energy, and the power supply stability information of each type of renewable energy; the power supply stability information of each type of renewable energy is obtained by analyzing a plurality of first power supply change curves corresponding to each type of renewable energy; S2: determining a plurality of first influencing factor type information according to the first power supply structure information, and acquiring first influencing factor information within a second preset time interval according to the plurality of first influencing factor type information; The first influencing factor type information includes meteorological factors and hydrological factors; S3: Obtaining first power supply quantity prediction information according to the first power supply structure information and the first influencing factor information; S4: Acquire first remaining power information according to the first power supply prediction information and the first power demand information; S5: Obtaining a first risk assessment result according to the first power supply structure information and the first remaining power information; The S1 includes the following sub-steps: S11: Acquire first power supply information of the first distribution network within a first preset time interval; wherein the first power supply information is in the form of a data sequence corresponding to a time sequence; S12: Determine first total power generation information of the first distribution network within the first preset time interval and first sub-power generation information corresponding to each type of new energy source based on the first power supply information; S13: Acquire first power supply structure information according to the first power supply information, the first total power generation information, and the plurality of first sub-power generation information; The S2 includes the following sub-steps: S21: Determine multiple first new energy power supply types according to the first power supply structure information; S22: Determine a plurality of first influencing factor type information according to the plurality of first new energy power supply types; S23: Acquire first influencing factor information corresponding to a plurality of first influencing factor type information within a second preset time interval.

2. The distribution network risk assessment method according to claim 1, characterized in that: The S13 includes the following sub-steps: S131: Utilize the ratio of each first sub-power generation information to the first total power generation information as first power generation proportion information; S132: Acquire a plurality of first power supply amount change curves corresponding to various types of new energy sources based on the first power supply information; S133: Inputting the plurality of first power supply amount variation curves into a power supply stability determination model respectively to determine a plurality of first power supply stability information; S134: Obtain the first power supply structure information according to the first total power generation information, the first power generation proportion information, and the plurality of first power supply stability information.

3. The distribution network risk assessment method according to claim 2, characterized in that: The S3 includes the following sub-steps: S31: Acquire multiple first historical power supply information from the first power supply structure information; S32: Acquire a plurality of second influencing factor information corresponding to each piece of the first historical power supply information from the first influencing factor information, thereby obtaining a plurality of first power supply information sets; S33: Inputting a plurality of first power supply information sets into a power supply quantity information determination model to obtain the first power supply quantity prediction information.

4. The distribution network risk assessment method according to claim 3, characterized in that: The S5 comprises the following sub-steps: S51: Determine power supply stability risk information based on the first power supply structure information; S52: Determine power flow reverse risk information based on the first surplus power information; S53: Determine islanding risk information based on the first remaining power information and the first maintenance plan; S54: Obtain the first risk assessment result according to the power supply stability risk information, the power flow reverse risk information, and the islanding risk information.

Images