A distributed new energy power distribution network optimization scheduling method and system

Abstract

The application relates to the technical field of power distribution scheduling, in particular to a distributed new energy power distribution network optimization scheduling method and system. The method is characterized in that: the power distribution network information model is constructed through step S1; the mutation prediction result and the mutation state are acquired through step S2; the scheduling cost is calculated and the target execution decision is output through step S3; and the scheduling optimization process is feedback optimized through step S4. The system comprises a power grid information acquisition module, a weather mutation prediction module, a new energy distribution monitoring module and a scheduling feedback optimization module. The application collects target power distribution network information, constructs a power distribution network information model according to the target power distribution network information, predicts weather mutation, and outputs the best target execution decision, so that the power distribution network can be efficiently optimized and scheduled, and the power generation cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of power distribution dispatching technology, and in particular to an optimized dispatching method and system for distributed new energy power distribution networks. Background Technology

[0002] The penetration rate of distributed new energy sources, represented by photovoltaic and wind power, in the distribution network is rapidly increasing. However, the volatility, randomness, and weak support of new energy sources pose severe challenges to the traditional distribution network dispatching mode. At the same time, the ever-changing weather conditions will also affect photovoltaic power generation in real time, further increasing dispatching costs. It is urgent to optimize dispatching technology to save dispatching costs and improve dispatching efficiency, so as to achieve safe and economical operation.

[0003] Chinese patent CN120454199A discloses a power dispatch optimization method and system for power distribution networks, addressing the technical problem of high computational complexity in traditional power dispatch optimization methods, leading to poor computational efficiency in power dispatch optimization schemes. The method includes constructing an initial classification tree model based on the optimal topology schemes and network operating states under multiple power distribution network operating scenarios; globally optimizing the tree structure of the initial classification tree model using a mixed integer programming method based on a preset feature selection and partitioning mechanism to generate an optimal classification tree model; and upon receiving the current operating state of the power distribution network, solving the corresponding optimization problem based on the current operating state using the optimal classification tree model and mathematical programming to generate a target power dispatch scheme. However, this scheme still suffers from problems such as a lack of prediction for sudden weather changes, resulting in poor forward-looking optimization, and a lack of optimization for dispatch costs, leading to high dispatch costs and low optimization efficiency. Summary of the Invention

[0004] To address this, the present invention provides an optimized scheduling method and system for distributed new energy distribution networks, which overcomes the problems of poor forward-looking optimization scheduling due to the lack of prediction of sudden weather changes in the prior art, as well as the lack of optimization of scheduling costs, resulting in high scheduling costs and low optimization scheduling efficiency.

[0005] To achieve the above objectives, the present invention provides an optimized scheduling method for distributed renewable energy distribution networks, the method comprising:

[0006] Step S1: Collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model;

[0007] Step S2 involves obtaining the mutation prediction results based on the distribution network information model, updating the distribution network information model based on the mutation prediction results, and optimizing the model update process based on the target distribution network information.

[0008] Step S3: Obtain the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state; calculate the scheduling cost based on the target distribution network information; and output the target execution decision based on the scheduling cost and the new energy distribution network scheduling decision.

[0009] Step S4: The cost calculation process of scheduling cost is corrected based on the number of charge and discharge cycles, and the cost correction process is calibrated based on the sudden change state. The distribution network information model is updated a second time based on the meteorological impact value of the distribution points, and the meteorological optimization is performed on the second update process based on the proportion of sudden change duration.

[0010] Further, in step S1, when constructing the distribution network information model based on the target distribution network information using the distribution network information model construction method, the distribution network information model construction method includes:

[0011] Step A01: Construct the distribution network information visualization model based on the layout coordinates to obtain the distribution network information visualization model;

[0012] Step A02: Based on the target distribution network information, perform data embedding processing on the distribution network information visualization model to obtain the distribution network information model.

[0013] Further, in step S2, when the mutation prediction result is obtained based on the target distribution network information and the distribution network information model is updated based on the mutation prediction result, the target distribution network information and cloud thickness are input into the power generation mutation prediction model to obtain the mutation prediction result and the predicted mutation time t output by the power generation mutation prediction model. The mutation prediction result includes whether a mutation exists or not.

[0014] When the mutation prediction result is that there is no mutation, the distribution network information model will not be updated;

[0015] When the mutation prediction result indicates the existence of a mutation, the predicted mutation occurrence time t is compared with the preset predicted mutation occurrence time t0. Based on the comparison result, a judgment is made regarding the predicted mutation occurrence time, and the distribution network information model is updated according to the judgment result. Wherein:

[0016] When t≥t0, the predicted time of sudden change is determined to be long, and the distribution network information model is not updated.

[0017] When t < t0, the predicted time of sudden change is determined to be short, and the distribution network information model is updated: the predicted time of sudden change and the single-point power generation FD are input into the power generation adjustment model to obtain the single-point power generation adjustment amount m output by the power generation adjustment model. The updated single-point power generation Fm is calculated based on the single-point power generation FD and the single-point power generation adjustment amount m, and Fm is set to FD + m to obtain the updated single-point power generation Fm. The single-point power generation FD is replaced with the updated single-point power generation Fm to obtain the updated target distribution network information. The data embedding process of the distribution network information visualization model is then re-performed based on the updated target distribution network information.

[0018] Further, in step S2, when optimizing the model based on the target distribution network information using a model optimization method, the model optimization method includes:

[0019] Step B01: Input the target distribution network information into the mutation area prediction model to obtain the mutation status output by the mutation area prediction model. The mutation status includes the mutation trend and the mutation area percentage S. The mutation trend includes an upward trend and a downward trend.

[0020] Step B02: When the mutation trend is downward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model update process based on the determination result, wherein:

[0021] When S≥S0, the mutation area ratio is determined to be large, and no model optimization is performed during the model update process.

[0022] When S < S0, the mutation area ratio is determined to be small, and the model update process is optimized by replacing the mutation prediction result of "mutation exists" with "mutation does not exist".

[0023] Step B03: When the mutation trend is upward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model for the preset predicted mutation time t0 based on the determination result, wherein:

[0024] When S < S0, the mutation area ratio is determined to be small, and the model is not optimized for the preset predicted mutation time t0.

[0025] When S≥S0, the mutation area ratio is determined to be large area. The model is then optimized based on the model optimization coefficient cv at the preset predicted mutation time t0, with cv = 1.47 - 0.26 × e -（S-S0）Where e is the base of the natural logarithm, the optimized preset predicted mutation time t0' is obtained, t0' is set to t0×cv, the preset predicted mutation time t0 is replaced with the optimized preset predicted mutation time t0', and the predicted mutation time t is re-compared with the preset predicted mutation time t0.

[0026] Furthermore, in step S3, when obtaining the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state, the target distribution network information and the sudden change state are input into the new energy distribution network scheduling decision tree to obtain the new energy distribution network scheduling decision output by the new energy distribution network scheduling decision tree.

[0027] In step S3, when calculating the dispatch cost based on the target distribution network information and outputting the target execution decision based on the dispatch cost and the new energy distribution network dispatch decision, the dispatch cost Y is calculated based on the point generation cost information TP={TP1,TP2,TP3,...,TPn}, and Y=TP1+TP2+TP3+……+TPn is set to obtain the dispatch cost Y;

[0028] The scheduling cost Y is compared with the preset scheduling cost Y0. Based on the comparison result, the status of the scheduling cost is determined, and based on the determination result, a target execution decision is output, wherein:

[0029] When Y < Y0, the scheduling cost is determined to be low, and the new energy distribution network scheduling decision is output as the target execution decision.

[0030] When Y≥Y0, the scheduling cost is determined to be high. The scheduling decision of the new energy distribution network is optimized by manual adjustment to obtain the optimized new energy distribution network scheduling decision, and the optimized new energy distribution network scheduling decision is output as the target execution decision.

[0031] Further, in step S4, when adjusting the scheduling cost calculation process based on the number of charge / discharge cycles, the first deployment point's charge / discharge cycle np1 is compared with the preset charge / discharge cycle np0. The status of the first deployment point's charge / discharge cycle is determined based on the comparison result, and the scheduling cost calculation process is adjusted based on the determination result. Wherein:

[0032] When np1≤np0, the state of the first deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0033] When np1 > np0, the state of the first deployment point's charging and discharging count is determined to be abnormal, and cost correction is performed on the scheduling cost calculation process: the generation scheduling cost TP1 of the first deployment point is corrected according to the cost correction coefficient cv, and cv = 1.33 - 0.22 × e -（np1-np0） Where e is the base of the natural logarithm, the power generation scheduling cost TP1' of the first deployment point after correction is obtained, TP1' = TP1 × cv is set, the power generation scheduling cost TP1 of the first deployment point is replaced with the power generation scheduling cost TP1' of the first deployment point after correction, and the scheduling cost Y is recalculated according to the power generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point;

[0034] The second charging / discharging count np2 is compared with the preset charging / discharging count np0. Based on the comparison result, the status of the second charging / discharging count is determined, and the scheduling cost calculation process is adjusted according to the determination result.

[0035] When np2≤np0, the state of the second deployment point charging and discharging times is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0036] When np2 > np0, the state of the second deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP2 of the second deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP2' of the second deployment point. TP2' is set to TP2 × cv. The power generation scheduling cost TP2 of the second deployment point is replaced with the corrected power generation scheduling cost TP2' of the second deployment point. The scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}.

[0037] The third deployment point's charge / discharge count np3 is compared with the preset charge / discharge count np0. Based on the comparison result, the status of the third deployment point's charge / discharge count is determined, and the scheduling cost calculation process is adjusted according to the determination result.

[0038] When np3≤np0, the state of the third deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0039] When np3 > np0, the state of the third deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP3 of the third deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP3' of the third deployment point. TP3' is set to TP3 × cv. The power generation scheduling cost TP3 of the third deployment point is replaced with the corrected power generation scheduling cost TP3' of the third deployment point, and the scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}.

[0040] ...

[0041] The number of charge / discharge cycles at the nth location (npn) is compared with the preset number of charge / discharge cycles (np0). Based on the comparison result, the status of the number of charge / discharge cycles at the nth location is determined, and the cost calculation process for scheduling costs is adjusted based on the determination result.

[0042] When npn≤np0, the state of the nth point charging and discharging count is determined to be normal, and no cost correction is performed in the calculation process of scheduling cost;

[0043] When npn > np0, the state of the number of charge and discharge cycles at the nth deployment point is determined to be an abnormal state, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TPn of the nth deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TPn' of the nth deployment point. TPn' is set to TPn × cv. The power generation scheduling cost TPn of the nth deployment point is replaced with the corrected power generation scheduling cost TPn' of the nth deployment point. The scheduling cost Y is recalculated according to the power generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point.

[0044] Furthermore, in step S4, when performing cost calibration for the cost correction process based on the mutation state, if the mutation state is on an upward trend and S≥S0, cost calibration is performed on the preset charge-discharge number np0 based on the calibration coefficient ck, where ck = 0.64 + 0.23 × e -（S-S0） Where e is the base of the natural logarithm, the preset charge-discharge count np0' after calibration is obtained, np0' = np0 × ck is set, and the preset charge-discharge count np0' after calibration is rounded to obtain the preset charge-discharge count np00 after calibration and rounding. The preset charge-discharge count np0 is replaced with the preset charge-discharge count np00 after calibration and rounding. The charge-discharge counts np1, np2, np3, ..., npn of the first point are compared with the preset charge-discharge count np0 again.

[0045] Further, in step S4, when updating the distribution network information model a second time based on the meteorological impact value at the designated location, the distribution information and cloud thickness are input into the meteorological impact model at the designated location to obtain the meteorological impact value BD output by the model. The meteorological impact value BD is then compared with a preset meteorological impact value BD0. Based on the comparison result, the state of the meteorological impact value at the designated location is determined, and the distribution network information model is updated a second time based on the determination result. Wherein:

[0046] When BD≤BD0, the status of the meteorological impact value at the distribution point is determined to be acceptable, and no secondary update is performed on the distribution network information model;

[0047] When BD > BD0, the meteorological impact value of the distribution point is determined to be unacceptable, and the distribution network information model is updated a second time: the boundary of the distribution point coordinates is expanded to obtain the updated distribution point coordinates, the distribution point coordinates are replaced with the updated distribution point coordinates, and the distribution network information visualization model is reconstructed based on the distribution point coordinates.

[0048] Further, in step S4, the secondary update process is meteorologically optimized based on the proportion of mutation duration. The proportion of mutation duration tp is calculated based on the short duration td and the total monitoring time tz, and tp = td / tz is set to obtain the mutation duration proportion tp. This mutation duration proportion tp is compared with a preset mutation duration proportion tp0. The state of the mutation duration proportion is judged based on the comparison result, and meteorological optimization is performed on the preset meteorological impact value BD0 based on the judgment result. Wherein:

[0049] When tp≤tp0, the percentage of sudden change duration is determined to be the normal percentage, and no meteorological optimization is performed on the preset meteorological impact value BD0.

[0050] When tp > tp0, the percentage of abrupt change duration is determined to be an abnormal percentage. Based on the meteorological optimization coefficient cq, the preset meteorological impact value BD0 is optimized, with cq set to 0.63 + 0.25 × e. -（tp-tp0） Where e is the base of the natural logarithm, the optimized preset meteorological impact value BD0' is obtained, BD0' is set to BD0×cq, the preset meteorological impact value BD0 is replaced with the optimized preset meteorological impact value BD0', and the meteorological impact value BD0 is re-compared with the preset meteorological impact value BD0.

[0051] On the other hand, the present invention also provides a system for an optimized scheduling method of a distributed renewable energy distribution network, the system comprising:

[0052] The power grid information acquisition module is used to collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model.

[0053] The weather change prediction module acquires change prediction results based on the distribution network information model, updates the distribution network information model based on the change prediction results, acquires the change status based on the target distribution network information, and optimizes the model based on the change status during the model update process.

[0054] The new energy deployment monitoring module acquires new energy distribution network scheduling decisions based on target distribution network information and sudden change status, calculates scheduling costs based on target distribution network information, and outputs target execution decisions based on scheduling costs and new energy distribution network scheduling decisions.

[0055] The scheduling feedback optimization module corrects the scheduling cost calculation process based on the number of charging and discharging cycles, calibrates the cost correction process based on sudden change states, updates the distribution network information model a second time based on the meteorological impact value of the distribution points, and optimizes the second update process based on the proportion of sudden change duration.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows: The method collects target distribution network information and constructs a distribution network information model in step S1, so as to intuitively represent the target distribution network information in the form of three-dimensional map annotation, thereby improving the efficiency of distribution network optimization scheduling. The method also obtains the mutation prediction results and mutation status in step S2, so as to respond to possible future weather mutations in a timely manner and ensure the stable operation of the distribution network under adverse weather conditions, thereby improving the forward-looking nature and environmental adaptability of the distribution network. The method also calculates the scheduling cost and outputs the target execution decision in step S3, so as to evaluate the scheduling decision of the new energy distribution network and further obtain the target execution decision that saves scheduling costs. The method also performs feedback optimization on the scheduling optimization process in step S4, so as to calibrate the importance of individual deployment points in the scheduling cost calculation, thereby further improving the control of scheduling costs and improving the efficiency of scheduling optimization. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating the optimized scheduling method for distributed new energy distribution networks in this embodiment.

[0058] Figure 2 This is a flowchart illustrating the distribution network information model construction method in this embodiment;

[0059] Figure 3 This is a flowchart illustrating the model optimization method in this embodiment;

[0060] Figure 4 This is a schematic diagram of the system structure of the optimized scheduling method for distributed new energy distribution networks in this embodiment. Detailed Implementation

[0061] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0062] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0063] It should be noted that in the description of this invention, the terms "upper," "lower," "left," "right," "inner," and "outer," etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is merely for ease of description and does not indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0064] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0065] Please see Figure 1 As shown, this is a flowchart illustrating the optimized scheduling method for a distributed renewable energy distribution network in this embodiment. The method includes:

[0066] Step S1: Collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model;

[0067] Step S2 involves obtaining the mutation prediction results based on the distribution network information model, updating the distribution network information model based on the mutation prediction results, and optimizing the model update process based on the target distribution network information.

[0068] Step S3: Obtain the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state; calculate the scheduling cost based on the target distribution network information; and output the target execution decision based on the scheduling cost and the new energy distribution network scheduling decision.

[0069] Step S4: The cost calculation process of scheduling cost is corrected based on the number of charge and discharge cycles, and the cost correction process is calibrated based on the sudden change state. The distribution network information model is updated a second time based on the meteorological impact value of the distribution points, and the meteorological optimization is performed on the second update process based on the proportion of sudden change duration.

[0070] Specifically, the optimized scheduling method for distributed new energy distribution networks is applied to terminals of new energy distribution network optimized scheduling systems, such as solar photovoltaic panels and wind turbines. The method collects target distribution network information, constructs a distribution network information model based on this information, predicts sudden weather changes, and outputs the optimal target execution decision. This facilitates efficient optimized scheduling of the distribution network and reduces power generation costs. Specifically, step S1 involves collecting target distribution network information and constructing a distribution network information model to visually represent the target distribution network information using a 3D map, thereby improving the efficiency of optimized scheduling of the distribution network. The method also acquires the mutation prediction results and mutation status in step S2 to respond promptly to possible future weather changes and ensure the stable operation of the distribution network under adverse weather conditions, thereby improving the forward-looking nature and environmental adaptability of the distribution network. The method also calculates the scheduling cost and outputs the target execution decision in step S3 to evaluate the scheduling decision of the new energy distribution network and further obtain the target execution decision that saves scheduling costs. The method also performs feedback optimization on the scheduling optimization process in step S4 to calibrate the importance of individual deployment points in the scheduling cost calculation, thereby further improving the control of scheduling costs and improving the efficiency of scheduling optimization.

[0071] Specifically, in step S1, target distribution network information is collected. This target distribution network information includes distribution information and point-based generation cost information TP={TP1,TP2,TP3,...,TPn}, where TP1 is the generation dispatch cost of the first point, TP2 is the generation dispatch cost of the second point, TP3 is the generation dispatch cost of the third point, TPn is the generation dispatch cost of the nth point, and n is the order of generation dispatch costs and is a positive integer. The distribution information includes point coordinates, single-point energy storage, single-point charging and discharging status, single-point generation power, current total energy storage, current total power, and number of charging and discharging operations np={np1,np2,np3,...,npn}, where np1 is the number of charging and discharging operations of the first point, np2 is the number of charging and discharging operations of the second point, np3 is the number of charging and discharging operations of the third point, and npn is the number of charging and discharging operations of the nth point.

[0072] Specifically, the deployment coordinates refer to the geographical location of a single deployment point in space; a single deployment point refers to an electrical device deployed at a specific location in the distribution network according to the deployment coordinates; single-point energy storage refers to the rated total electrical energy that the energy storage system of a single deployment point can store; single-point charge / discharge status refers to the charging and discharging status of a single deployment point at the current moment; single-point power generation refers to the real-time active power output value of a single deployment point; current total energy storage refers to the total electrical energy stored by all single deployment points at the current moment; current total power refers to the sum of the power generation of all single points at the current moment; deployment point power generation cost information refers to the set of power generation scheduling costs of a single deployment point in the distribution network; and the number of charge / discharge cycles refers to the set of the number of times all single deployment points perform charge / discharge operations. This embodiment does not limit the specific methods for obtaining deployment point power generation cost information, deployment coordinates, single-point energy storage, single-point charge / discharge status, single-point power generation, current total energy storage, current total power, and number of charge / discharge cycles. Those skilled in the art can freely choose according to actual needs, such as obtaining them through the energy management system of the distribution network.

[0073] Specifically, in step S1, target distribution network information is collected to facilitate subsequent optimized scheduling of the distribution network based on the target distribution network information, thereby improving the efficiency of optimized scheduling.

[0074] Specifically, in step S1, when constructing the distribution network information model based on the target distribution network information using the distribution network information model construction method, the distribution network information model construction method includes:

[0075] Step A01: Construct the distribution network information visualization model based on the layout coordinates to obtain the distribution network information visualization model;

[0076] Step A02: Based on the target distribution network information, perform data embedding processing on the distribution network information visualization model to obtain the distribution network information model.

[0077] Specifically, the distribution network information visualization model refers to a three-dimensional model that can intuitively reflect the distribution location of points in the distribution network, obtained by constructing a basic map of the distribution network based on the coordinates of the distribution points. This embodiment does not limit the specific construction method of the distribution network information visualization model. Those skilled in the art can freely choose according to actual needs, such as constructing the distribution network information visualization model using 3ds Max software. 3ds Max refers to professional three-dimensional computer graphics software, its full name is Autodesk 3ds Max. The data embedding process refers to the process of marking the specific location of the target distribution network information in the distribution network information visualization model. This embodiment does not limit the specific method of data embedding process. Those skilled in the art can freely choose according to actual needs, such as using the text tool of 3ds Max to create 3D text labels containing the target distribution network information and binding the 3D text labels containing the target distribution network information to the distribution point coordinates.

[0078] Specifically, in step S1, a distribution network information model is constructed to visually represent the target distribution network information using a three-dimensional map annotation, thereby improving the efficiency of optimizing and scheduling the distribution network.

[0079] Specifically, in step S2, when the mutation prediction result is obtained based on the target distribution network information and the distribution network information model is updated based on the mutation prediction result, the target distribution network information and cloud thickness are input into the power generation mutation prediction model to obtain the mutation prediction result and the predicted mutation time t output by the power generation mutation prediction model. The mutation prediction result includes whether a mutation exists or not.

[0080] When the mutation prediction result is that there is no mutation, the distribution network information model will not be updated;

[0081] When the mutation prediction result indicates the existence of a mutation, the predicted mutation occurrence time t is compared with the preset predicted mutation occurrence time t0. Based on the comparison result, a judgment is made regarding the predicted mutation occurrence time, and the distribution network information model is updated according to the judgment result. Wherein:

[0082] When t≥t0, the predicted time of sudden change is determined to be long, and the distribution network information model is not updated.

[0083] When t < t0, the predicted time of sudden change is determined to be short, and the distribution network information model is updated: the predicted time of sudden change and the single-point power generation FD are input into the power generation adjustment model to obtain the single-point power generation adjustment amount m output by the power generation adjustment model. The updated single-point power generation Fm is calculated based on the single-point power generation FD and the single-point power generation adjustment amount m, and Fm is set to FD + m to obtain the updated single-point power generation Fm. The single-point power generation FD is replaced with the updated single-point power generation Fm to obtain the updated target distribution network information. The data embedding process of the distribution network information visualization model is then re-performed based on the updated target distribution network information.

[0084] Specifically, the cloud thickness refers to the thickness of the clouds in the sky at the current moment. This embodiment does not limit the specific method for obtaining the cloud thickness; those skilled in the art can freely choose according to actual needs, such as obtaining the cloud thickness through a laser cloud height meter and millimeter-wave cloud radar. The power generation mutation prediction model refers to a recurrent neural network model that takes the target distribution network information and cloud thickness as input data and the mutation prediction result and the predicted mutation time as output data. This embodiment does not limit the specific construction method of the power generation mutation prediction model; those skilled in the art can freely choose according to actual needs, such as using historical target distribution network information and cloud thickness as first mutation prediction data, using historical mutation prediction results and the predicted mutation time as second mutation prediction data, and using the first mutation prediction data and its corresponding second mutation prediction data as a training set to train the recurrent neural network model to obtain the power generation mutation prediction model. The predicted mutation time refers to the predicted time length of future abnormal power generation in the distribution network caused by weather mutations, obtained according to the power generation mutation prediction model. The preset predicted mutation time refers to the situation where the predicted mutation time is set. The preset value for the judgment is not limited in this embodiment. Those skilled in the art can freely choose the value based on actual needs. For example, this embodiment sets t0=2h based on the required prediction accuracy. The predicted time of sudden change refers to the length of the predicted time of sudden change judged based on the difference between the predicted time of sudden change and the preset predicted time of sudden change. The predicted time of sudden change includes both long and short periods. The power generation adjustment model is a recurrent neural network model that uses the predicted time of sudden change and the single-point power generation as input data and the single-point power generation adjustment amount as output data. This embodiment does not limit the specific construction method of the power generation adjustment model. Those skilled in the art can freely choose the method based on actual needs. For example, historical predicted times of sudden change and single-point power generation can be used as historical power adjustment data, and the historical power adjustment data and their corresponding single-point power generation adjustment amounts can be used as a training set to train the recurrent neural network model to obtain the power generation adjustment model. The single-point power generation adjustment amount refers to the quantified value of the degree to which the single-point power generation needs to be adjusted to cope with sudden weather changes, obtained from the power generation adjustment model.

[0085] Specifically, in step S2, the distribution network information model is updated to respond promptly to possible sudden weather changes in the future, ensuring the stable operation of the distribution network under adverse weather conditions, thereby improving the forward-looking nature and environmental adaptability of the distribution network.

[0086] Specifically, in step S2, when optimizing the model based on the target distribution network information using a model optimization method, the model optimization method includes:

[0087] Step B01: Input the target distribution network information into the mutation area prediction model to obtain the mutation status output by the mutation area prediction model. The mutation status includes the mutation trend and the mutation area percentage S. The mutation trend includes an upward trend and a downward trend.

[0088] Step B02: When the mutation trend is downward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model update process based on the determination result, wherein:

[0089] When S≥S0, the mutation area ratio is determined to be large, and no model optimization is performed during the model update process.

[0090] When S < S0, the mutation area ratio is determined to be small, and the model update process is optimized by replacing the mutation prediction result of "mutation exists" with "mutation does not exist".

[0091] Step B03: When the mutation trend is upward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model for the preset predicted mutation time t0 based on the determination result, wherein:

[0092] When S < S0, the mutation area ratio is determined to be small, and the model is not optimized for the preset predicted mutation time t0.

[0093] When S≥S0, the mutation area ratio is determined to be large area. The model is then optimized based on the model optimization coefficient cv at the preset predicted mutation time t0, with cv = 1.47 - 0.26 × e -（S-S0） Where e is the base of the natural logarithm, the optimized preset predicted mutation time t0' is obtained, t0' is set to t0×cv, the preset predicted mutation time t0 is replaced with the optimized preset predicted mutation time t0', and the predicted mutation time t is re-compared with the preset predicted mutation time t0.

[0094] Specifically, the mutation area prediction model refers to a recurrent neural network model that takes target distribution network information as input data and mutation state as output data. This embodiment does not limit the specific construction method of the mutation area prediction model. Those skilled in the art can freely choose according to actual needs. For example, the recurrent neural network model can be trained using historical target distribution network information and its corresponding mutation state as training set to obtain the mutation area prediction model. The mutation trend includes an upward trend and a downward trend. The upward trend refers to the trend that the weather mutation will continue to increase according to the mutation area prediction model. The downward trend refers to the trend that the weather mutation will be alleviated according to the mutation area prediction model. The preset mutation area ratio refers to a preset value for judging the state of the mutation area ratio. This embodiment does not limit the specific value setting of the preset mutation area ratio S0. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, S0 is set to 0.6 based on historical experience. The state of the mutation area ratio refers to the area size state of the mutation area ratio judged according to the mutation area ratio and the preset mutation area ratio. The state of the mutation area ratio includes large area and small area.

[0095] Specifically, in step S2, the mutation state is acquired, and the model is optimized based on the mutation state during the model update process. When the mutation trend is downward and the mutation area ratio is small, the mutation prediction result is determined to be that there is no mutation, so as to save power generation costs. When the mutation trend is upward and the mutation area ratio is large, the preset predicted mutation time t0' is increased according to the model optimization coefficient, which increases from 1.21 to 1.47 and remains constant. The constant term 1.47 of the model optimization coefficient is the maximum value that the model optimization coefficient can reach, and the constant coefficient 0.26 represents the change range of the model optimization coefficient, so as to optimize the model update process in a timely manner according to the mutation state, reduce the impact of the mutation state on the distribution network information model, and reasonably optimize the scheduling, thereby improving the efficiency of distribution network optimization scheduling.

[0096] Specifically, in step S3, when obtaining the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state, the target distribution network information and the sudden change state are input into the new energy distribution network scheduling decision tree to obtain the new energy distribution network scheduling decision output by the new energy distribution network scheduling decision tree.

[0097] Specifically, the new energy distribution network scheduling decision tree refers to a decision tree model that takes target distribution network information and sudden change states as input data and new energy distribution network scheduling decisions as output data. This embodiment does not limit the specific construction method of the new energy distribution network scheduling decision tree. Those skilled in the art can freely choose according to actual needs. For example, the new energy distribution network scheduling decision tree can be constructed by using historical target distribution network information - historical sudden change states - new energy distribution network scheduling decisions as the execution path of the decision tree. The new energy distribution network scheduling decision refers to the scheme decision for power scheduling in the new energy distribution network obtained from the new energy distribution network scheduling decision tree.

[0098] Specifically, in step S3, by acquiring the scheduling decisions of the new energy distribution network, it is possible to evaluate the scheduling decisions of the new energy distribution network and further obtain the target execution decision to save scheduling costs, thereby improving the efficiency of new energy distribution network scheduling.

[0099] Specifically, in step S3, when calculating the scheduling cost based on the target distribution network information and outputting the target execution decision based on the scheduling cost and the new energy distribution network scheduling decision, the scheduling cost Y is calculated based on the point generation cost information TP={TP1,TP2,TP3,...,TPn}, and Y=TP1+TP2+TP3+……+TPn is set to obtain the scheduling cost Y;

[0100] The scheduling cost Y is compared with the preset scheduling cost Y0. Based on the comparison result, the status of the scheduling cost is determined, and based on the determination result, a target execution decision is output, wherein:

[0101] When Y < Y0, the scheduling cost is determined to be low, and the new energy distribution network scheduling decision is output as the target execution decision.

[0102] When Y≥Y0, the scheduling cost is determined to be high. The scheduling decision of the new energy distribution network is optimized by manual adjustment to obtain the optimized new energy distribution network scheduling decision, and the optimized new energy distribution network scheduling decision is output as the target execution decision.

[0103] Specifically, the power generation cost weight set refers to a set of coefficients that measure the importance of the dispatch cost of a single power generation point in the overall dispatch cost. Here, w1 is the weight coefficient of the dispatch cost of the first power generation point in the weight set, w2 is the weight coefficient of the dispatch cost of the second power generation point, w3 is the weight coefficient of the dispatch cost of the third power generation point, and wn is the weight coefficient of the dispatch cost of the nth power generation point. This embodiment does not limit the specific value of the power generation cost weight set; those skilled in the art can freely choose according to actual needs, as long as the requirement of w1+w2+w3+……+wn=1 is met. For example, when the maximum value of n is... At time 4, w1=0.3, w2=0.3, w3=0.2, w4=0.2. The preset scheduling cost refers to the preset value for judging the state of the scheduling cost. This embodiment does not limit the specific value of the preset scheduling cost Y0. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, Y0 is set to 103,000 yuan / day. The state of the scheduling cost refers to the high or low state of the scheduling cost judged based on the scheduling cost and the preset scheduling cost. The state of the scheduling cost includes low cost and high cost. The manual adjustment refers to the process by which relevant technicians reduce the cost of scheduling decisions for the new energy distribution network, such as installing a static var generator at the end of a heavily loaded line. The static var generator refers to a power electronic device that reduces reactive power transmission losses.

[0104] Specifically, in step S3, the status of scheduling costs is judged and the target execution decision is output so as to control the scheduling costs of the new energy distribution network scheduling decision, thereby reducing the power generation cost.

[0105] Specifically, in step S4, when adjusting the scheduling cost calculation based on the number of charge / discharge cycles, the first deployment point's charge / discharge cycle np1 is compared with the preset charge / discharge cycle np0. Based on the comparison result, the status of the first deployment point's charge / discharge cycle is determined, and the scheduling cost calculation is adjusted based on the determination result. Wherein:

[0106] When np1≤np0, the state of the first deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0107] When np1 > np0, the state of the first deployment point's charging and discharging count is determined to be abnormal, and cost correction is performed on the scheduling cost calculation process: the generation scheduling cost TP1 of the first deployment point is corrected according to the cost correction coefficient cv, and cv = 1.33 - 0.22 × e -（np1-np0）The generation scheduling cost TP1' of the first deployment point after correction is obtained. TP1' = TP1 × cv is set. The generation scheduling cost TP1 of the first deployment point is replaced with the generation scheduling cost TP1' of the first deployment point after correction. The scheduling cost Y is recalculated based on the generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point.

[0108] The second charging / discharging count np2 is compared with the preset charging / discharging count np0. Based on the comparison result, the status of the second charging / discharging count is determined, and the scheduling cost calculation process is adjusted according to the determination result.

[0109] When np2≤np0, the state of the second deployment point charging and discharging times is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0110] When np2 > np0, the state of the second deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP2 of the second deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP2' of the second deployment point. TP2' is set to TP2 × cv. The power generation scheduling cost TP2 of the second deployment point is replaced with the corrected power generation scheduling cost TP2' of the second deployment point. The scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}.

[0111] The third deployment point's charge / discharge count np3 is compared with the preset charge / discharge count np0. Based on the comparison result, the status of the third deployment point's charge / discharge count is determined, and the scheduling cost calculation process is adjusted according to the determination result.

[0112] When np3≤np0, the state of the third deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process;

[0113] When np3 > np0, the state of the third deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP3 of the third deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP3' of the third deployment point. TP3' is set to TP3 × cv. The power generation scheduling cost TP3 of the third deployment point is replaced with the corrected power generation scheduling cost TP3' of the third deployment point, and the scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}.

[0114] ...

[0115] The number of charge / discharge cycles at the nth location (npn) is compared with the preset number of charge / discharge cycles (np0). Based on the comparison result, the status of the number of charge / discharge cycles at the nth location is determined, and the cost calculation process for scheduling costs is adjusted based on the determination result.

[0116] When npn≤np0, the state of the nth point charging and discharging count is determined to be normal, and no cost correction is performed in the calculation process of scheduling cost;

[0117] When npn > np0, the state of the charging and discharging count of the nth deployment point is determined to be an abnormal state, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TPn of the nth deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TPn' of the nth deployment point. TPn' is set to TPn × cv. The power generation scheduling cost TPn of the nth deployment point is replaced with the corrected power generation scheduling cost TPn' of the nth deployment point, and the scheduling cost Y is recalculated according to the power generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point.

[0118] Specifically, the preset number of charge-discharge cycles refers to a preset value used to judge the state of the first charge-discharge cycle, the second charge-discharge cycle, the third charge-discharge cycle, ..., the nth charge-discharge cycle. This embodiment does not limit the specific value of the preset number of charge-discharge cycles np0. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, np0 is set to 50 times. The state of the first charge-discharge cycle refers to the normality of the first charge-discharge cycle judged based on the first charge-discharge cycle and the preset number of charge-discharge cycles. The state of the first charge-discharge cycle includes normal and abnormal states. The state of the second charge-discharge cycle refers to the normality of the second charge-discharge cycle judged based on the second charge-discharge cycle and the preset number of charge-discharge cycles. The state of the second charge-discharge cycle includes normal and abnormal states. The state of the third charge-discharge cycle refers to the normality of the third charge-discharge cycle judged based on the third charge-discharge cycle and the preset number of charge-discharge cycles. The state of the third charge-discharge cycle includes normal and abnormal states.

[0119] Specifically, in step S4, the state of the charging and discharging counts of a single deployment point is judged so that when the state of the charging and discharging counts of a deployment point is abnormal, the power generation dispatch cost of that deployment point is increased according to the cost correction coefficient which increases from 1.11 to 1.33. Here, the constant term of the cost correction coefficient, 1.33, is the maximum value that the cost correction coefficient can reach, and the constant coefficient, 0.22, is the change range of the cost correction coefficient. This allows the dispatch cost to be refined on a per-deployment-point basis, reducing the impact of abnormal charging and discharging counts on the dispatch cost of a single deployment point, thereby further improving the control of dispatch costs.

[0120] Specifically, in step S4, when performing cost calibration for the cost correction process based on the mutation state, if the mutation state is on an upward trend and S≥S0, cost calibration is performed on the preset charge-discharge number np0 based on the calibration coefficient ck, where ck = 0.64 + 0.23 × e -（S-S0） Where e is the base of the natural logarithm, the preset charge-discharge count np0' after calibration is obtained, np0' = np0 × ck is set, and the preset charge-discharge count np0' after calibration is rounded to obtain the preset charge-discharge count np00 after calibration and rounding. The preset charge-discharge count np0 is replaced with the preset charge-discharge count np00 after calibration and rounding. The charge-discharge counts np1, np2, np3, ..., npn of the first point are compared with the preset charge-discharge count np0 again.

[0121] Specifically, the rounding process refers to the process of rounding the preset number of charge and discharge cycles after calibration to the nearest integer. This embodiment does not limit the specific method of rounding. Those skilled in the art can freely choose according to actual needs, such as rounding according to the rounding principle.

[0122] Specifically, in step S4, the preset number of charge-discharge cycles is reduced by a calibration coefficient that decreases from 0.87 to 0.64 while remaining constant when the mutation state is on the rise and S≥S0. The constant term 0.64 of the calibration coefficient is the minimum value that the calibration coefficient can reach, and the constant coefficient 0.23 represents the change range of the calibration coefficient. This is to reduce the impact on the cost correction process when the weather mutation trend is aggravated and the mutation area is large, thereby improving the efficiency of scheduling optimization.

[0123] Specifically, in step S4, when updating the distribution network information model a second time based on the meteorological impact value at the designated locations, the distribution information and cloud thickness are input into the meteorological impact model at the designated locations to obtain the meteorological impact value BD output by the model. The meteorological impact value BD is then compared with a preset meteorological impact value BD0. Based on the comparison result, the state of the meteorological impact value at the designated locations is determined, and the distribution network information model is updated a second time based on the determination result. Wherein:

[0124] When BD≤BD0, the status of the meteorological impact value at the distribution point is determined to be acceptable, and no secondary update is performed on the distribution network information model;

[0125] When BD > BD0, the meteorological impact value of the distribution point is determined to be unacceptable, and the distribution network information model is updated a second time: the boundary of the distribution point coordinates is expanded to obtain the updated distribution point coordinates, the distribution point coordinates are replaced with the updated distribution point coordinates, and the distribution network information visualization model is reconstructed based on the distribution point coordinates.

[0126] Specifically, the meteorological impact model refers to a recurrent neural network model that uses power distribution information and cloud thickness as input data and meteorological impact values ​​at designated locations as output data. This embodiment does not limit the specific construction method of the meteorological impact values ​​at designated locations; those skilled in the art can freely choose according to actual needs. For example, historical power distribution information and cloud thickness can be used as historical meteorological data at designated locations, and the historical meteorological data at designated locations and their corresponding meteorological impact values ​​can be used as a training set to train the recurrent neural network model to obtain the meteorological impact model at designated locations. The meteorological impact value at designated locations refers to the value obtained from the meteorological impact model that reflects the degree of influence of weather changes on all individual locations. The preset meteorological impact value at designated locations... The meteorological impact value refers to a preset value used to judge the state of the meteorological impact value of the sampling point. This embodiment does not limit the specific value setting of the preset meteorological impact value BD0. Those skilled in the art can freely choose according to actual needs. For example, this embodiment sets BD0=0.6 based on historical experience. The state of the meteorological impact value of the sampling point refers to the degree of acceptability of the meteorological impact value of the sampling point judged based on the meteorological impact value of the sampling point and the preset meteorological impact value of the sampling point. The state of the meteorological impact value of the sampling point includes acceptable and unacceptable. The boundary expansion process refers to the process of expanding the boundary of the sampling point coordinates. For example, the sampling point coordinates are set as the center of a circle, and a circular range is drawn according to the radius d. The circular range is used as the updated sampling point coordinates.

[0127] Specifically, in step S4, the status of the meteorological impact value of the sampling points is judged. When the status of the meteorological impact value of the sampling points is unacceptable, the boundary of the sampling point coordinates is expanded to increase the influence of the sampling point coordinates on the prediction results of sudden changes and the predicted time of sudden changes, and the prediction accuracy is calibrated in a timely manner, thereby improving the efficiency of scheduling optimization.

[0128] Specifically, in step S4, the meteorological optimization of the secondary update process is performed based on the proportion of mutation duration. The proportion of mutation duration tp is calculated based on the short duration td and the total monitoring time tz, and tp = td / tz is set to obtain the mutation duration proportion tp. This mutation duration proportion tp is compared with a preset mutation duration proportion tp0. The state of the mutation duration proportion is judged based on the comparison result, and the preset meteorological impact value BD0 is optimized based on the judgment result. Wherein:

[0129] When tp≤tp0, the percentage of sudden change duration is determined to be the normal percentage, and no meteorological optimization is performed on the preset meteorological impact value BD0.

[0130] When tp > tp0, the percentage of abrupt change duration is determined to be an abnormal percentage. Based on the meteorological optimization coefficient cq, the preset meteorological impact value BD0 is optimized, with cq set to 0.63 + 0.25 × e. -（tp-tp0） Where e is the base of the natural logarithm, the optimized preset meteorological impact value BD0' is obtained, BD0' is set to BD0×cq, the preset meteorological impact value BD0 is replaced with the optimized preset meteorological impact value BD0', and the meteorological impact value BD0 is re-compared with the preset meteorological impact value BD0.

[0131] Specifically, the short duration refers to the total duration of the predicted mutation time when the mutation prediction result indicates the existence of a mutation. This embodiment does not limit the specific method of obtaining the short duration; those skilled in the art can freely choose according to actual needs, such as obtaining the short duration through system logs. The total monitoring time refers to the preset time length for monitoring the mutation prediction result, calculated based on the mutation duration percentage. The preset mutation duration percentage refers to a preset value for judging the state of the mutation duration percentage. This embodiment does not limit the specific values ​​of the total monitoring time tz and the preset mutation duration percentage tp0; those skilled in the art can freely choose according to actual needs, such as setting tz=24h and tp0=0.4 in this embodiment. The state of the mutation duration percentage refers to the normality of the mutation duration percentage judged based on the mutation duration percentage and the preset mutation duration percentage. The state of the mutation duration percentage includes normal percentage and abnormal percentage.

[0132] Specifically, in step S4, by judging the state of the percentage of mutation duration, when the state of the percentage of mutation duration is an abnormal percentage, the meteorological impact value of the preset distribution points is reduced according to the meteorological optimization coefficient which decreases from 0.88 to 0.63 and remains constant. The constant term 0.63 of the meteorological optimization coefficient is the minimum value that the meteorological optimization coefficient can reach, and the constant coefficient 0.25 is the change range of the meteorological optimization coefficient, so as to reasonably reduce the impact of the abnormal percentage of mutation duration on the secondary update process, thereby improving the efficiency of distribution network dispatch optimization.

[0133] Please see Figure 4 As shown, this is a schematic diagram of the system structure of the optimized scheduling method for distributed new energy distribution networks in this embodiment. The system includes:

[0134] The power grid information acquisition module is used to collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model.

[0135] The weather change prediction module is used to acquire change prediction results based on the distribution network information model, update the distribution network information model based on the change prediction results, acquire change status based on the target distribution network information, and optimize the model during the model update process based on the change status. The weather change prediction module is connected to the power grid information acquisition module.

[0136] The new energy deployment monitoring module is used to acquire new energy distribution network scheduling decisions based on target distribution network information and sudden change status, calculate scheduling costs based on target distribution network information, and output target execution decisions based on scheduling costs and new energy distribution network scheduling decisions. The new energy deployment monitoring module is connected to the weather sudden change prediction module.

[0137] The scheduling feedback optimization module is used to correct the scheduling cost calculation process based on the number of charge and discharge cycles, and to calibrate the cost correction process based on the sudden change state. It is also used to perform a secondary update of the distribution network information model based on the meteorological impact value of the site, and to optimize the meteorological process of the secondary update based on the proportion of the sudden change duration. The scheduling feedback optimization module is connected to the new energy site monitoring module.

[0138] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An optimized scheduling method for a distributed renewable energy distribution network, characterized in that, The method includes: Step S1: Collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model; Step S2 involves obtaining the mutation prediction results based on the distribution network information model, updating the distribution network information model based on the mutation prediction results, and optimizing the model update process based on the target distribution network information. Step S3: Obtain the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state; calculate the scheduling cost based on the target distribution network information; and output the target execution decision based on the scheduling cost and the new energy distribution network scheduling decision. Step S4: The cost calculation process of scheduling cost is corrected based on the number of charging and discharging cycles, and the cost correction process is calibrated based on the sudden change state. The distribution network information model is updated a second time based on the meteorological impact value of the distribution points, and the meteorological optimization is performed on the second update process based on the proportion of sudden change duration. In step S4, when updating the distribution network information model a second time based on the meteorological impact value at the designated location, the distribution information and cloud thickness are input into the meteorological impact model at the designated location to obtain the meteorological impact value BD output by the model. The meteorological impact value BD is then compared with the preset meteorological impact value BD0. Based on the comparison result, the state of the meteorological impact value at the designated location is determined, and the distribution network information model is updated a second time based on the determination result. Wherein: When BD≤BD0, the status of the meteorological impact value at the distribution point is determined to be acceptable, and no secondary update is performed on the distribution network information model; When BD > BD0, the meteorological impact value of the distribution point is determined to be unacceptable, and the distribution network information model is updated a second time: the boundary of the distribution point coordinates is expanded to obtain the updated distribution point coordinates, the distribution point coordinates are replaced with the updated distribution point coordinates, and the distribution network information visualization model is reconstructed based on the distribution point coordinates. In step S4, the meteorological optimization of the secondary update process is performed based on the proportion of abrupt changes in duration. The proportion of abrupt changes in duration tp is calculated based on the short duration td and the total monitoring time tz, with tp = td / tz. This proportion is then compared with a preset proportion of abrupt changes in duration tp0. The state of the proportion of abrupt changes in duration is judged based on the comparison result, and the preset meteorological impact value BD0 is optimized based on the judgment result. When tp≤tp0, the percentage of sudden change duration is determined to be the normal percentage, and no meteorological optimization is performed on the preset meteorological impact value BD0. When tp > tp0, the percentage of abrupt change duration is determined to be an abnormal percentage. Based on the meteorological optimization coefficient cq, the preset meteorological impact value BD0 is optimized, with cq set to 0.63 + 0.25 × e. -（tp-tp0） Where e is the base of the natural logarithm, the optimized preset meteorological impact value BD0' is obtained, BD0' is set to BD0×cq, the preset meteorological impact value BD0 is replaced with the optimized preset meteorological impact value BD0', and the meteorological impact value BD0 is re-compared with the preset meteorological impact value BD0.

2. The optimized scheduling method for distributed new energy distribution networks according to claim 1, characterized in that, In step S1, when constructing the distribution network information model based on the target distribution network information using the distribution network information model construction method, the distribution network information model construction method includes: Step A01: Construct the distribution network information visualization model based on the layout coordinates to obtain the distribution network information visualization model; Step A02: Based on the target distribution network information, perform data embedding processing on the distribution network information visualization model to obtain the distribution network information model.

3. The optimized scheduling method for distributed new energy distribution networks according to claim 2, characterized in that, In step S2, when the mutation prediction result is obtained based on the target distribution network information and the distribution network information model is updated based on the mutation prediction result, the target distribution network information and cloud thickness are input into the power generation mutation prediction model to obtain the mutation prediction result and the predicted mutation time t output by the power generation mutation prediction model. The mutation prediction result includes whether a mutation exists or not. When the mutation prediction result is that there is no mutation, the distribution network information model will not be updated; When the mutation prediction result indicates the existence of a mutation, the predicted mutation occurrence time t is compared with the preset predicted mutation occurrence time t0. Based on the comparison result, a judgment is made regarding the predicted mutation occurrence time, and the distribution network information model is updated according to the judgment result. Wherein: When t≥t0, the predicted time of sudden change is determined to be long, and the distribution network information model is not updated. When t < t0, the predicted time of sudden change is determined to be short, and the distribution network information model is updated: the predicted time of sudden change and the single-point power generation FD are input into the power generation adjustment model to obtain the single-point power generation adjustment amount m output by the power generation adjustment model. The updated single-point power generation Fm is calculated based on the single-point power generation FD and the single-point power generation adjustment amount m, and Fm is set to FD + m to obtain the updated single-point power generation Fm. The single-point power generation FD is replaced with the updated single-point power generation Fm to obtain the updated target distribution network information. The data embedding process of the distribution network information visualization model is then re-performed based on the updated target distribution network information.

4. The optimized scheduling method for distributed new energy distribution networks according to claim 3, characterized in that, In step S2, when the model is optimized based on the target distribution network information during the model update process using a model optimization method, the model optimization method includes: Step B01: Input the target distribution network information into the mutation area prediction model to obtain the mutation status output by the mutation area prediction model. The mutation status includes the mutation trend and the mutation area percentage S. The mutation trend includes an upward trend and a downward trend. Step B02: When the mutation trend is downward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model update process based on the determination result, wherein: When S≥S0, the mutation area ratio is determined to be large, and no model optimization is performed during the model update process. When S < S0, the mutation area ratio is determined to be small, and the model update process is optimized by replacing the mutation prediction result of "mutation exists" with "mutation does not exist". Step B03: When the mutation trend is upward, compare the mutation area percentage S with the preset mutation area percentage S0, determine the state of the mutation area percentage based on the comparison result, and optimize the model for the preset predicted mutation time t0 based on the determination result, wherein: When S < S0, the mutation area ratio is determined to be small, and the model is not optimized for the preset predicted mutation time t0. When S≥S0, the mutation area ratio is determined to be large area. The model is then optimized based on the model optimization coefficient cv at the preset predicted mutation time t0, with cv = 1.47 - 0.26 × e -（S-S0） Where e is the base of the natural logarithm, the optimized preset predicted mutation time t0' is obtained, t0' is set to t0×cv, the preset predicted mutation time t0 is replaced with the optimized preset predicted mutation time t0', and the predicted mutation time t is re-compared with the preset predicted mutation time t0.

5. The optimized scheduling method for distributed new energy distribution networks according to claim 4, characterized in that, In step S3, when obtaining the new energy distribution network scheduling decision based on the target distribution network information and the sudden change state, the target distribution network information and the sudden change state are input into the new energy distribution network scheduling decision tree to obtain the new energy distribution network scheduling decision output by the new energy distribution network scheduling decision tree. In step S3, when calculating the dispatch cost based on the target distribution network information and outputting the target execution decision based on the dispatch cost and the new energy distribution network dispatch decision, the dispatch cost Y is calculated based on the point generation cost information TP={TP1,TP2,TP3,...,TPn}, and Y=TP1+TP2+TP3+……+TPn is set to obtain the dispatch cost Y; The scheduling cost Y is compared with the preset scheduling cost Y0. Based on the comparison result, the status of the scheduling cost is determined, and based on the determination result, a target execution decision is output, wherein: When Y < Y0, the scheduling cost is determined to be low, and the new energy distribution network scheduling decision is output as the target execution decision. When Y≥Y0, the scheduling cost is determined to be high. The scheduling decision of the new energy distribution network is optimized by manual adjustment to obtain the optimized new energy distribution network scheduling decision, and the optimized new energy distribution network scheduling decision is output as the target execution decision.

6. The optimized scheduling method for distributed new energy distribution networks according to claim 5, characterized in that, In step S4, when adjusting the scheduling cost calculation based on the number of charge / discharge cycles, the first deployment point's charge / discharge cycle np1 is compared with the preset charge / discharge cycle np0. Based on the comparison result, the status of the first deployment point's charge / discharge cycle is determined, and the scheduling cost calculation is adjusted based on the determination result. Specifically: When np1≤np0, the state of the first deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process; When np1 > np0, the state of the first deployment point's charging and discharging count is determined to be abnormal, and cost correction is performed on the scheduling cost calculation process: the generation scheduling cost TP1 of the first deployment point is corrected according to the cost correction coefficient cv, and cv = 1.33 - 0.22 × e -（np1-np0） Where e is the base of the natural logarithm, the power generation scheduling cost TP1' of the first deployment point after correction is obtained, TP1' = TP1 × cv is set, the power generation scheduling cost TP1 of the first deployment point is replaced with the power generation scheduling cost TP1' of the first deployment point after correction, and the scheduling cost Y is recalculated according to the power generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point; The second charging / discharging count np2 is compared with the preset charging / discharging count np0. Based on the comparison result, the status of the second charging / discharging count is determined, and the scheduling cost calculation process is adjusted according to the determination result. When np2≤np0, the state of the second deployment point charging and discharging times is determined to be normal, and no cost correction is performed in the scheduling cost calculation process; When np2 > np0, the state of the second deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP2 of the second deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP2' of the second deployment point. TP2' is set to TP2 × cv. The power generation scheduling cost TP2 of the second deployment point is replaced with the corrected power generation scheduling cost TP2' of the second deployment point. The scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}. The third deployment point's charge / discharge count np3 is compared with the preset charge / discharge count np0. Based on the comparison result, the status of the third deployment point's charge / discharge count is determined, and the scheduling cost calculation process is adjusted according to the determination result. When np3≤np0, the state of the third deployment point's charging and discharging count is determined to be normal, and no cost correction is performed in the scheduling cost calculation process; When np3 > np0, the state of the third deployment point's charging and discharging count is determined to be abnormal, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TP3 of the third deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TP3' of the third deployment point. TP3' is set to TP3 × cv. The power generation scheduling cost TP3 of the third deployment point is replaced with the corrected power generation scheduling cost TP3' of the third deployment point, and the scheduling cost Y is recalculated according to the deployment point power generation cost information TP = {TP1, TP2, TP3, ..., TPn}. …… The number of charge / discharge cycles at the nth location (npn) is compared with the preset number of charge / discharge cycles (np0). Based on the comparison result, the status of the number of charge / discharge cycles at the nth location is determined, and the cost calculation process for scheduling costs is adjusted based on the determination result. When npn≤np0, the state of the nth point charging and discharging count is determined to be normal, and no cost correction is performed in the calculation process of scheduling cost; When npn > np0, the state of the charging and discharging count of the nth deployment point is determined to be an abnormal state, and the cost calculation process of the scheduling cost is corrected: the power generation scheduling cost TPn of the nth deployment point is corrected according to the cost correction coefficient cv to obtain the corrected power generation scheduling cost TPn' of the nth deployment point. TPn' is set to TPn × cv. The power generation scheduling cost TPn of the nth deployment point is replaced with the corrected power generation scheduling cost TPn' of the nth deployment point, and the scheduling cost Y is recalculated according to the power generation cost information TP = {TP1, TP2, TP3, ..., TPn} of the deployment point.

7. The optimized scheduling method for distributed new energy distribution networks according to claim 6, characterized in that, In step S4, when the cost correction process is performed based on the mutation state, if the mutation state is on an upward trend and S≥S0, the cost is calibrated based on the preset charge-discharge number np0 according to the calibration coefficient ck, and ck=0.64+0.23×e -（S-S0） Where e is the base of the natural logarithm, the preset charge-discharge count np0' after calibration is obtained, np0' = np0 × ck is set, and the preset charge-discharge count np0' after calibration is rounded to obtain the preset charge-discharge count np00 after calibration and rounding. The preset charge-discharge count np0 is replaced with the preset charge-discharge count np00 after calibration and rounding. The charge-discharge counts np1, np2, np3, ..., npn of the first point are compared with the preset charge-discharge count np0 again.

8. A system for the optimized scheduling method of a distributed renewable energy distribution network as described in any one of claims 1-7, the system comprising: The power grid information acquisition module is used to collect target distribution network information and construct a distribution network information model based on the target distribution network information to obtain the distribution network information model. The weather change prediction module acquires change prediction results based on the distribution network information model, updates the distribution network information model based on the change prediction results, acquires the change status based on the target distribution network information, and optimizes the model based on the change status during the model update process. The new energy deployment monitoring module acquires new energy distribution network scheduling decisions based on target distribution network information and sudden change status, calculates scheduling costs based on target distribution network information, and outputs target execution decisions based on scheduling costs and new energy distribution network scheduling decisions. The scheduling feedback optimization module corrects the scheduling cost calculation process based on the number of charging and discharging cycles, calibrates the cost correction process based on sudden change states, updates the distribution network information model a second time based on the meteorological impact value of the distribution points, and optimizes the second update process based on the proportion of sudden change duration.

Images