A regional power grid prediction scheduling method and system based on topological balance

Abstract

The application discloses a kind of based on topological balance regional power grid prediction scheduling method and system, it is related to power grid prediction scheduling technical field, including: the monitoring model reflecting whole network real-time power flow is established;Based on the monitoring model, the new energy output scene library is constructed in combination with probability statistics method;Based on the scene library, the critical operating state of power grid under each probability scene is obtained, and the mapping relationship between transmission capacity and scene is formed;According to real-time operating state matching mapping relationship, the adaptive transmission limit is obtained by correction and joint optimization;The application solves the problem that traditional static transmission limit cannot accurately reflect high proportion new energy fluctuation and power grid operating state change by dynamic scheduling method based on real-time monitoring, scenario analysis and joint optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid prediction scheduling, and more particularly to a regional power grid prediction scheduling method and system based on topology balance. BACKGROUND

[0002] With the continuous increase of new energy generation proportion, the large-scale access of intermittent power sources such as wind power and photovoltaic in regional power grids makes power grid operation face more and more uncertainties and dynamic challenges. Traditional power grid scheduling and prediction methods mostly rely on historical load prediction and statistical rules of unit output, and perform operation optimization through static power flow calculation and experience scheduling rules. However, these methods have several technical bottlenecks.

[0003] The traditional scheduling method mostly uses static or empirical safety margin when calculating the transmission channel capacity, ignores the critical state and instantaneous bearable capacity difference under different operation scenarios, and is difficult to provide accurate constraints for new energy output adjustment. Moreover, high proportion of new energy output fluctuation is large and has strong uncertainty, and the characteristics of power flow distribution and power transmission are increasingly complex and changeable, the maximum transmission capacity is closely related to the operation mode of the power grid and the adjustment mode of the new energy output, and the off-line limit formulated based on the long-period time scale is difficult to adapt to the dramatic changes in the operation mode of the power grid. In view of the above problems, the present application provides a solution. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a regional power grid prediction scheduling method and system based on topology balance, which solves the problem that the traditional static transmission limit cannot accurately reflect the high proportion of new energy fluctuation and the change of power grid operation state through a dynamic scheduling method based on real-time monitoring, scenario analysis and joint optimization.

[0005] To achieve the above object, the present application provides the following technical scheme:

[0006] In a first aspect, the present application provides a regional power grid prediction scheduling method based on topology balance, which comprises: establishing a monitoring model reflecting real-time power flow of the whole network; based on the monitoring model, combining a probability statistical method to construct a new energy output scenario library; based on the scenario library, obtaining the critical operation state of the power grid under each probability scenario to form a mapping relationship between transmission capacity and scenario, the mapping relationship is obtained by weighting and averaging the transmission capacity matrix through the correction probability weight, and a scenario-capacity mapping table is generated; according to the real-time operation state, the mapping relationship is matched, corrected and jointly optimized to obtain adaptive transmission limit, wherein the correction is corrected by calculating the variance of the confidence interval, and the optimization is optimized by constructing a multi-objective joint optimization function and constraint condition in combination with the confidence interval correction.

[0007] In one of the embodiments, a monitoring model reflecting real-time power flow of the whole network is established, specifically: the real-time power grid data is preprocessed to form a standardized input data set; a node-branch network model is constructed based on the power grid topology and dynamic assignment is performed to form a power flow monitoring framework; the standardized input data set is subjected to parameter matching and parameter mapping, and is input into the power flow monitoring framework, and an improved Newton-Raphson iteration method is used for power flow calculation; and a real-time power flow monitoring model of the whole network containing node voltage, branch power flow and system power state is generated according to the calculation result.

[0008] In one of the embodiments, the standardized input data is subjected to parameter matching and parameter mapping, specifically: the standardized input data is initialized as a particle set, and an initial weight is assigned; the prediction state of each particle is calculated based on the power grid topology; real-time measurement data is obtained and compared with the particle prediction state, the updated weight of each particle is calculated, and the particle set is resampled; the resampled particle set is obtained by weighted average to obtain the optimal particle state, and is mapped to the node-branch network model.

[0009] In one of the embodiments, an improved Newton-Raphson iteration method is used for power flow calculation, specifically: an initial power flow state vector is constructed according to the matched standardized input data set; a nonlinear equation of the whole network power flow is established based on the power grid topology structure and the whole network power flow monitoring framework; the Jacobian matrix is constructed by sparse matrix and the initial power flow state vector is iteratively solved, and the power flow calculation result containing node voltage, line power flow and system power is output.

[0010] In one of the embodiments, based on the monitoring model, a new energy output scene is constructed by combining a probability statistical method, specifically: a first data set is extracted based on the real-time monitoring model, and the first data set is subjected to probability distribution fitting; an initial random scene set is generated from the probability distribution by using a random sampling method; the initial random scene set is subjected to dimension reduction and clustering processing, representative scenes are screened, and an initial probability weight is assigned to each scene; multi-scene power flow calculation and verification are performed based on the representative scenes to generate a new energy output scene library containing node voltage, branch power flow and system power balance state.

[0011] In one of the embodiments, the critical operating state of the power grid under each probability scene is obtained based on the scene library, specifically: time series power flow calculation is performed on each representative scene in the new energy output scene library to obtain dynamic operating data; dynamic margin indicators of each scene are calculated based on the dynamic operating data; and the critical operating state under each scene is identified according to the dynamic margin indicators.

[0012] In one of the embodiments, the mapping relationship is obtained by weighted average processing of the power transmission capability matrix by correcting the probability weight, and specifically: the limit power transmission capability of the power transmission channel under each scene is analyzed to form a scene power transmission capability matrix; a conditional probability correction model is constructed based on the initial probability weight of each scene, and the corrected probability weight is output through the probability correction model; the corrected probability weight is combined with the power transmission capability matrix corresponding to each scene, and a weighted average method is used to obtain a corrected weighted power transmission capability matrix; and a scene-capability mapping table is generated based on the corrected weighted power transmission capability matrix.

[0013] In one of the embodiments, the adaptive power transmission limit is obtained by matching the mapping relationship according to the real-time operation state, correcting and jointly optimizing, wherein the correction step is: a high-dimensional state vector is constructed based on the real-time operation state, and a state vector corresponding to a historical scene is extracted; the current high-dimensional state vector and the corresponding historical state vector are preliminarily screened, and based on the Mahalanobis distance calculation method, a number of scenes with the smallest distance are selected as a subset; the power transmission capability data in the subset is predicted by using a multivariate linear regression model to obtain a preliminary power transmission capability; the preliminary power transmission capability is input into a pre-trained random forest regression model, and the final power transmission capability prediction value is obtained by multi-decision tree integration prediction; the variance is calculated based on the final power transmission capability prediction value, the confidence interval is corrected based on the variance, and the corrected actual available power transmission capability is output.

[0014] In one of the embodiments, the adaptive power transmission limit is obtained by matching the mapping relationship according to the real-time operation state, correcting and jointly optimizing, and the optimization step is: a multi-objective joint optimization function is constructed, which includes a system total power loss, a renewable energy output consumption amount, and a dynamic margin index of power grid safety, and a constraint condition is set, and the confidence interval of the corrected actual available power transmission capability is used to constrain the instantaneous maximum power transmission capability of each power transmission channel; a rolling time domain optimization algorithm is used to solve the optimization function to obtain an optimal adjustment scheme of the new energy output strategy and the power grid operation mode; the adaptive limit of each power transmission channel is calculated and issued to the dispatching system, and a closed-loop updating mechanism is established to realize dynamic adjustment of the power transmission capability.

[0015] In a second aspect, the application provides a regional power grid predictive scheduling system based on topological balance, which comprises a model construction module for establishing a monitoring model reflecting real-time power flow of the whole grid; a scenario library construction module for constructing a new energy output scenario library based on the monitoring model and combining a probability and statistics method; a mapping table generation module for obtaining critical operating states of the power grid under each probability scenario based on the scenario library, forming a mapping relationship between power transmission capacity and scenarios, and generating a scenario-capacity mapping table by performing weighted average processing on the power transmission capacity matrix by correcting the probability weight; and a power transmission limit generation module for matching the mapping relationship according to a real-time operating state, correcting and jointly optimizing to obtain adaptive power transmission limits, wherein the correction is performed by correcting a confidence interval by calculating a variance, and the optimization is performed by constructing a multi-objective joint optimization function and constraint conditions in combination with the confidence interval correction.

[0016] The application can reflect the grid state in real time under the condition of a high proportion of new energy, scientifically describe the fluctuation characteristics of new energy, reduce the calculation complexity, and enhance the reliability and flexibility of scheduling decisions, thereby providing a solid data foundation and technical support for dynamic power transmission capacity calculation, scheduling optimization and risk assessment. The overall scheme not only improves the safety, stability and flexibility of the power grid under the condition of a high proportion of new energy, but also enhances the new energy consumption capacity and significantly optimizes the real-time and accuracy of scheduling decisions, thereby providing scientific and operable technical support for modern power grid operation. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of a regional power grid predictive scheduling method based on topological balance is provided for the embodiments of the application.

[0018] Figure 2 A structural diagram of a regional power grid predictive scheduling system based on topological balance is provided for the embodiments of the application.

[0019] Figure 3 A mapping diagram of instantaneous maximum power transmission capacity of a power transmission channel under new energy output scenarios is provided for the embodiments of the application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0021] Referring to Figure 1 The application provides a flowchart of a regional power grid predictive scheduling method based on topological balance, which comprises the following steps:

[0022] S1, a monitoring model reflecting real-time power flow of the whole network is established, specifically including obtaining real-time power grid data, and based on power grid topology structure and operation mode, a real-time monitoring model of power flow distribution state of the whole network is established, the real-time power grid data including new energy output data, conventional unit operation data and key operation parameters.

[0023] Among them, through the power dispatch automation system and the new energy station monitoring system, the new energy output data of the new energy unit is collected in real time, the new energy output data including active power output, reactive power output and available installed capacity; the conventional thermal power, hydropower and nuclear power unit operation data are obtained by using the state monitoring device and the energy management system, the conventional unit operation data including real-time output level, standby capacity and start-stop state; through the power grid SCADA system (supervisory control and data acquisition system), the key operation parameters including voltage, current, frequency and power flow direction are synchronously obtained, so as to ensure the consistency and comparability of the new energy output data and the conventional unit operation data; the new energy output data, the conventional unit operation data and the key operation parameters are time-synchronized and data-cleaned to form a standardized real-time operation database for subsequent establishment of a dynamic monitoring model of the whole network power flow distribution state.

[0024] In this embodiment, real-time power grid data is obtained, and based on power grid topology structure and operation mode, a real-time monitoring model of power flow distribution state of the whole network is established, the real-time power grid data including new energy output data, conventional unit operation data and key operation parameters, specifically:

[0025] The real-time power grid data is preprocessed, the preprocessing including time alignment and data cleaning, eliminating outliers and completing missing values to form a standardized input data set;

[0026] Based on the power grid topology information, the buses, substations and important load points in the power grid are modeled as nodes in the graph, and the transmission lines, transformers and controllable branches are modeled as branches in the graph, wherein each node and branch contains electrical parameter information, the electrical parameter information including impedance, rated capacity, rated voltage and power flow control device parameters, forming a complete node-branch network model;

[0027] According to the actual operation mode of the power grid, the node-branch network model is dynamically assigned node and branch parameters to construct a whole network power flow monitoring framework that can reflect the current operation state of the power grid;

[0028] Exemplarily, the dynamic assignment can be understood as follows: according to the current actual operation of the power grid, the parameters of the nodes and branches are updated to the current values in real time or according to time steps, instead of using fixed and pre-set parameters, wherein the node and branch parameters include the unit start state and output power, the load distribution and power factor, the power flow control strategy parameters such as reactive power regulation, phase angle control, power shunt device setting, and the purpose of this is to make the power flow monitoring framework truly reflect the current state of the power grid, and the power flow calculation result can be updated in real time along with the unit start-stop, load change or control strategy adjustment, instead of static calculation.

[0029] Parameter matching and parameter mapping are performed on the standardized input data set, so that the standardized input data set corresponds to the corresponding nodes and branches in the node-branch network model;

[0030] The matched standardized input data set is input into the full-network power flow monitoring framework, and the improved Newton-Raphson iteration method is used to perform power flow calculation, to solve the voltage amplitude and phase angle of each node, the active power flow and reactive power flow of each line, and the power balance state of the whole system;

[0031] According to the power flow calculation result, a real-time monitoring model of the full-network power flow distribution state is constructed, including the voltage of each node, the power flow of each branch and the system power state.

[0032] It should be noted that the power grid topology information refers to a complete information set describing the structure and connection relationship of the power grid, including all nodes, branches and their attributes. The node mainly refers to the bus, substation, load point and the position and capacity of the accessible new energy or energy storage device, each node contains parameters such as voltage level and power demand; the branch mainly refers to the power transmission channel such as transmission line, transformer and other power transmission channels, and its attributes include impedance, rated capacity, transformation ratio and switch state; and the connection relationship between the nodes and branches is also included, reflecting the topology structure of the power grid.

[0033] Further, parameter matching and parameter mapping are performed on the standardized input data set, specifically as follows:

[0034] The standardized input data set is initialized as a particle set, each particle representing a state in the node-branch network model, and the initial weight of each particle is set according to the initial uncertainty, and the initial weight may be , wherein N is the total number of particles;

[0035] Based on the power grid topology structure and the node-branch network model, the predicted state of each particle is calculated by a power flow approximation method, the predicted state including the injected power and voltage of each node, the power flow size and direction of each branch, forming a particle predicted state set;

[0036] The real-time measurement data actually collected is compared with the particle predicted state, and the updated weight of each particle is calculated. The weight update can adopt a Gaussian likelihood function form;

[0037] The updated weight is specifically calculated according to the following formula:

[0038]

[0039] In the formula, is the updated weight of the i th particle at the time step k, is the weight of the i th particle at the previous time step k-1, is an exponential function for mapping the observation error into a weight attenuation factor, is a standard form of a Gaussian likelihood function, which measures the difference between the particle predicted state and the real-time measurement data, is the real-time measurement data at the time step k, which includes all measurable grid parameters such as node voltage amplitude, voltage phase angle, branch power flow, active / reactive power, is the particle predicted state of the i th particle at the time step k, and R is a measurement noise covariance matrix, which reflects the uncertainty and noise intensity of each real-time measurement data.

[0040] The particle set is resampled, and the particles with an updated weight lower than a preset weight threshold are eliminated, and the particles higher than the preset weight threshold are retained, and new particles are generated, so that the number of particles remains N, to enhance the robustness to abnormal data and measurement noise;

[0041] The resampled particle set is obtained by weighted average to obtain the optimal particle state, and is mapped into the node-branch network model as the optimal estimation value of the node voltage amplitude, phase angle and branch power flow, to realize the matching of the standardized input data and the node-branch network model.

[0042] It should be noted that the particle filtering scheme has the advantages of parameter matching and parameter mapping, which can accurately correspond the multi-source, real-time, possibly missing or noisy new energy and conventional unit operation data to the grid node-branch model, and through the prediction, weight update and weighted average processing of the particle set, the optimal estimation of the node voltage, phase angle and branch power flow is realized, so that the input data is completely matched with the network model in structure and value, the accuracy, robustness and real-time of the power flow calculation are improved, and the new energy output fluctuation, unit start-stop and topology change are dynamically responded to, to provide a reliable foundation for real-time monitoring, dynamic power transmission capacity calculation and dispatching optimization.

[0043] Further, the matched standardized input data set is input into the full-network power flow monitoring framework, and the improved Newton-Raphson iteration method is adopted for power flow calculation, specifically:

[0044] The matched standardized input data set is structured into an initial power flow state vector, which includes an initial value of a voltage amplitude, an initial value of a phase angle, injected active power, injected reactive power of each node, and impedance parameters and rated capacity of each branch;

[0045] Based on the power grid topology and the whole network power flow monitoring framework, a whole network power flow nonlinear equation, i.e., a node power imbalance function f(x)=0, is established, wherein:

[0046]

[0047] In the formula, is the injected active power of the i-th node, is the injected reactive power of the i-th node, is the voltage amplitude of the i-th node, is the voltage amplitude of the j-th node, is the difference between the voltage phase angles of the i-th node and the j-th node, is the real part of the admittance matrix, representing the active transmission capacity between the i-th node and the j-th node, is the imaginary part of the admittance matrix, representing the reactive transmission capacity between the i-th node and the j-th node, and x is the initial power flow state vector, including the voltage amplitudes and phase angles of all nodes, is the total number of nodes in the power grid.

[0048] The Jacobian matrix J(x) is constructed by a sparse matrix, only the non-zero elements are saved, and in each iteration, if the system state change is less than a preset average change value, the previous Jacobian matrix is reused, otherwise only part of the sub-matrix is updated, so as to reduce the matrix construction overhead;

[0049] Based on the improved Newton-Raphson iteration method, the initial power flow state vector is iteratively updated until a preset maximum iteration number is reached, and convergence is performed;

[0050] The improved Newton-Raphson iteration method has the following specific calculation formula:

[0051]

[0052] In the formula, is the initial power flow state vector at the n-th iteration, is the initial power flow state vector at the n+1-th iteration, is the node power imbalance function at the n-th iteration, is the Jacobian matrix of the node power imbalance function with respect to the initial power flow state variable at the n-th iteration.

[0053] When the iteration converges, the voltage amplitude and phase angle of each node, the active and reactive power flow of each line, and the power balance state of the whole system are obtained, forming a complete power flow calculation result.

[0054] It should be noted that in each iteration, the power imbalance in the current state is calculated The linear approximation relationship of the power equation is described by using the Jacobian matrix The correction amount is obtained The state is updated using to gradually approach the solution that makes the power imbalance function zero.

[0055] By improving the Newton-Raphson iteration method for power flow calculation, the calculation efficiency and convergence stability can be significantly improved while ensuring the calculation accuracy. The role is to accurately combine the mapped standardized input data with the whole network power flow monitoring framework, and quickly solve the voltage amplitude and phase angle of each node, the active and reactive power flow of each branch, and the power balance state of the whole system. The advantage is that the use of sparse matrix technology and partial Jacobian update mechanism reduces the calculation amount of large-scale power grid operation, improves algorithm efficiency, and meets the real-time monitoring demand.

[0056] S2, based on the monitoring model, a new energy output scenario library is constructed by combining a probability statistical method, specifically including quantifying the uncertainty of new energy output based on the real-time monitoring model, and constructing a plurality of new energy output scenario libraries by combining a probability statistical method.

[0057] In this embodiment, based on the real-time monitoring model, the uncertainty of new energy output is quantified, and a plurality of new energy output scenario libraries are constructed by combining a probability statistical method, specifically:

[0058] Based on the real-time monitoring model, a first data set related to new energy is extracted, and the first data set includes a historical power curve and real-time output data of each new energy unit;

[0059] The first data set is subjected to probability distribution fitting to obtain the probability distribution of new energy output, and the probability distribution includes marginal probability distribution and joint probability distribution, which are used to quantify the randomness and fluctuation range of the output;

[0060] Wherein, for single-unit output, a Gaussian mixture model or a non-parametric kernel density estimation method can be used to obtain the marginal probability distribution, and for multi-unit joint output, a Copula function can be used to construct a joint probability distribution;

[0061] A random sampling method is used to generate an initial random scenario set from the probability distribution, and each scenario corresponds to the power output state of each new energy unit;

[0062] The random sampling method can be Monte Carlo sampling, Latin hypercube sampling or importance sampling.

[0063] The initial random scene set is subjected to dimension reduction and clustering processing, representative scenes are screened out, and an initial probability weight is assigned to each scene to represent the probability of occurrence thereof;

[0064] The PCA principal component analysis method is used to extract the most important power fluctuation mode (a core mode describing the overall change trend of new energy output and capable of explaining most of the fluctuations, used to reduce complexity and retain key fluctuation information of the system), the K-means clustering, spectral clustering or hierarchical clustering algorithm is applied to the dimension-reduced data, representative scenes are screened out, and the original power curve corresponding to each cluster center is retained as a representative scene;

[0065] Based on the representative scenes, real-time monitoring models are combined to perform multi-scenario power flow calculation and verification, and node voltage, branch power flow and system power balance state are output;

[0066] For each scene, the improved Newton-Raphson method is used to calculate the power flow of the whole network, the voltage amplitude and phase angle of each node, the active and reactive power flow of each branch and the power balance state of the whole system are output, the scene feasibility is verified, including voltage out-of-limit, line overload and power balance constraints, and scenes that do not meet the safety constraints are eliminated;

[0067] According to the multi-scenario power flow calculation results, a plurality of new energy output scene libraries are generated, which include new energy output and corresponding power flow state under different time scales.

[0068] It should be noted that by quantifying the uncertainty of new energy output and generating a plurality of representative output scenes, the impact of new energy power fluctuation on power grid dispatching and operation can be fully considered. The specific effects are as follows: firstly, through probability distribution fitting, the randomness of each unit output can be accurately described; secondly, the representative scenes generated by random sampling and dimension reduction clustering not only retain the main fluctuation trend of the system, but also greatly reduce the calculation complexity; then, combined with real-time power flow calculation, the power balance and safety constraint verification is performed on each scene to ensure that the generated scenes are not only feasible but also representative; finally, the plurality of new energy output scene libraries can provide real-time and reliable data support for power grid dispatching optimization, risk assessment and power transmission capacity calculation, significantly improving the safety, reliability and flexibility of the power grid under high proportion of new energy output, and reducing the dependence on offline prediction and artificial experience.

[0069] S3, obtaining critical operation states of the power grid under each probability scenario based on the scenario library, to form a mapping relationship between the power transmission capacity and the scenario, specifically including obtaining the critical operation states of the power grid under each scenario according to the multiple new energy output scenario library, and determining the instantaneous maximum power transmission capacity of each power transmission channel, to form a scenario-capacity mapping table.

[0070] In this embodiment, the critical operation states of the power grid under each scenario are obtained according to the multiple new energy output scenario library, specifically including:

[0071] Each representative scenario in the multiple new energy output scenario library is subjected to power flow calculation to obtain dynamic operation data of each scenario on a continuous time sequence, the dynamic operation data including time sequence data of node voltage amplitude, phase angle, and active and reactive power flow of each branch, and the power flow calculation is performed by improved Newton-Raphson iteration method;

[0072] For each scenario, dynamic margin indicators are calculated according to the dynamic operation data, the dynamic margin indicators including node voltage dynamic margin, branch power flow dynamic margin, and system power balance deviation dynamic margin;

[0073] Among them, the node voltage dynamic margin is the difference between the node voltage at each time step and the upper and lower limits, and the minimum margin is recorded; the branch power flow dynamic margin is the difference between the branch power at each time step and the rated capacity, and the maximum occupation ratio and the extreme value are recorded; the system power balance deviation dynamic margin is the extreme value and the change rate of the difference between the total power generation and the load and loss at each time step;

[0074] The critical operation state under each scenario is identified according to the dynamic margin indicators;

[0075] The dynamic margin indicators of each scenario are statistically analyzed according to the time sequence to extract key indicators, the key indicators including the minimum margin, the maximum occupation ratio, the change rate, and the extreme fluctuation time point;

[0076] The critical operation state under each scenario is identified based on the key indicators, the critical operation state including the time when the node or branch appears the minimum margin or the maximum occupation ratio, the time point when the system power balance deviation reaches the preset critical threshold, and the voltage or power flow instantaneous over-limit situation possibly caused by short-time rapid fluctuation.

[0077] It should be noted that the purpose of identifying the critical operating state is to find the limit operating condition that the power grid may reach under the high fluctuation condition of new energy, and the dynamic margin index can accurately quantify the time and amplitude at which the node voltage, branch power flow and system power balance approach the critical state. This not only helps to identify the weakest link and potential risks of the system, but also provides reliable data basis for subsequent dispatch optimization and transmission capacity adjustment, so that the power grid can take timely measures when facing rapid output fluctuation or extreme scenarios, improve safety, stability and operation flexibility, and reduce the risk of accidents caused by overload or voltage out-of-limit.

[0078] Further, the instantaneous maximum transmission capacity of each transmission channel is determined, and a scenario-capacity mapping table is formed, specifically:

[0079] According to each scenario and its corresponding critical operating state, the transmission channels are analyzed one by one, and the transmission power of the transmission channel is gradually increased or decreased until the preset limit condition is reached, and the instantaneous maximum transmission capacity of each transmission channel under the current scenario is obtained, as shown in Figure 3

[0080] Among them, the preset limit condition is to gradually increase the transmission power with the branch power flow margin as the constraint until any node or branch reaches its safety limit, while keeping the active and reactive power balance of the system within the preset allowable deviation range, and keeping the voltage of each node within the allowable safety range.

[0081] Repeat the analysis operation for all transmission channels under each scenario to form the transmission capacity matrix corresponding to the scenario, which records the instantaneous maximum transmission capacity of each transmission channel under the scenario;

[0082] Based on the initial probability weight of each scenario, a conditional probability correction model is constructed to output a corrected probability weight, and the correction model uses the Bayesian conditional probability formula;

[0083] The conditional probability correction model has the following specific calculation formula:

[0084]

[0085] In the formula, is the corrected probability weight of scenario under the current power grid operating state Y, is the conditional probability of the occurrence of operating state Y under scenario , is the initial probability weight, is the normalization processing of all scenarios.

[0086] ​​The power grid operation state includes voltage amplitude and phase angle of each node, power flow of each branch, active / reactive power, load demand, and output level of each new energy unit.

[0087] The modified probability weight is substituted for the initial probability weight, combined with the power transmission capacity matrix corresponding to each scenario, and a weighted average method is used to obtain a modified weighted power transmission capacity matrix.

[0088] The modified weighted power transmission capacity matrix has the following specific calculation formula:

[0089]

[0090] In the formula, is a probability-modified weighted power transmission capacity matrix, is a power transmission capacity matrix under a scenario.

[0091] The scenario-capacity mapping table is generated based on the modified weighted power transmission capacity matrix.

[0092] It should be noted that the generation of the scenario-capacity mapping table can associate the power transmission channel instantaneous maximum capacity under each scenario with the scenario occurrence probability and the operation state, provide a quantitative basis for the new energy output adjustment strategy and the power grid operation mode, enable the dispatching system to accurately judge the power transmission limit that the power grid can bear under different operation conditions, and thus realize joint optimization. Based on the mapping table, the system can dynamically adjust the power transmission limit, so that it is updated adaptively with the change of the new energy output and the power grid state, ensures the safe and stable operation of the power grid, maximizes the consumption capacity of renewable energy, and improves the overall operation efficiency and flexibility.

[0093] S4, according to the real-time operation state, matching the mapping relationship, correcting and jointly optimizing to obtain an adaptive power transmission limit, specifically including:

[0094] S41, obtaining the real-time power grid operation state and matching and correcting with the scenario-capacity mapping table to obtain the corrected actual available power transmission capacity under the current power grid operation condition,

[0095] S42, and jointly optimizing the new energy output adjustment strategy and the power grid operation mode to generate a power transmission limit adaptive to the power grid state.

[0096] S41, obtaining the real-time power grid operation state and matching and correcting with the scenario-capacity mapping table to obtain the corrected actual available power transmission capacity under the current power grid operation condition, specifically:

[0097] ​Obtaining real-time power grid operation states, and constructing a high-dimensional state vector, the real-time power grid operation states including voltage amplitude and phase angle of each node, power flow, active / reactive power, load demand of each branch, and output level of each new energy unit;

[0098] Extracting a corresponding historical state vector for each historical scene in the scene-capability mapping table;

[0099] Preliminarily screening the current high-dimensional state vector and the historical state vector, obtaining a multi-dimensional similarity measure through Mahalanobis distance calculation, and sorting the multi-dimensional similarity measure from small to large, and selecting a number of scenes with the smallest distance as a subset;

[0100] Predicting the preliminary power transmission capacity of the subset by using a multiple linear regression model, specifically:

[0101] Pairing the power grid operation state of each scene in the subset with the corresponding power transmission capacity matrix to form a training data set;

[0102] Constructing a multiple linear regression model, taking the power transmission capacity as the dependent variable, and taking each element of the power grid operation state as the independent variable, and establishing a regression model;

[0103] The regression model has the following specific calculation formula:

[0104]

[0105] In the formula, is the predicted preliminary power transmission capacity, is the intercept, is the regression coefficient, is the jth power grid operation state characteristic value of the ith scene, is the residual error, is the state vector dimension.

[0106] Training the model parameters , ,..., using the least squares method to minimize the error between the predicted value and the true power transmission capacity, and obtaining the optimal model parameters;

[0107] Inputting the current real-time power grid operation state into the trained regression model to obtain the preliminary power transmission capacity.

[0108] Inputting the preliminary power transmission capacity into the pre-trained random forest regression model, independently predicting the corresponding power transmission capacity by each regression tree to obtain a single tree prediction value set;

[0109] The pre-trained random forest regression model uses state vectors of a historical scenario subset and corresponding power transmission capacities as training data to train multiple regression decision trees, and generates different training samples through a bootstrap method to ensure diversity and robustness of the model.

[0110] The final power transmission capacity prediction value output by the random forest regression model is obtained by averaging the prediction results of all decision trees.

[0111] The final power transmission capacity prediction value is specifically calculated according to the following formula:

[0112]

[0113] In the formula, is the final power transmission capacity prediction value of the random forest regression model, is the total number of decision regression trees contained in the random forest, is the prediction output of the jth regression tree for the current input state vector.

[0114] Based on the variance of the final power transmission capacity prediction value of each tree The modified actual available power transmission capacity with an uncertainty boundary is obtained by correcting the confidence interval of the random forest output result.

[0115] The variance is specifically calculated according to the following formula:

[0116]

[0117] The modified actual available power transmission capacity is specifically calculated according to the following formula:

[0118]

[0119] In the formula, is the modified actual available power transmission capacity, is a preset confidence coefficient, which is found according to a required confidence level in a standard normal distribution table, is the standard deviation of the prediction values of each tree of the random forest.

[0120] It should be noted that the real-time operation state and the high-dimensional characteristics of the historical scene of the current power grid are comprehensively considered, and a multiple linear regression model is used for preliminary prediction, and a random forest regression model is used for further optimization, and the prediction uncertainty is quantified and corrected by the confidence interval, so as to obtain more stable and reliable power transmission capacity estimation. This method can effectively reflect the dynamic operation characteristics of the power grid under different new energy output conditions, ensure that the constraints of the power transmission channel are more scientific and reasonable under high volatility and uncertainty, provide reliable basis for the subsequent joint optimization of new energy output adjustment strategy and power grid operation mode, improve the power grid safety margin and new energy consumption capacity, and enhance the real-time and accuracy of the dispatching decision.

[0121] In S42, the new energy output adjustment strategy and the power grid operation mode are jointly optimized to generate a power transmission limit adaptive to the state of the power grid, specifically:

[0122] A joint optimization objective function is constructed, the joint optimization objective function includes total system power loss, renewable energy output consumption amount and dynamic margin index of power grid safety, and constraint conditions are set, and the confidence interval of the corrected actual available power transmission capacity is used to constrain the instantaneous maximum power transmission capacity of each power transmission channel;

[0123] The joint optimization objective function is specifically calculated as follows:

[0124]

[0125] In the formula, to minimize the joint optimization objective function, , , are weight coefficients respectively, is the total system power loss, is the renewable energy output consumption amount, is the dynamic margin index of power grid safety.

[0126] The constraint conditions include node voltage, branch power flow and power balance constraints, specifically:

[0127]

[0128]

[0129]

[0130] In the formula, is the voltage amplitude, , are the allowed minimum and maximum values of the node voltage respectively, is the active power flow of the jth power transmission line (branch) in the power grid, the corrected actual available transmission capacity of the jth transmission line, the sum of the active power output of all generators (conventional units and new energy), the sum of the active power of all load nodes.

[0131] Based on the rolling time domain optimization algorithm, the current power grid operating state and constraint conditions are input to solve the joint optimization objective function to obtain the optimal output adjustment strategy of each new energy unit and the optimal adjustment scheme of the power grid operating mode;

[0132] According to the joint optimization result, the adaptive limit of each transmission channel is calculated, which is dynamically updated with the fluctuation of new energy output and the change of power grid state, reflecting the instantaneous carrying capacity of the power grid under the current operating condition;

[0133] Among them, the joint optimization model solving result is obtained, including the optimal output adjustment amount of each new energy unit and the power grid operating mode adjustment scheme. The optimization result is calculated by the improved Newton-Raphson iteration method, the power flow distribution and node voltage of each branch are recalculated, and the instantaneous carrying capacity of each transmission channel under the current power grid state is obtained. The aforementioned instantaneous carrying capacity is combined with the confidence interval of the corrected actual available transmission capacity to obtain the adaptive limit, which is recorded as the transmission channel constraint value for the reference of the dispatching and control system, and the limit is dynamically updated when the new energy output or the power grid state changes, ensuring that the limit changes adaptively with the power grid operating conditions;

[0134] The adaptive transmission limit is real-time issued to the dispatching and control center, and a closed-loop updating mechanism is established: according to the new real-time operating state, periodic or event-triggered updating is performed to ensure that the limit is always consistent with the power grid state, realizing dynamic adjustment and safe operation of the transmission capacity.

[0135] The present application realizes real-time acquisition of new energy and conventional unit operating data, and combines the power grid topology structure and operating mode to construct a high-precision, dynamically updated whole network power flow distribution state real-time monitoring model; particle filtering is used to realize accurate matching of standardized input data and node-branch network model, ensuring the accuracy and robustness of power flow calculation; based on the improved Newton-Raphson iteration method, efficient power flow calculation is performed to improve the operation speed and convergence stability of large-scale power grids; the uncertainty of new energy output is further quantized, representative scenario library is generated through probability distribution fitting, random sampling and dimensionality reduction clustering, and power flow verification is performed on each scenario to ensure the feasibility and representativeness of the scenario.

[0136] The high-precision and dynamically updated power flow monitoring model and the particle filtering data matching are used to realize the accurate correspondence between the input data and the node-branch network model, and ensure the accuracy and robustness of the power flow calculation; the uncertainty of the new energy is quantified based on multiple output scenarios, and the critical operating state under each scenario is identified to form a scenario-capacity mapping table, thereby providing a scientific basis for the instantaneous carrying capacity of the power transmission channel; further, the actual available transmission capacity is obtained through the combination of the multiple linear regression and random forest regression models and the confidence interval correction; finally, the transmission limit is generated by the joint optimization and rolling horizon algorithm, and the dynamic adjustment and safe operation of the transmission capacity are realized.

[0137] Referring to Figure 2 As shown in the figure, the application provides a regional power grid predictive scheduling system structure based on topology balance, which comprises a model construction module, a scenario library construction module, a mapping table generation module and a transmission limit generation module, and there is a connection between the modules:

[0138] The model construction module is used to establish a monitoring model reflecting the real-time power flow of the whole network.

[0139] The scenario library construction module is used to construct a new energy output scenario library based on the monitoring model and in combination with a probability statistical method.

[0140] The mapping table generation module is used to obtain the critical operating state of the power grid under each probability scenario based on the scenario library, form the mapping relationship between the transmission capacity and the scenario, and obtain the scenario-capacity mapping table through the weighted average processing of the transmission capacity matrix by correcting the probability weight.

[0141] The transmission limit generation module is used to match the mapping relationship according to the real-time operating state, correct and jointly optimize to obtain the adaptive transmission limit, wherein the correction is performed by correcting the confidence interval through variance calculation, and the optimization is performed by constructing a multi-objective joint optimization function and constraint condition in combination with the confidence interval correction.

[0142] The above formulas are all dimensionless values, and the formulas are obtained by collecting a large amount of data to simulate the latest real situation, and the preset parameters in the formulas are set by the person skilled in the art according to the actual situation.

[0143] The above embodiments can be realized by software, hardware, firmware or any combination thereof, in whole or in part. When realized by software, the above embodiments can be realized in the form of a computer program product in whole or in part.

[0144] Those skilled in the art can understand that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0145] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0146] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0147] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A predictive dispatch method for regional power grids based on topology balance, characterized in that, include: Establish a monitoring model that reflects the real-time power flow of the entire network; Based on the monitoring model, a new energy output scenario library is constructed using probabilistic statistical methods; Based on the scenario library, the critical operating state of the power grid under each probability scenario is obtained, forming a mapping relationship between transmission capacity and scenario. The mapping relationship is processed by weighted averaging of the transmission capacity matrix by correcting the probability weights to generate a scenario-capacity mapping table. Based on the real-time operating status matching mapping relationship, an adaptive transmission limit is derived through correction and joint optimization. The correction includes adjusting the confidence interval by calculating the variance to obtain the corrected actual available transmission capacity, specifically: A high-dimensional state vector is constructed based on the real-time running status, and the state vector corresponding to the historical scene is extracted. The current high-dimensional state vector is initially screened by comparing it with the corresponding historical state vectors. Based on the Mahalanobis distance calculation method, several scenarios with the smallest distance are selected as a subset. The transmission capacity data in the subset is predicted using a multiple linear regression model to obtain the preliminary transmission capacity; The initial power transmission capacity is input into a pre-trained random forest regression model, and the final power transmission capacity prediction value is obtained through multi-decision tree ensemble prediction. Variance is calculated based on the final predicted transmission capacity, confidence interval is corrected based on the variance, and the corrected actual available transmission capacity is output. The optimization is achieved by combining confidence interval correction to construct a multi-objective joint optimization function and constraints, specifically as follows: A multi-objective joint optimization function is constructed, which includes dynamic margin indicators of total system power loss, renewable energy output and consumption, and grid security. Constraints are set, and the confidence interval of the modified actual available transmission capacity is used to constrain the instantaneous maximum transmission capacity of each transmission channel. The optimal adjustment scheme for new energy output strategy and power grid operation mode is obtained by solving the optimization function using the rolling time-domain optimization algorithm. The adaptive limit for each transmission channel is calculated and sent to the dispatching system, and a closed-loop update mechanism is established to realize the dynamic adjustment of transmission capacity.

2. The regional power grid predictive dispatch method based on topology balancing according to claim 1, characterized in that, The establishment of a monitoring model reflecting the real-time power flow of the entire network specifically includes: Real-time power grid data is preprocessed to form a standardized input dataset; A node-branch network model is constructed based on the power grid topology and dynamically assigned values ​​to form a power flow monitoring framework. The standardized input dataset is matched and mapped with parameters, and then fed into the power flow monitoring framework. The improved Newton-Raphson iterative method is used to calculate the power flow. Based on the calculation results, a real-time power flow monitoring model for the entire network is generated, which includes node voltage, branch power flow, and system power status.

3. The regional power grid predictive dispatch method based on topology balancing according to claim 2, characterized in that, The parameter matching and parameter mapping of the standardized input dataset specifically involves: Initialize the normalized input data into a set of particles and assign initial weights; Predicted states of each particle are calculated based on the power grid topology. Acquire real-time measurement data and compare it with the predicted state of the particles, calculate the updated weight of each particle, and resample the particle set; The optimal particle state is obtained by weighted averaging the resampled particle set and then mapped to the node-branch network model.

4. The regional power grid predictive dispatch method based on topology balancing according to claim 2, characterized in that, The improved Newton-Raphson iterative method is used for power flow calculation, specifically as follows: Construct an initial power flow state vector based on the matched standardized input dataset; Based on the power grid topology and the whole-network power flow monitoring framework, establish the whole-network power flow nonlinear equation; The Jacobian matrix is ​​constructed using sparse matrices, and the initial power flow state vector is solved iteratively to output power flow calculation results including node voltages, line power flow, and system power.

5. The regional power grid predictive dispatch method based on topology balancing according to claim 4, characterized in that, The construction of a new energy output scenario library based on the monitoring model and combining probabilistic statistical methods specifically includes: The first dataset is extracted based on the real-time monitoring model, and a probability distribution is fitted to the first dataset. An initial set of random scenes is generated from a probability distribution using a random sampling method. The initial random scene set is subjected to dimensionality reduction and clustering to select representative scenes and assign initial probability weights to each scene; Based on representative scenarios, multi-scenario power flow calculations and verifications are performed to generate a new energy output scenario library that includes node voltage, branch power flow, and system power balance status.

6. The regional power grid predictive dispatch method based on topology balancing according to claim 5, characterized in that, The process of obtaining the critical operating state of the power grid under each probability scenario based on the scenario library specifically includes: Time-series power flow calculations are performed on representative scenarios in the new energy power output scenario library to obtain dynamic operation data; The dynamic margin index for each scenario is calculated based on the dynamic operating data. Identify critical operating states in various scenarios based on dynamic margin indicators.

7. The regional power grid predictive dispatch method based on topology balancing according to claim 1, characterized in that, The mapping relationship is obtained by weighting the transmission capacity matrix with corrected probability weights to generate a scenario-capacity mapping table. The specific steps are as follows: Analyze the maximum power transmission capacity of power transmission channels under various scenarios to form a scenario power transmission capacity matrix; A conditional probability correction model is constructed based on the initial probability weights for each scenario, and the corrected probability weights are output through the probability correction model. The corrected probability weights are combined with the transmission capacity matrix corresponding to each scenario, and the weighted average method is used to obtain the corrected weighted transmission capacity matrix. A scenario-capacity mapping table is generated based on the modified weighted transmission capacity matrix.

8. A system using the topology-balanced regional power grid predictive dispatch method as described in any one of claims 1-7, characterized in that, include: The model building module is used to build a monitoring model that reflects the real-time power flow of the entire network. The scenario library construction module is used to construct a new energy output scenario library based on the monitoring model and combined with probabilistic statistical methods; The mapping table generation module is used to obtain the critical operating state of the power grid under each probability scenario based on the scenario library, form a mapping relationship between transmission capacity and scenario, and generate a scenario-capacity mapping table by performing weighted averaging on the transmission capacity matrix after correcting the probability weights. The power transmission limit generation module is used to match the mapping relationship according to the real-time operating status, correct and jointly optimize to obtain the adaptive power transmission limit. The correction includes correcting the confidence interval by calculating the variance to obtain the corrected actual available power transmission capacity. The optimization is carried out by constructing a multi-objective joint optimization function and constraints by combining the confidence interval correction.

Images