CN-121998406-A - New energy access risk assessment method and device, electronic equipment and storage medium

CN121998406ACN 121998406 ACN121998406 ACN 121998406ACN-121998406-A

Abstract

The application discloses a new energy access risk assessment method, a device, electronic equipment and a storage medium, wherein the method comprises the steps of S1, collecting multi-source heterogeneous data in real time, S2, calculating the short-circuit ratio of a new energy multi-station in real time through multi-source heterogeneous data fusion, S3, extracting and constructing a practical risk feature vector of engineering, including transient overvoltage features and corresponding network structure risk features, S4, constructing a nonlinear dynamic mapping model between the short-circuit ratio of the new energy multi-station and the transient overvoltage key features by using a deep learning model, S5, carrying out risk assessment and grading according to the short-circuit ratio of the new energy multi-station and the nonlinear dynamic mapping model calculated according to the data collected in real time, and S6, when the risk grade reaches an early warning threshold or the transient overvoltage predicted value exceeds a limit value, automatically triggering early warning and generating a decision suggestion by the system. The risk assessment method and the risk assessment system remarkably improve the efficiency, accuracy and engineering applicability of risk assessment and improve the automation and intelligence level.

Inventors

DENG JIAOJIAO
ZHENG XIAOMING
ZHANG YONGHAO
WANG YAO
FU ZIYU
HU YINGYING
SHI ZHENG
SHI JINKAI
LIANG YAN
WANG KAIKAI
REN AIPING

Assignees

国网山西省电力有限公司经济技术研究院

Dates

Publication Date: 20260508
Application Date: 20251218

Claims (10)

1. The new energy access risk assessment method is characterized by comprising the following steps: S1, acquiring multi-source heterogeneous data from SCADA/EMS, WAMS/PMU, a new energy station monitoring system, a weather information system and an equipment parameter library in real time through a data access module; s2, calculating a new energy multi-station short-circuit ratio MRSCR of a core index representing the strength of the power grid in real time through multi-source heterogeneous data fusion; S3, extracting and constructing engineering practical risk feature vectors from different dimensions according to relevant standards, wherein the engineering practical risk feature vectors comprise transient overvoltage TOV features and corresponding network structure risk features; S4, constructing a nonlinear dynamic mapping model between the short-circuit ratio MRSCR of the new energy multi-station and the key characteristics of the transient overvoltage TOV by using the deep learning model; S5, performing real-time dynamic risk assessment and risk classification according to a new energy multi-station short-circuit ratio, a new energy multi-station short-circuit ratio MRSCR and a nonlinear dynamic mapping model between transient overvoltage TOV key features, which are calculated according to data acquired in real time; And S6, when the risk level reaches an early warning threshold value or the transient overvoltage TOV predicted value exceeds a limit value, automatically triggering early warning by the system and generating a decision suggestion.
2. The new energy access risk assessment method according to claim 1, wherein the step S1 specifically includes the steps of: S11, collecting multi-source heterogeneous data from SCADA/EMS, WAMS/PMU, new energy station monitoring system, weather information system and equipment parameter library in real time through a data access module, wherein the collected key data comprises active output of each new energy station Reactive power And operating state, grid bus voltage The system topology, the power flow distribution and the meteorological information comprising wind speed and illumination intensity; S12, the collected original multi-source heterogeneous data is subjected to normalization through a preprocessing module, wherein the normalization comprises time sequence alignment, anomaly detection and elimination, missing data repair and per unit processing so as to eliminate dimension influence and provide consistent and reliable data input for subsequent calculation and analysis.
3. The new energy access risk assessment method according to claim 2, wherein the step S2 specifically includes the steps of: s21, calculating equivalent impedance of a target collection point based on real-time power grid topology and running state The value impedance is obtained through a network node impedance matrix and a multipoint network equivalent calculation engine; S22, calculating the equivalent short-circuit capacity of the target collection point The calculation formula is as follows: ; Wherein, the As a reference voltage, the reference voltage is set, Conjugation for voltage phasors; s23, calculating a new energy multi-station short circuit ratio MRSCR of a core index representing the strength of the power grid: ; Wherein, the And Respectively are bus bars And bus bar The new energy injection power of the (a) is set, Is a bus bar Is a multi-point equivalent self-impedance of davin, Is a bus bar And bus bar The multi-point Thevenin equivalent transimpedance therebetween.
4. The new energy access risk assessment method according to claim 3, wherein the step S3 specifically includes the steps of: S31, transient overvoltage TOV (total internal voltage) characteristic extraction, namely based on real-time power grid topology and running state, simulating various expected fault types by adopting a high-precision electromechanical transient simulation platform, generating a large amount of corresponding system transient response data under different short circuit ratio levels, and further generating disturbance windows And (3) extracting peak value, duration and overrun energy of the transient overvoltage, wherein: Peak transient voltage The method comprises the following steps: ; in the formula, In order to evaluate the peak voltage within the window, Is a node The voltage timing profile is a function of the voltage, As a threshold value of the transient voltage, Detecting a time window range; overrun duration (S) is: ; Wherein, the Taking 1 as an indication function to meet the condition, otherwise taking 0; overrun energy The method comprises the following steps: ; in the formula, Measures the comprehensive damage degree of the overvoltage intensity and the persistence, ; S32, extracting risk features of the network structure, wherein the extracting comprises the steps of calculating an electrical distance, and the calculation formula is as follows: ; where V i represents the bus i voltage amplitude and Q j represents the bus j injection of reactive power, the electrical distance reflecting the degree of electrical coupling between the nodes.
5. The new energy access risk assessment method according to claim 4, wherein the step S4 specifically includes the steps of: S41, constructing a nonlinear dynamic mapping model between a short-circuit ratio MRSCR of a new energy multi-station and key characteristics of transient overvoltage TOV by adopting a long-short-term memory network LSTM suitable for time sequence data processing, wherein: Input sequence of the nonlinear dynamic mapping model A status feature within a pre-fault time window, comprising: ; Wherein, the Short-circuit ratio of multiple stations for new energy; Is the first Active power output of the individual stations; Is the first Reactive power output of the individual stations; is the electrical distance; is weather data; The network state of the nonlinear dynamic mapping model is updated by LSTM updating the cell state through the internal gating mechanism Hidden state Thereby capturing the long-term dependency between the running state and the subsequent transient response, the state update formula of the LSTM network is as follows: ; ; ; ; ; ; Wherein, the , , The states of the forget gate, the input gate and the output gate are updated respectively, As a result of the candidate memory cell, In order to be in the output state, 、、、 Respectively a weight matrix input to each gate, 、、、 The cyclic weight matrix from the last hidden state to each gate, 、、、 Respectively corresponding offset vectors; The output of the nonlinear dynamic mapping model is the predicted value of TOV characteristics: ; Wherein, the In order to predict the transient overvoltage peak value, For the duration of time it is possible, Is predicted overrun energy; the training process of the nonlinear dynamic mapping model is performed by minimizing the following loss function: ; Wherein, the The weight coefficient of the loss term is used for adjusting the importance of different indexes; The nonlinear dynamic mapping model has online updating capability, continuously absorbs new data through incremental learning, adapts to power grid structure change and novel equipment access, and has an updating formula as follows: ; Wherein, the As a parameter of the model, it is possible to provide, In order for the rate of learning to be high, Gradient of the loss function with respect to model parameters.
6. The new energy access risk assessment method according to claim 5, wherein the step S5 specifically includes the steps of: s51, carrying out real-time dynamic risk assessment according to a nonlinear dynamic mapping model between the new energy multi-station short-circuit ratio, the new energy multi-station short-circuit ratio MRSCR and the transient overvoltage TOV key characteristics, which are calculated according to data acquired in real time, and calculating a comprehensive risk index: ; Wherein w 1 、w 2 、w 3 represents the weight of each sub-risk; The MRSCR sub-risk calculation is: ; The TOV sub-risk calculation is: ; Wherein β 1 、β 2 、β 3 、β 4 represents a weighted portion of each risk indicator; the calculation formula of the loss sub-risk of the wind discarding quoted flat value is as follows: ; wherein, the calculation formula of the loss value of the abandoned wind abandoned quoted flat value is as follows: ; Wherein: discarding the electric quantity of the light for discarding wind; Is that Time period new energy available output; Is that The actual power is sent out in the period; for the period length (h), Is that The new energy average online electricity price or market price in time interval; S52, risk grade division, wherein the risk level of the power grid is divided into different grades by the comprehensive risk index RI, and the specific division method is as follows: ; The gamma 1 、γ 2 、γ 3 、γ 4 is a preset threshold value set according to the safety standard, the historical data or the expert experience of the power grid, dynamic risk classification is carried out according to the preset threshold value, the risk level is classified into 1 to 5 levels, and the risk level corresponds to the safety state, the attention state, the early warning state, the high-police state and the critical state respectively, so that clear situation perception is provided for operators.
7. The new energy access risk assessment method according to claim 6, wherein the step S6 specifically includes the steps of: S61, when the risk level reaches an early warning threshold value or the TOV predicted value exceeds a limit value, automatically triggering early warning by the system and generating decision advice, wherein the generated decision advice comprises real-time regulation advice and supporting planning configuration advice, and the real-time regulation advice and the supporting planning configuration advice are: The real-time regulation suggestion is based on a rule base or a rapid optimization algorithm, and executable regulation measures are output: (1) Reactive power regulation advice: ; in the formula, Represent the first The new energy station/reactive compensation device is provided with a current reactive set value, Indicating the recommended reactive power adjustment quantity, 、 Respectively represent the first The reactive lower/upper limit allowed by the individual object, The formula provides an optimization strategy for reactive power regulation for a cut-off function, and ensures the voltage stability of a power grid; (2) Active limited send advice: Sequencing the voltage sensitivity of the collection points according to all the stations, and preferentially limiting the station output force with the weakest voltage support; The supporting planning configuration proposal is used for fundamentally improving the strength of the power grid, providing the capacity configuration proposal of the supporting power supply and the equivalent short circuit supporting capacity required to be supplemented The calculation formula is as follows: ; Wherein, the For the lowest threshold value of the safe operating requirement, And Respectively are bus bars And bus bar The new energy injection power of the (a) is set, Is a bus bar Is a multi-point equivalent self-impedance of davin, Is a bus bar And bus bar The multi-point inter-impedance of the equivalent of the Thevenin, Equivalent short-circuit capacity of the target collection point.
8. New energy access risk assessment device, characterized by comprising: The real-time data acquisition and preprocessing module is used for acquiring multi-source heterogeneous data from the SCADA/EMS, WAMS/PMU, the new energy station monitoring system, the weather information system and the equipment parameter library in real time through the data access module; the short-circuit ratio online calculation module is used for calculating the new energy multi-station short-circuit ratio MRSCR of the core index representing the power grid strength in real time through multi-source heterogeneous data fusion; The multi-dimensional risk feature extraction and structuring characterization module extracts and constructs engineering practical risk feature vectors from different dimensions according to relevant standards, wherein the risk feature vectors comprise transient overvoltage TOV and corresponding electrical distances; The mapping model construction module is used for constructing a nonlinear dynamic mapping model between the short-circuit ratio MRSCR of the new energy multi-station and the transient overvoltage TOV key characteristics by using the deep learning model; The real-time risk assessment and dynamic grading module is used for carrying out real-time dynamic risk assessment and risk grading according to a nonlinear dynamic mapping model between the new energy multi-station short-circuit ratio, the new energy multi-station short-circuit ratio MRSCR and the transient overvoltage TOV key characteristics, which are calculated according to the data acquired in real time; And the automatic early warning and accurate decision suggestion generation module is used for automatically triggering early warning and generating decision suggestions when the risk level reaches an early warning threshold value or the transient overvoltage TOV predicted value exceeds a limit value.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the new energy access risk assessment method according to any one of claims 1 to 7 when executing the computer program.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the new energy access risk assessment method according to any one of claims 1 to 7.

Description

New energy access risk assessment method and device, electronic equipment and storage medium Technical Field The application relates to the technical field of safety and stability analysis and risk assessment of a new energy power system, in particular to a new energy access risk assessment method, a new energy access risk assessment device, electronic equipment and a storage medium. Background With the transformation of the global energy structure to clean low carbon, new energy power generation such as wind power, photovoltaic and the like is connected into a power system at an unprecedented speed. The centralized grid connection of the large-scale and high-proportion new energy sources enables the power system to show the double-high characteristic of the high-proportion new energy sources and the high-proportion power electronic equipment. The inherent volatility, intermittence and randomness of new energy power generation and the weak inertia, low damping and low short-circuit capacity characteristics of a power electronic interface thereof bring serious challenges to the safe and stable operation of a power system. Among them, a significant reduction in short-circuit ratio (SCR) is one of the core problems caused by high-proportion new energy access. The low short-circuit ratio means that the strength of the power grid is weakened, so that the power grid voltage is more sensitive to reactive disturbance, the voltage regulation sensitivity is reduced, the system voltage stability margin is reduced, and the problems of transient voltage instability and the like are easily caused. In addition, when the power grid fails, particularly when the direct current transmission system fails in commutation or is subjected to serious disturbance such as direct current blocking, a serious transient overvoltage problem can occur at the side of the new energy station. Dynamic control characteristics (such as current limiting and reactive compensation strategies) of the power electronic converter of the new energy unit during fault ride-through and nonlinear response of reactive compensation equipment (such as SVC (static var compensator) and SVG/STATCOM) in the station can aggravate amplitude and duration of transient overvoltage, so that the new energy unit is disconnected on a large scale, and safe operation of a power grid is seriously threatened. At present, in order to alleviate the problems of insufficient short-circuit capacity and transient overvoltage of a double-high power system, the stability of the system is often maintained by limiting the output level of a new energy station in practice, which severely restricts the grid-connected power generation capacity and the utilization efficiency of the new energy. Therefore, a method and a system for real-time, automatic and quantitative evaluation of the short-circuit ratio and the transient overvoltage risk of a new energy access point are urgently needed, so that the evaluation efficiency and accuracy are improved, the power grid planning, the operation and the optimal configuration of a supporting power source (such as a camera and a grid-structured energy storage) are guided, and the safe and stable operation of a novel power system and the efficient consumption of new energy are ensured. Most of researches concentrate on the aspect of new energy multi-scene output in the existing new energy access risk assessment field, cluster is conducted aiming at wind-light-load data characteristics, and typical representative days are selected to solve the problems of uncertainty and relevance of new energy access. These methods estimate the system index by enumerating (parsing) or randomly sampling (monte carlo simulation) the system state. Despite the advances made in the prior art in new energy access risk assessment, significant drawbacks remain: 1) The evaluation efficiency is low, real-time response is difficult, and most traditional risk evaluation methods, particularly a Monte Carlo simulation method and detailed manual simulation, have huge calculation amount and long calculation time, and are difficult to meet the real-time response requirement of rapid change of a power system in a new energy access scene. This limits the timeliness of risk early warning and decision making; 2) The nonlinear mapping relation of the Short Circuit Ratio (SCR) and the Transient Overvoltage (TOV) cannot be effectively quantized, and the new energy access has complex nonlinear influence on the power grid Strength (SCR) and the transient voltage behavior. The traditional evaluation model depends on a simplified linear relation or an empirical rule, so that the nonlinear relation of dynamic coupling between SCR and TOV under the access background of a new energy multi-station is difficult to accurately reflect, and the risk evaluation precision is insufficient; 3) The existing assessment method is mostly dependent on manually setting models and parameters, and lacks a unified fram