Probabilistic load flow calculation accuracy evaluation method for power grid

Abstract

The invention discloses a probabilistic load flow calculation accuracy evaluation method for a power grid, and belongs to the technical field of probabilistic load flow calculation accuracy evaluation, and the method comprises the steps: recognizing a power grid region corresponding to probabilistic load flow calculation, and segmenting the power grid region in combination with the probabilistic load flow calculation to obtain a plurality of unit regions; generating a regional information graph according to the unit region and the power grid region; the method comprises the following steps: acquiring analysis result data of analysis in a power grid region based on probabilistic load flow calculation in real time, acquiring corresponding standard result data according to the analysis result data, calibrating the analysis result data according to the standard result data, and identifying an abnormal position corresponding to the analysis result data which does not meet a calibration standard; marking the abnormal position in the regional information graph; identifying a unit area corresponding to the abnormal position in the area information graph, and marking the unit area as a traceability area; and determining an abnormal reason of probabilistic load flow calculation based on the traceability area, and optimizing and adjusting the probabilistic load flow calculation based on the abnormal reason.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of probability power flow calculation accuracy evaluation, and specifically relates to a probability power flow calculation accuracy evaluation method for a power grid. BACKGROUND

[0002] In the actual operation of a power system, network topology structure, transformer ratio, node injection power and other factors are not fixed but have many uncertainties. These uncertain factors pose a potential threat to the safe and stable operation of the power system. With the continuous development of the power industry, large-scale access of new energy represented by solar energy and wind energy to the power grid has brought obvious intermittency and randomness to the power grid. The emergence of new concepts such as microgrid, distributed power supply and electric vehicle in the distribution network has greatly enhanced the interaction between power supply, load and the power grid, further exacerbating the uncertainty of the power system. Under this background, probability power flow calculation emerged as the times require. Probability power flow calculation combines probability theory and can fully consider various uncertain factors in the power system, thereby more comprehensively reflecting the system operation condition.

[0003] However, the current probability power flow calculation accuracy evaluation method for a power grid still has some deficiencies. The existing evaluation method may not be able to comprehensively and accurately measure the deviation between the probability power flow calculation result and the actual system operation condition, and it is difficult to effectively evaluate the processing capability of the algorithm for various uncertain factors and the coverage degree of the complex characteristics of the power grid. Based on this, in order to solve the problem of probability power flow calculation accuracy evaluation, the application provides a probability power flow calculation accuracy evaluation method for a power grid. SUMMARY

[0004] In order to solve the problems existing in the above scheme, the application provides a probability power flow calculation accuracy evaluation method for a power grid.

[0005] The object of the application can be achieved by the following technical scheme: A probability power flow calculation accuracy evaluation method for a power grid comprises: Step 1: identify the power grid area corresponding to the probability power flow calculation, segment the power grid area according to the probability power flow calculation, and obtain a plurality of unit areas; generate a region information map according to the unit areas and the power grid area.

[0006] Further, the method for segmenting the power grid area according to the probability power flow calculation comprises: Set a unit area standard, and set a corresponding unit calibration item according to the unit area standard; Collect the region information of the power grid area, determine the calibration priority order of each unit calibration item, divide the power grid area according to the calibration priority order of the unit calibration item and the region information, and obtain the corresponding unit area.

[0007] Further, the method for determining the calibration priority of each unit calibration item comprises: presetting a power grid simulation area; identifying unit calibration items, arranging the unit calibration items according to different priority orders to obtain several candidate priority sequences; simulating and dividing the power grid simulation area according to the candidate priority sequences to obtain the division efficiency corresponding to the candidate priority sequences; selecting the candidate priority sequence with the highest division efficiency as a target priority sequence, and determining the calibration priority of each unit calibration item according to the target priority sequence.

[0008] Further, the method for dividing the power grid area according to the calibration priority of the unit calibration item and the area information comprises: identifying the calibration priority, and marking the unit calibration items as i according to the calibration priority, i=1, 2, …, n, n being the number of the calibration priority; identifying the calibration item standard corresponding to the unit calibration item i=1 according to the unit area standard, calibrating the area information according to the calibration item standard to obtain a candidate area satisfying the calibration item standard in the current power grid area; identifying the calibration item standard corresponding to the unit calibration item i=2 according to the unit area standard, calibrating the candidate area according to the calibration item standard and the area information, and eliminating the candidate area not satisfying the calibration item standard; identifying the calibration item standard corresponding to the unit calibration item i=3 according to the unit area standard, calibrating the candidate area according to the calibration item standard and the area information, and eliminating the candidate area not satisfying the calibration item standard; and so on, until identifying the calibration item standard corresponding to the unit calibration item i=n according to the unit area standard, calibrating the candidate area according to the calibration item standard and the area information, and marking the candidate area satisfying the calibration item standard as a unit area.

[0009] Further, the method for judging whether the corresponding candidate area satisfies the calibration item standard comprises: extracting the features of the area information according to the unit calibration item to obtain the candidate area information of the corresponding candidate area; establishing a unit calibration model, analyzing the candidate area information corresponding to the corresponding unit calibration item and the calibration item standard through the unit calibration model to obtain the calibration judgment result of the corresponding candidate area, the calibration judgment result including satisfying the calibration item standard and not satisfying the calibration item standard.

[0010] Further, the expression of the unit calibration model is: ; In the formula, (sj , BZ j ) is input data, s j represents the candidate region information of the corresponding unit calibration item, j represents the corresponding unit calibration item, j = 1, 2, …, m, and m is the number of calibration priorities; BZ j represents the calibration item standard of the corresponding unit calibration item; s j → BZ j represents that the corresponding candidate region information meets the calibration item standard; the output data is a single calibration value DP(s j , BZ j ), and the unit calibration value is 1 or 0; When the unit calibration value is 1, the corresponding candidate region meets the calibration item standard; When the unit calibration value is 0, the corresponding candidate region does not meet the calibration item standard.

[0011] Step two: real-time acquisition of analysis result data based on probabilistic power flow calculation in the power grid region, acquisition of corresponding standard result data according to the analysis result data, calibration of the analysis result data according to the standard result data, identification of abnormal positions corresponding to the analysis result data that do not meet the calibration standard; and marking of the abnormal positions in the region information map.

[0012] Further, another method for determining abnormal positions includes: presetting variable verification data, the variable verification data including input variable adjustment data and an output result allowable range; and establishing a verification library by summarizing the variable verification data according to whether they can be applied to verification of corresponding unit regions; real-time identification of analysis result data based on probabilistic power flow calculation in the power grid region, classification of the analysis result data according to unit regions to form unit result data of the corresponding unit regions; and identification of unit input data of the corresponding unit regions according to the unit result data; matching corresponding variable verification data from the verification library according to the unit input data, identification of input variable adjustment data and an output result allowable range corresponding to the variable verification data, simulation analysis according to the input variable adjustment data and the unit input data to obtain simulation analysis result data of the unit region; calibration of the simulation analysis result data and the unit result data according to the output result allowable range, and judgment of whether the corresponding unit region is abnormal; when it is judged that the unit region is normal, no corresponding operation is performed; when it is judged that the unit region is abnormal, determination of corresponding abnormal positions according to the simulation analysis result data and the unit result data.

[0013] Step three: identifying the unit area corresponding to the abnormal position in the regional information map, marking the unit area as a traceable area; determining the abnormal reason of the probabilistic power flow calculation based on the traceable area, and optimizing and adjusting the probabilistic power flow calculation based on the abnormal reason.

[0014] A probabilistic power flow calculation accuracy evaluation device for a power grid comprises: A memory for storing a computer program; A processor for executing the computer program to implement the steps of the above-mentioned method for evaluating the accuracy of probabilistic power flow calculation for a power grid.

[0015] Compared with the prior art, the present application has the following advantages: By reasonably dividing the power grid region based on the probabilistic power flow technology, the originally huge overall power grid calculation task is divided into a plurality of relatively independent sub-regions, which facilitates rapid determination of abnormal regions, and also facilitates reduction of the impact of subsequent optimization and adjustment, helping users to determine the normality of other regions; by determining the calibration priority of each unit calibration item, the efficiency of dividing the power grid region is improved, analysis of regions that do not meet the requirements is avoided, and the analysis efficiency is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0017] Figure 1 The method flowchart of the present application. DETAILED DESCRIPTION

[0018] The technical solutions of the present application will be described in detail below with reference to the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] As shown in Figure 1 A method for evaluating the accuracy of probabilistic power flow calculation for a power grid comprises: Step one: identifying the power grid region corresponding to the probabilistic power flow calculation, dividing the power grid region in combination with the probabilistic power flow calculation, and obtaining a plurality of unit regions; generating a regional information map based on the unit regions and the power grid region, i.e. directly displaying the power grid region and each unit region through the information map.

[0020] The power grid region is segmented in combination with the probabilistic power flow calculation. The core of the probabilistic power flow calculation is to analyze the influence of uncertainties (such as load fluctuation and new energy output) on the state of the power grid. If the input variables (such as load and distributed power supply) in a region are relatively independent and weakly coupled with other regions, the probabilistic power flow calculation result of the region can be approximately independently verified, and the region can be segmented into a unit region. The segmentation can be performed based on the existing method according to the standard.

[0021] In one embodiment, the method for segmenting the power grid region in combination with the probabilistic power flow calculation includes: Setting a unit region standard, such as input independence, topological decoupling, and calculation target consistency. For example, the input independence: the correlation coefficient (such as the Pearson correlation coefficient) between the uncertainty factors (such as load and wind power) in a region and other regions is less than 0.3; the topological decoupling: the electrical connection between regions is weak, or the external influence can be ignored through equivalent methods (such as Thevenin equivalent); the calculation target consistency: the accuracy evaluation of the state variables (such as voltage and power) in a region does not depend on external regions.

[0022] Setting a corresponding unit calibration item according to the unit region standard. Each unit calibration item is set for a specific unit region standard, such as a unit calibration item corresponding to input independence; Collecting the region information of the power grid region, determining the calibration priority order of each unit calibration item, and gradually determining the corresponding candidate region of the power grid region according to the calibration priority order of the unit calibration item and the region information until the corresponding candidate region meets the unit region standard. The corresponding candidate region is marked as a unit region.

[0023] By determining the calibration priority order of each unit calibration item, the segmentation efficiency of the power grid region is improved, the analysis of regions that do not meet the requirements is avoided, and the analysis efficiency is reduced.

[0024] In one embodiment, the calibration priority order of each unit calibration item can be determined based on the existing method, such as direct manual setting.

[0025] In one embodiment, the method for determining the calibration priority order of the unit calibration item includes: According to the historical data, a power grid simulation region convenient for simulation analysis is preset. Generally, the existing power grid region is selected, and preferably the power grid region applying the probabilistic power flow calculation is applied.

[0026] Identifying the unit calibration items, arranging the unit calibration items in different priority orders, and obtaining the candidate priority sequences with various possibilities; According to the corresponding candidate priority sequence, the power grid simulation region is simulated and divided, and the division efficiency corresponding to the corresponding candidate priority sequence is obtained. The calibration priority of each unit calibration item is determined according to the selected priority sequence with the highest division efficiency.

[0027] In an embodiment, the selected area corresponding to the power grid area is determined step by step according to the calibration priority of the unit calibration item and the area information, comprising: The calibration priority is identified, and the unit calibration item is marked as i according to the calibration priority, i=1, 2, …, n, n is the number of calibration priorities, for example, if the calibration priority is from 1 to 10, then n is 10; The calibration item standard corresponding to the unit calibration item i=1 is identified according to the unit area standard, the area information is calibrated according to the calibration item standard, and the selected area that meets the calibration item standard is determined for the current power grid area; The calibration item standard corresponding to the unit calibration item i=2 is identified according to the unit area standard, and the selected area is calibrated according to the calibration item standard and the area information, and the selected area that does not meet the calibration item standard is removed; The calibration item standard corresponding to the unit calibration item i=3 is identified according to the unit area standard, and the selected area is calibrated according to the calibration item standard and the area information, and the selected area that does not meet the calibration item standard is removed; In this way, until the calibration item standard corresponding to the unit calibration item i=n is identified according to the unit area standard, the selected area is calibrated according to the calibration item standard and the area information, and the selected area that meets the calibration item standard is marked as the unit area.

[0028] In an embodiment, the judgment of whether the corresponding selected area meets the calibration item standard and the calibration of the area information according to the calibration item standard can be realized by using existing judgment and calibration methods, such as commonly used machine learning and deep learning algorithms to establish corresponding intelligent models for intelligent calibration.

[0029] In an embodiment, the method for judging whether the corresponding selected area meets the calibration item standard comprises: The feature of the unit calibration item is extracted according to the area information, and the area information of the selected area and the calibration item standard is extracted and marked as selected area information; The unit calibration model is established, and the expression of the unit calibration model is: ; In the formula, (s j , BZ j ) is the input data, s j represents the selected area information of the corresponding unit calibration item, j represents the corresponding unit calibration item, j=1, 2, …, m, m is the number of calibration priorities; BZ j represents the calibration item standard of the corresponding unit calibration item; sj → BZ j indicates that the corresponding candidate region information meets the calibration item standard; the output data is a single-item calibration value DP(s j , BZ j ), and the unit calibration value is 1 or 0; the corresponding historical region information is used to set the corresponding training data for training.

[0030] The candidate region information and the calibration item standard of the corresponding unit calibration item are analyzed by the unit calibration model to obtain the unit calibration value of the corresponding candidate region; When the unit calibration value is 1, the corresponding candidate region meets the calibration item standard; When the unit calibration value is 0, the corresponding candidate region does not meet the calibration item standard.

[0031] Step two: real-time acquisition of analysis result data based on probabilistic power flow calculation for analysis in the grid region, acquisition of corresponding standard result data according to the analysis result data, the standard result data being deterministic power flow or high-precision measured data; calibration of the analysis result data according to the standard result data, identification of the region position corresponding to the analysis result data that does not meet the calibration standard, marking as an abnormal position, if it cannot be accurately located to a specific position, it can be marked as a general region, or even the corresponding unit region as an abnormal position; if the 95% quantile of the probability distribution of the voltage of a certain node deviates from the measured value by more than 5%, it does not meet the calibration standard, and the calibration standard is set according to the accuracy requirement of the probabilistic power flow calculation; the abnormal position is marked in the region information map.

[0032] In one embodiment, the determination of the abnormal position can also be performed in the following manner, comprising: preset variable verification data, the variable verification data being set for the grid and the probabilistic power flow calculation, including input variable adjustment data and output result allowable range, that is, the corresponding input data is adjusted, and the result should be within the output result allowable range, otherwise it is an abnormal situation; for example, the input data is wind power output fluctuation rate and load standard deviation, and the output data is voltage out-of-limit probability, the input variable adjustment data is formed by changing the input data, and then the change of the output data is determined, and it is integrated into the output result allowable range within the expected range; for example, if the wind power output fluctuation rate increases by 10% and leads to an increase in voltage out-of-limit probability exceeding the expected threshold, it is an abnormal situation.

[0033] The variable verification data is summarized and a verification library is established according to whether it can be applied to the verification of the corresponding unit region, that is, it is classified and stored according to the corresponding unit region.

[0034] Real-time identification of analysis result data based on probability flow calculation for analysis in the grid area, classifying the analysis result data according to the unit area, forming the unit result data of the corresponding unit area; identifying the corresponding input data according to the unit result data, and marking it as unit input data; matching the corresponding variable verification data from the verification library according to the unit input data, that is, matching in combination with the input variable adjustment data, one is to directly match the unit input data, and two is that the corresponding adjustment range is within the allowed range, updating the corresponding output result allowed range according to the corresponding adjustment range, and ensuring that the matched variable verification data is for the unit input data of the unit area; Identifying the input variable adjustment data and the output result allowed range corresponding to the variable verification data, simulating and analyzing according to the input variable adjustment data and the unit input data to obtain the simulation analysis result data of the unit area; Calibrating the simulation analysis result data and the unit result data according to the output result allowed range to determine whether the corresponding unit area is abnormal; When judging that the unit area is normal, no corresponding operation is performed; When judging that the unit area is abnormal, determining the corresponding abnormal position according to the difference between the simulation analysis result data and the unit result data.

[0035] Step three: identifying the unit area corresponding to the abnormal position in the regional information map, marking the unit area as a trace area; determining the abnormal reason of the probability flow calculation based on the trace area, and optimizing and adjusting the probability flow calculation based on the abnormal reason.

[0036] In one embodiment, the specific abnormal reason can be determined in combination with existing reason analysis techniques, such as matching the reason with corresponding historical abnormal data; the abnormal reason can also be determined by artificial means; when the abnormal reason is clear, the specific optimization and adjustment is to adjust the corresponding abnormal reason by artificial means and existing methods.

[0037] In one embodiment, the present application provides a probability flow calculation accuracy evaluation device for a power grid, comprising: A memory for storing a computer program; A processor for executing the computer program to realize the steps of the above-mentioned embodiment of the method for evaluating the accuracy of the probability flow calculation for the power grid.

[0038] The evaluation device provided in the embodiment can include but is not limited to a smartphone, a tablet computer, a notebook computer or a desktop computer, etc.

[0039] The above formulas are dimensionless values calculated by removing dimensions, the formulas are obtained by collecting a large amount of data to simulate software to obtain a formula closest to the actual situation, and the preset parameters and the preset threshold in the formula are set by a person skilled in the art according to the actual situation or obtained by a large amount of data simulation.

[0040] The above embodiments are only used to illustrate the technical method of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A method for evaluating the accuracy of probabilistic power flow calculations for a power grid, characterized in that: include: Step 1: Identify the grid area corresponding to the probabilistic power flow calculation, divide the grid area into several unit areas based on the probabilistic power flow calculation, and generate a regional information map based on the unit areas and grid areas. Step 2: Real-time acquisition of analysis result data based on probabilistic power flow calculations in the power grid area, obtaining corresponding standard result data based on the analysis result data, calibrating the analysis result data based on the standard result data, identifying abnormal locations corresponding to the analysis result data that do not meet the calibration standards; and marking the abnormal locations in the regional information map; Step 3: Identify the unit area corresponding to the abnormal position in the regional information map, and mark the unit area as the tracing area; determine the abnormal cause of the probabilistic power flow calculation based on the tracing area, and optimize and adjust the probabilistic power flow calculation based on the abnormal cause.

2. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 1, characterized in that: Methods for segmenting power grid areas using probabilistic power flow calculations include: Set the unit area standard and set the corresponding unit calibration items according to the unit area standard; The regional information of the power grid area is collected, the calibration priority of each unit calibration item is determined, the power grid area is divided according to the calibration priority of the unit calibration item and the regional information, and the corresponding unit area is obtained.

3. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 2, characterized in that: Methods for determining the calibration priority of each unit calibration item include: Preset the power grid simulation area; identify the unit calibration items, arrange the unit calibration items according to different priority orders, and obtain several priority sequences to be selected; Performing simulation division of the power grid simulation area according to the candidate priority sequence to obtain the division efficiency corresponding to the candidate priority sequence; The candidate priority sequence with the highest division efficiency is selected as the target priority sequence, and the calibration priority order of each unit calibration item is determined according to the target priority sequence.

4. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 3, characterized in that: Methods for dividing power grid areas according to the calibration priority order of unit calibration items and regional information include: Identify the calibration priority, and mark the unit calibration items as i according to the calibration priority, where i=1, 2, ..., n, and n is the number of calibration priorities; Identify the calibration item standard corresponding to the unit calibration item with i=1 according to the unit area standard, calibrate the area information according to the calibration item standard, and obtain a candidate area in the current power grid area that meets the calibration item standard; Identify the calibration item standard corresponding to the unit calibration item with i=2 according to the unit area standard, calibrate the candidate area according to the calibration item standard and area information, and eliminate the candidate area that does not meet the calibration item standard; Identify the calibration item standard corresponding to the unit calibration item with i=3 according to the unit area standard, calibrate the candidate area according to the calibration item standard and area information, and eliminate the candidate area that does not meet the calibration item standard; The process is deduced in this way until the calibration item standard corresponding to the unit calibration item i=n is identified according to the unit area standard, the selected area is calibrated according to the calibration item standard and area information, and the selected area that meets the calibration item standard is marked as a unit area.

5. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 4, characterized in that: Methods for determining whether the corresponding candidate area meets the calibration item standard include: Extract features of the region information according to the unit calibration item to obtain the candidate region information of the corresponding candidate region; A unit calibration model is established, and the candidate area information and calibration item standards corresponding to the corresponding unit calibration items are analyzed through the unit calibration model to obtain the calibration judgment results of the corresponding candidate areas. The calibration judgment results include whether the calibration item standards are met or not met.

6. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 5, characterized in that: The expression of the unit calibration model is: ； Where: (s j , BZ j ) is the input data, s j Indicates the candidate area information of the corresponding unit calibration item, j represents the corresponding unit calibration item, j=1, 2, ..., m, m is the number of calibration priority; BZ j Indicates the calibration item standard of the corresponding unit calibration item; s j →BZ j Indicates that the corresponding candidate area information meets the calibration item standard; the output data is the single calibration value DP (s j , BZ j ), the unit calibration value is 1 or 0; When the unit calibration value is 1, the corresponding selected area meets the calibration item standard; When the unit calibration value is 0, the corresponding candidate area does not meet the calibration item standard.

7. A method for evaluating the accuracy of probabilistic power flow calculation for a power grid according to claim 1, characterized in that: Another method of anomaly location determination includes: Preset variable verification data, which includes input variable adjustment data and output result allowable range; summarize the variable verification data according to whether it can be applied to the verification of the corresponding unit area to establish a verification library; Real-time identification of analysis result data of the power grid area based on probabilistic power flow calculation, classification of the analysis result data by unit area, forming unit result data of the corresponding unit area; identification of unit input data of the corresponding unit area based on the unit result data; Matching corresponding variable verification data from a verification library according to the unit input data, identifying input variable adjustment data and an output result allowable range corresponding to the variable verification data, performing simulation analysis based on the input variable adjustment data and the unit input data, and obtaining simulation analysis result data of the unit area; Calibrate the simulation analysis result data and unit result data according to the allowable range of the output results to determine whether the corresponding unit area has analysis anomalies; When the unit area analysis is judged to be normal, no corresponding operation is performed; When it is determined that the unit area analysis is abnormal, the corresponding abnormal location is determined based on the simulation analysis result data and the unit result data.

8. A device for evaluating the accuracy of probabilistic power flow calculations in a power grid, characterized in that: include: memory for storing computer programs; A processor is configured to implement a method for evaluating the accuracy of probabilistic power flow calculation for a power grid as claimed in any one of claims 1 to 7 when executing the computer program.

Images