Power grid project risk assessment method, system and equipment based on artificial intelligence, and medium

Abstract

The invention relates to the field of electronic digital data processing, in particular to a power grid project risk assessment method, system and device based on artificial intelligence and a medium, and the method comprises the steps: calculating the distribution consistency of meteorological data of a region where a power grid project is located; the meteorological data are multi-dimensional data points including at least two dimensions of a temperature dimension, a humidity dimension and a wind speed dimension; taking any meteorological data as an independent clustering cluster, for any two clustering clusters, acquiring a mean value of distribution consistency of all the meteorological data in the clustering clusters, merging the clustering clusters with the mean value greater than a preset consistency threshold value, and repeating until all the clustering clusters cannot be merged, so as to obtain a plurality of optimal clustering clusters; and performing risk assessment on the meteorological data according to the optimal clustering cluster. The method has the effect of improving the accuracy of the power grid environment data anomaly detection result.

Description

Technical Field

[0001] This application relates to the field of electronic digital data processing, and particularly to an artificial intelligence-based power grid project risk assessment method, system, device, and medium. Background Art

[0002] During the risk assessment process of power grid projects, uncontrollable natural factors, such as climate change and natural disasters, are often regarded as key environmental risks. These factors are not only potential threats during the construction and operation of the power grid, but may also have a serious impact on the safety, reliability, and long-term stability of the power grid. Specifically, climate change and natural disasters may cause damage to power equipment, line interruptions, grid overloads, etc. In severe cases, it may result in long-term outages, energy supply interruptions, and economic losses.

[0003] For example, extreme changes in climate warming or cold weather may cause power equipment (such as transformers, switches, wires, etc.) to operate overloaded, shortening their service life and even causing equipment failures; high-temperature weather may cause transmission lines to deform due to thermal expansion, thereby increasing the risk of short circuits and power outages, while extremely cold weather may cause the lines to freeze, increasing the burden on the lines and even causing breaks. Therefore, it is necessary to analyze environmental data to achieve risk assessment of power grid projects.

[0004] The density peak clustering algorithm is an unsupervised learning algorithm for clustering based on the density distribution of data points. This algorithm automatically identifies points with higher density as clustering centers by calculating the local density and relative distance of data points, and clusters the data based on these centers.

[0005] However, for power grid projects, due to the low probability of extreme weather occurrences, if extreme weather occurs, the density of data points corresponding to extreme weather is small. During the clustering process, the data points corresponding to extreme weather will be annexed by the clustering cluster where normal weather is located, resulting in inaccurate density clustering results and further inaccurate anomaly detection results. Summary of the Invention

[0006] To solve the problem of low accuracy of anomaly detection results for power grid environmental data, this application provides an artificial intelligence-based power grid project risk assessment method, system, device, and medium.

[0007] In a first aspect, this application provides an artificial intelligence-based power grid project risk assessment method, adopting the following technical solution:

[0008] The artificial intelligence-based power grid project risk assessment method includes:

[0009] Calculate the distribution consistency of meteorological data in the area where the power grid project is located; the meteorological data is multi-dimensional data points including at least two dimensions among the temperature dimension, humidity dimension, and wind speed dimension;

[0010] The calculation formula for distribution consistency is:

[0011]

[0012] In the formula, represents the distribution consistency between meteorological data i1 and meteorological data i2; sort the K2 meteorological data with the closest Euclidean distance to meteorological data i1 to construct a sequence to obtain a neighboring sequence, calculate the differences between each meteorological data in the neighboring sequence and meteorological data i1 in each dimension, and construct a neighboring vector sequence of meteorological data i1 according to the differences; similarly, obtain the neighboring vector sequence of meteorological data i2; represents the m-th vector in the neighboring vector sequence of meteorological data i1; represents the m-th vector in the neighboring vector sequence of meteorological data i2; cos( ) represents the cosine similarity of vectors; represents the similarity between meteorological data i1 and meteorological data i2; norm( ) represents the standard normalization function;

[0013] Take any meteorological data as a separate clustering cluster. For any two clustering clusters, obtain the mean value of the distribution consistency of all meteorological data in the clustering clusters, merge the clustering clusters with the mean value greater than the preset consistency threshold, and repeat until all clustering clusters cannot be merged to obtain multiple optimal clustering clusters;

[0014] Conduct risk assessment on meteorological data according to the optimal clustering clusters.

[0015] The beneficial effects are as follows: By clustering the meteorological data, the data within each clustering cluster is made to have consistency in distribution, and by gradually merging the clustering clusters with the mean value greater than the preset consistency threshold, the clustering results can be dynamically adjusted to ensure that the finally obtained optimal clustering clusters have high internal consistency and low external differences. This gradually optimized process improves the efficiency and accuracy of risk assessment.

[0016] In the calculation formula for distribution consistency, represents the cosine similarity of the corresponding vectors in the neighboring vector sequences of meteorological data i1 and meteorological data i2. The larger its value, the more similar the distribution of the data points around the two; therefore, if has a larger value and The larger the value, the higher the distribution consistency of the two meteorological data. Conversely, the distribution consistency is low. The application of the standard normalization function ensures the standardization of the distribution consistency value, making the results comparable under different datasets or different calculation conditions. This helps to conduct unified risk assessments between different projects and regions, improving the reliability and consistency of the assessment results. By comprehensively analyzing multi-dimensional data for risk assessment, the meteorological conditions in the area where the power grid project is located can be more comprehensively reflected. Compared with single-dimensional data analysis, multi-dimensional data fusion can more accurately identify meteorological factors that potentially affect the power grid project.

[0017] Optionally, the similarity calculation formula is:

[0018]

[0019] In the formula, represents the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; represents the collection time of meteorological data i1; represents the collection time of meteorological data i2; norm( ) represents the standard normalization function.

[0020] The beneficial effect is that by comprehensively considering the local density and collection time of meteorological data, this similarity calculation formula can effectively evaluate the similarity of two meteorological data in space and time, and through normalization processing, the similarity is made standardized and comparable.

[0021] Optionally, the similarity calculation formula is:

[0022]

[0023] In the formula, represents the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; norm() represents the standard normalization function.

[0024] The beneficial effect is that it provides a similarity calculation method with relatively less computational effort to quantify the similarity degree between meteorological data.

[0025] Optionally, the local density calculation formula is:

[0026]

[0027] In the formula, ρ iRepresents the local density of meteorological data i; set the neighborhood of meteorological data i, and K1 represents the total number of data points within the neighborhood of meteorological data i; X i，j Represents the data value of the j-th dimension of meteorological data i; n represents the total number of dimensions of meteorological data; x a，j Represents the data value of the j-th dimension of the a-th data point within the neighborhood of meteorological data i; σ j Represents the standard deviation of the data values of meteorological data i and the data points within its neighborhood in the j-th dimension; Similarly to calculating the local density of meteorological data i, calculate the local density of meteorological data i1 and the local density of meteorological data i2.

[0028] The beneficial effect is: The factor Represents the distance between meteorological data i and the a-th data point within the neighborhood. The smaller the value of this factor, the greater the local density of the meteorological data. Conversely, the smaller the local density of this meteorological data. In, by dividing by the standard deviation σ j , the influence brought by different dimensions and different units is avoided, making the calculation result of the local density of meteorological data i more accurate.

[0029] Optionally, conduct risk assessment on meteorological data according to the optimal clustering clusters, including:

[0030] Calculate the abnormality degree of the optimal clustering cluster, regard the meteorological data in the clustering cluster with an abnormality degree greater than the preset threshold as abnormal data, and use the ratio of the abnormal data to the total number of meteorological data as the risk assessment result.

[0031] Optionally, the calculation formula for the abnormality degree is:

[0032] Y b = exp(-N b ) × norm(σ(b));

[0033] In the formula, Y b Represents the abnormality degree of the b-th clustering cluster; N b Represents the total number of meteorological data in the b-th clustering cluster; σ(b) represents the mean value of the standard deviations of all meteorological data in the b-th clustering cluster in each dimension; exp() represents the exponential function with base e; norm() represents the standard normalization function.

[0034] The beneficial effect is: It provides a method for calculating the abnormality degree. This formula comprehensively considers two factors: the amount of data and the data difference, and can more comprehensively evaluate the abnormality degree of the clustering cluster. It is applicable to scenarios that require high-precision risk assessment.

[0035] Optionally, the calculation formula for the abnormality degree is

[0036] Y b= norm(σ(b));

[0037] In the formula, Y b represents the degree of abnormality of the b-th clustering cluster; σ(b) represents the mean standard deviation of all meteorological data in each dimension in the b-th clustering cluster; norm() represents the standard normalization function.

[0038] The beneficial effect is: providing another method for calculating the degree of abnormality, which is applicable to scenarios that require rapid evaluation and have low requirements for accuracy.

[0039] In the second aspect, the present application provides an artificial intelligence-based power grid project risk assessment system, adopting the following technical solutions:

[0040] The artificial intelligence-based power grid project risk assessment system includes:

[0041] A distribution consistency calculation module, which is used to calculate the distribution consistency of meteorological data in the area where the power grid project is located; the meteorological data is multi-dimensional data points including at least two dimensions of a temperature dimension, a humidity dimension, and a wind speed dimension;

[0042] The calculation formula for distribution consistency is:

[0043]

[0044] In the formula, represents the distribution consistency between meteorological data i1 and meteorological data i2. After sorting the K2 meteorological data with the closest Euclidean distance to meteorological data i1 to construct a sequence to obtain an adjacent sequence, calculate the difference between each meteorological data in the adjacent sequence and meteorological data i1 in each dimension, and construct an adjacent vector sequence of meteorological data i1 according to the difference; similarly, obtain an adjacent vector sequence of meteorological data i2; represents the m-th vector in the adjacent vector sequence of meteorological data i1; represents the m-th vector in the adjacent vector sequence of meteorological data i2; cos() represents the cosine similarity of vectors; represents the similarity between meteorological data i1 and meteorological data i2; norm() represents the standard normalization function;

[0045] An optimal clustering cluster acquisition module, which is used to take any meteorological data as a separate clustering cluster. For any two clustering clusters, obtain the mean value of the distribution consistency of all meteorological data in the clustering clusters, merge the clustering clusters with the mean value greater than the preset consistency threshold, and repeat until all clustering clusters cannot be merged to obtain multiple optimal clustering clusters;

[0046] A risk assessment module, which is used to perform risk assessment on meteorological data according to the optimal clustering clusters.

[0047] In a third aspect, the present application provides an artificial intelligence-based power grid project risk assessment device, adopting the following technical solution:

[0048] The artificial intelligence-based power grid project risk assessment device includes a memory and a processor;

[0049] The memory is used to store computer program code and transmit the computer program code to the processor;

[0050] The processor is used to execute the above-mentioned artificial intelligence-based power grid project risk assessment method according to the instructions in the computer program code.

[0051] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution:

[0052] A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the above-mentioned artificial intelligence-based power grid project risk assessment method.

[0053] The beneficial effect is that the above-mentioned artificial intelligence-based power grid project risk assessment method is generated into a computer program and stored in the memory to be loaded and executed by the processor. Thus, a system is made according to the memory and the processor, which is convenient to use.

[0054] The present application has the following technical effects:

[0055] 1. By clustering the meteorological data, the data within each clustering cluster is made consistent in distribution. And by gradually merging the clustering clusters with a mean greater than a preset consistency threshold, the clustering result can be dynamically adjusted to ensure that the finally obtained optimal clustering cluster has high internal consistency and low external difference. This gradually optimized process improves the efficiency and accuracy of risk assessment.

[0056] 2. By comprehensively conducting risk assessment with multi-dimensional data, it can more comprehensively reflect the meteorological conditions in the area where the power grid project is located. Compared with single-dimensional data analysis, multi-dimensional data fusion can more accurately identify the meteorological factors that have potential impacts on the power grid project. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] By referring to the accompanying drawings and reading the following detailed description, the above and other purposes, features, and advantages of the exemplary embodiments of the present application will become easily understandable. In the drawings, several embodiments of the present application are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0058] Figure 1 It is a flowchart of the artificial intelligence-based power grid project risk assessment method according to an embodiment of the present application.

[0059] Figure 2 is the structural block diagram of the risk assessment system for the artificial intelligence power grid project according to the embodiments of the present application.

[0060] Figure 3 is the structural block diagram of the risk assessment device for the artificial intelligence power grid project according to the embodiments of the present application. Specific Embodiments

[0061] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0062] It should be understood that when the claims, specifications, and drawings of the present application use terms such as "first" and "second", they are only used to distinguish different objects and not to describe a specific order. The terms "including" and "comprising" used in the specifications and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0063] The embodiments of the present application disclose a risk assessment method for an artificial intelligence power grid project. Refer to Figure 1 and include steps S1 - S3, specifically as follows:

[0064] S1: Calculate the distribution consistency of the meteorological data in the area where the power grid project is located.

[0065] The meteorological data includes at least two dimensions among the temperature dimension, humidity dimension, and wind speed dimension, and the meteorological data is a multi-dimensional data point.

[0066] In one embodiment, the meteorological data of the location of the power grid project is monitored by arranging meteorological sensors (such as temperature sensors, humidity sensors, anemometers, etc.). Among them, these sensors can be deployed at key nodes of the power grid (such as substations, along transmission lines, etc.). The preset acquisition frequency is 0.2 Hz (i.e., once every 5 S), and the data collected at the same moment is used as a multi-dimensional meteorological data point (temperature, humidity, wind speed). Meteorological conditions (such as temperature, humidity, wind speed, etc.) affect the working load and durability of power grid equipment (such as transformers, wires, circuit breakers, etc.). For example, high temperature may cause equipment overheating, and severe cold weather may affect the freezing or damage of power facilities. Therefore, by monitoring meteorological data, abnormal events can be detected in a timely manner, thereby evaluating the potential risks to power grid facilities.

[0067] This application will be described by taking the analysis of a single data collection point as an example, that is, comparing meteorological data at the same location at different times. Details will not be elaborated hereinafter.

[0068] By calculating the local density difference between any meteorological data and any meteorological data in its vicinity (here, the vicinity refers to multiple adjacent times), the similarity between the two is calculated; then, based on the similarity and distribution of this data point and its neighboring data points, the distribution consistency between the two is calculated; clustering clusters are divided according to the distribution consistency between meteorological data to complete clustering.

[0069] For any multi-dimensional meteorological data, for data points at adjacent times, if the fluctuations of the meteorological data in this area are large, it indicates that the risk of power grid projects in this area is greater. By measuring the local density of meteorological data, it can help analyze the fluctuation of meteorological data in this area. Therefore, this application analyzes the local density of any meteorological data.

[0070] Exemplarily, set the number of neighboring data points to 10. For any meteorological data, obtain the 10 data points with the closest Euclidean distance to this meteorological data, and calculate the local density of this meteorological data according to the distance difference between this meteorological data point and its neighboring data points.

[0071] In one embodiment, the calculation formula for local density is:

[0072]

[0073] In the formula, ρ i represents the local density of meteorological data i; set the neighborhood of meteorological data i, K1 represents the total number of data points within the neighborhood of meteorological data i; X i，j represents the data value of the j-th dimension of meteorological data i; n represents the total number of dimensions of meteorological data; x a，j represents the data value of the j-th dimension of the a-th data point within the neighborhood of meteorological data i; σ j represents the standard deviation of the data values of meteorological data i and the data points within its neighborhood in the j-th dimension.

[0074] Among them, since the factor represents the distance between meteorological data i and the a-th data point within the neighborhood, the smaller the value of this factor, the greater the local density of the meteorological data. Conversely, the local density of this meteorological data is smaller; in j , by dividing by the standard deviation σ, the influence brought by different dimensions and different dimensions is avoided, making the calculation result of the local density of meteorological data i more accurate.

[0075] Similarly to calculating the local density of meteorological data i, calculate the local density of meteorological data i1 and the local density of meteorological data i2.

[0076] The working principle of the density peak clustering algorithm is as follows: data points are assigned to the clustering cluster with a greater density that is closest to them. For meteorological data, if there are extreme weather data, when using the density peak clustering algorithm to cluster meteorological data, the number of clustering clusters where they are located is small and the density is low. When using density peak clustering, it will cause misclassification of the clustering cluster where the extreme weather data is located, that is, the clustering cluster where the extreme weather data is located will be annexed by the clustering cluster where the normal weather data is located.

[0077] In order to reduce the above misclassification situation, the present application calculates the similarity between two meteorological data according to the local density of the meteorological data and the data around it. At the same time, considering the distribution of other data points around these two meteorological data, the distribution consistency between the two is calculated by calculating the distribution of other data points around them.

[0078] In one embodiment, the calculation formula for similarity is:

[0079]

[0080] In the formula, represents the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; represents the acquisition time of meteorological data i1; represents the acquisition time of meteorological data i2; norm() represents the standard normalization function.

[0081] In order to reduce the calculation amount, the calculation formula for similarity can also be:

[0082]

[0083] In the formula, represents the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; norm() represents the standard normalization function.

[0084] After obtaining the similarity between meteorological data, the distribution consistency of meteorological data is calculated according to the similarity. In one embodiment, the calculation formula for distribution consistency is:

[0085]

[0086] In the formula, represents the distribution consistency between meteorological data i1 and meteorological data i2; It represents the m-th vector in the neighboring vector sequence of meteorological data i1; It represents the m-th vector in the neighboring vector sequence of meteorological data i2; cos() represents the cosine similarity of vectors; It represents the similarity between meteorological data i1 and meteorological data i2; norm() represents the standard normalization function.

[0087] Among them, the construction method of the neighboring vector sequence is as follows: for any meteorological data, obtain the K2 meteorological data with the closest Euclidean distance to it, and preset K2 = 5; arrange the K2 meteorological data with the closest Euclidean distance to this meteorological data in ascending order of Euclidean distance to obtain a neighboring sequence. After sorting the K2 meteorological data with the closest Euclidean distance to meteorological data i1, construct a sequence to obtain a neighboring sequence, calculate the difference between each meteorological data in the neighboring sequence and meteorological data i1 in each dimension, and construct the neighboring vector sequence of meteorological data i1 according to the difference. Similarly, obtain the neighboring vector sequence of meteorological data i2.

[0088] In the calculation formula of distribution consistency, It represents the cosine similarity of the corresponding vectors in the neighboring vector sequences of meteorological data i1 and meteorological data i2. The larger its value, the more similar the distribution of the data points around the two is; therefore, if the value is larger and the value is larger, it indicates that the distribution consistency of these two meteorological data is higher. On the contrary, the distribution consistency is low. The application of the standard normalization function ensures the standardization of the distribution consistency value, making the results comparable under different data sets or different calculation conditions. This helps to conduct unified risk assessments between different projects and regions, and improves the reliability and consistency of the assessment results.

[0089] S2: Take any meteorological data as a separate clustering cluster. For any two clustering clusters, obtain the mean value of the distribution consistency of all meteorological data in the clustering clusters, and merge the clustering clusters with the mean value greater than the preset consistency threshold. Repeat until all clustering clusters cannot be merged to obtain multiple optimal clustering clusters.

[0090] In one embodiment, first, take any meteorological data as a separate clustering cluster, set the consistency threshold to 0.7. For any two clustering clusters, obtain the mean value of the consistency of all meteorological data in these two clustering clusters. If the mean value is greater than the preset consistency threshold, then merge these two clustering clusters. Repeat the above steps until all clustering clusters cannot be merged.

[0091] Group the meteorological data through clustering so that the data within each cluster has consistency in distribution. By gradually merging clusters with a mean greater than a preset consistency threshold, the clustering results can be dynamically adjusted to ensure that the final optimal clusters have high internal consistency and low external differences. This process of gradual optimization improves the efficiency and accuracy of risk assessment.

[0092] S3: Conduct risk assessment on the meteorological data according to the optimal clusters.

[0093] Specifically, calculate the degree of abnormality of the optimal clusters. Take the meteorological data in the clusters with a degree of abnormality greater than the preset threshold as abnormal data, and take the ratio of the abnormal data to the total number of meteorological data as the risk assessment result.

[0094] The formula for calculating the degree of abnormality is:

[0095] Y b =exp(-N b )×norm(σ(b));

[0096] In the formula, Y b represents the degree of abnormality of the b-th cluster; N b represents the total number of meteorological data in the b-th cluster; σ(b) represents the mean standard deviation of all meteorological data in the b-th cluster in each dimension; exp() represents the exponential function with base e; norm() represents the standard normalization function.

[0097] In the above formula, the term exp(-N b ) makes the smaller the number N b of meteorological data in the cluster, the higher the degree of abnormality Y b , which reflects that clusters with less data are more likely to contain abnormal data. The term norm(σ(b)) represents the normalized value of the mean standard deviation of meteorological data in the cluster in each dimension. The larger the standard deviation, the greater the difference between data points, and the higher the degree of abnormality Y b . This formula comprehensively considers two factors: the amount of data and the data difference, and can more comprehensively evaluate the degree of abnormality of the clusters. It is applicable to scenarios that require high-precision risk assessment.

[0098] For any cluster, if the number of data in the cluster is less and the difference between data points in the cluster is greater, it means that the cluster is more abnormal.

[0099] The preset threshold is 0.7. Mark the clusters with a degree of abnormality greater than the preset abnormality threshold as abnormal clusters, and mark the meteorological data in the abnormal clusters as abnormal data points. Denote the ratio of the number of abnormal data points to the total number of meteorological data as the risk assessment result. The larger the ratio, the more abnormal data points and the higher the risk.

[0100] In other embodiments, the calculation formula for the degree of abnormality can also be

[0101] Y b = norm(σ(b));

[0102] In the formula, Y b represents the degree of abnormality of the b-th clustering cluster; σ(b) represents the mean value of the standard deviations of all meteorological data in the b-th clustering cluster in each dimension; norm() represents the standard normalization function. This embodiment is applicable to scenarios that require rapid evaluation and have low requirements for accuracy.

[0103] See Figure 2 , this application embodiment also discloses an artificial intelligence-based power grid project risk assessment system, which is applied to an artificial intelligence-based power grid project risk assessment method described in the above embodiment. The system includes:

[0104] A distribution consistency calculation module, configured to calculate the distribution consistency of meteorological data in the area where the power grid project is located; the meteorological data is multi-dimensional data points including at least two dimensions of a temperature dimension, a humidity dimension, and a wind speed dimension;

[0105] The calculation formula for the distribution consistency is:

[0106]

[0107] In the formula, represents the distribution consistency between meteorological data i1 and meteorological data i2. The K2 meteorological data with the closest Euclidean distance to meteorological data i1 are sorted to construct a sequence to obtain an adjacent sequence. Calculate the differences between each meteorological data in the adjacent sequence and meteorological data i1 in each dimension, and construct an adjacent vector sequence of meteorological data i1 according to the differences; similarly, obtain an adjacent vector sequence of meteorological data i2; represents the m-th vector in the adjacent vector sequence of meteorological data i1; represents the m-th vector in the adjacent vector sequence of meteorological data i2; cos() represents the cosine similarity of vectors; represents the similarity between meteorological data i1 and meteorological data i2; norm() represents the standard normalization function;

[0108] An optimal clustering cluster acquisition module, configured to use any meteorological data as a separate clustering cluster. For any two clustering clusters, obtain the mean value of the distribution consistency of all meteorological data in the clustering clusters, merge the clustering clusters with the mean value greater than the preset consistency threshold, and repeat until all clustering clusters cannot be merged, to obtain multiple optimal clustering clusters;

[0109] A risk assessment module, configured to perform risk assessment on meteorological data according to the optimal clustering clusters.

[0110] See Figure 3 , an embodiment of the present application also discloses an artificial intelligence-based power grid project risk assessment device, including a memory and a processor;

[0111] The memory is used to store computer program code and transmit the computer program code to the processor;

[0112] The processor is configured to execute the above-mentioned artificial intelligence-based power grid project risk assessment method according to the instructions in the computer program code.

[0113] An embodiment of the present application also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned artificial intelligence-based power grid project risk assessment method is implemented.

[0114] Generally speaking, the computer instructions for implementing the method of the present invention can be carried by any combination of one or more computer-readable storage media. A non-transitory computer-readable storage medium may include any computer-readable medium except for a signal propagating temporarily by itself.

[0115] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EKROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0116] Computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. In particular, the Python language suitable for neural network computing and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., connected through the Internet using an Internet service provider).

[0117] For the above-mentioned devices and non-transitory computer-readable storage media, reference may be made to the specific description of a method for risk assessment of an artificial intelligence power grid project and its beneficial effects, which will not be elaborated here.

[0118] Although this specification has shown and described multiple embodiments of the present application, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and idea of the present application. It should be understood that various alternative solutions to the embodiments of the present application described herein may be adopted in the practice of the present application.

[0119] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. The risk assessment method of power grid projects based on artificial intelligence is characterized by: include: Calculate the distribution consistency of meteorological data in the area where the power grid project is located; meteorological data is a multidimensional data point including at least two dimensions of temperature, humidity and wind speed; The calculation formula for distribution consistency is: In the formula, Indicates the distribution consistency of meteorological data i1 and meteorological data i2; sort the K2 meteorological data with the closest Euclidean distance to meteorological data i1 and construct a sequence to obtain a neighboring sequence, calculate the difference between each meteorological data in the neighboring sequence and meteorological data i1 in each dimension, and construct the neighboring vector sequence of meteorological data i1 according to the difference; similarly, obtain the neighboring vector sequence of meteorological data i2; Represents the mth vector in the neighboring vector sequence of meteorological data i1; represents the mth vector in the neighboring vector sequence of meteorological data i2; cos() represents the cosine similarity of the vector; Indicates the similarity between meteorological data i1 and meteorological data i2; norm() indicates the standard normalization function; Take any meteorological data as a separate cluster. For any two clusters, obtain the mean of the distribution consistency of all meteorological data in the clusters, merge the clusters whose means are greater than the preset consistency threshold, and repeat until all clusters cannot be merged, and obtain multiple optimal clusters. Risk assessment of meteorological data is performed based on the optimal clustering clusters.

2. The artificial intelligence-based power grid project risk assessment method according to claim 1 is characterized in that: The similarity calculation formula is: In the formula, Indicates the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; Indicates the collection time of meteorological data i1; represents the collection time of meteorological data i2; norm( ) represents the standard normalization function.

3. The artificial intelligence-based power grid project risk assessment method according to claim 1 is characterized in that: The similarity calculation formula is: In the formula, Indicates the similarity between meteorological data i1 and meteorological data i2; represents the local density of meteorological data i1; represents the local density of meteorological data i2; norm() represents the standard normalization function.

4. The artificial intelligence-based power grid project risk assessment method according to claim 2 or 3 is characterized in that: The local density is calculated as: In the formula, ρ i represents the local density of meteorological data i; sets the neighborhood of meteorological data i, K1 represents the total number of data points in the neighborhood of meteorological data i; X i，j represents the data value of the jth dimension of meteorological data i; n represents the total number of dimensions of meteorological data; x a，j represents the data value of the jth dimension of the ath data point in the neighborhood of meteorological data i; σ j Represents the standard deviation of the data values ​​of meteorological data i and the data points in the neighborhood of meteorological data i in the jth dimension; similarly to the calculation of the local density of meteorological data i, the local density of meteorological data i1 and the local density of meteorological data i2 are calculated.

5. The artificial intelligence-based power grid project risk assessment method according to claim 1 is characterized in that: Risk assessment of meteorological data based on optimal clustering, including: The abnormality degree of the optimal cluster is calculated, and the meteorological data in the cluster with an abnormality degree greater than the preset threshold is taken as abnormal data. The ratio of abnormal data to the total number of meteorological data is taken as the risk assessment result.

6. The artificial intelligence-based power grid project risk assessment method according to claim 5 is characterized in that: The calculation formula for the degree of abnormality is: Y b =exp(-N b )×norm(σ(b)); Where Y b Indicates the abnormality of the b-th cluster; N b represents the total number of meteorological data in the b-th cluster; σ(b) represents the mean standard deviation of all meteorological data in the b-th cluster in each dimension; exp() represents the exponential function with e as the base; norm() represents the standard normalization function.

7. The artificial intelligence-based power grid project risk assessment method according to claim 5 is characterized in that: The calculation formula for abnormality degree is: Y b =norm(σ(b)); Where Y b represents the abnormality of the b-th cluster; σ(b) represents the mean standard deviation of all meteorological data in the b-th cluster in each dimension; norm() represents the standard normalization function.

8. The artificial intelligence-based power grid project risk assessment system is characterized by: include: A distribution consistency calculation module is used to calculate the distribution consistency of meteorological data in the area where the power grid project is located; the meteorological data is a multidimensional data point including at least two dimensions of temperature, humidity and wind speed; The calculation formula for distribution consistency is: In the formula, Indicates the distribution consistency of meteorological data i1 and meteorological data i2. After sorting the K2 meteorological data with the closest Euclidean distance to meteorological data i1, a sequence is constructed to obtain a neighboring sequence. The difference between each meteorological data in the neighboring sequence and meteorological data i1 in each dimension is calculated, and the neighboring vector sequence of meteorological data i1 is constructed according to the difference. Similarly, the neighboring vector sequence of meteorological data i2 is obtained. Represents the mth vector in the neighboring vector sequence of meteorological data i1; represents the mth vector in the neighboring vector sequence of meteorological data i2; cos() represents the cosine similarity of the vector; Indicates the similarity between meteorological data i1 and meteorological data i2; norm() indicates the standard normalization function; The optimal cluster acquisition module is used to take any meteorological data as a separate cluster, and for any two clusters, obtain the mean of the distribution consistency of all meteorological data in the clusters, merge the clusters whose means are greater than a preset consistency threshold, and repeat until all clusters cannot be merged, so as to obtain multiple optimal clusters; The risk assessment module is used to perform risk assessment on meteorological data based on the optimal clustering clusters.

9. Artificial intelligence-based power grid project risk assessment equipment, characterized by: including memory and processor; The memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 7 according to instructions in the computer program code.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images