CN-122022573-A - Power grid space data analysis result evaluation method, device and equipment

CN122022573ACN 122022573 ACN122022573 ACN 122022573ACN-122022573-A

Abstract

The invention provides a power grid space data analysis result evaluation method, a device and equipment, wherein the method comprises the steps of obtaining a target power grid space data analysis result set; according to the target power grid space data analysis result set, single indexes in a power grid space data analysis result evaluation index system are quantized to obtain quantized score vectors, the quantized score vectors are randomly sampled to obtain virtual samples, anti-fact conflict injection is conducted on the virtual samples to obtain conflict enhancement samples, causal effect processing is conducted on the virtual samples and the conflict enhancement samples to obtain average causal effect intensity vectors, initial weights are linearly processed according to the average causal effect intensity vectors to obtain optimal weight vectors, and according to the quantized score vectors and the optimal weight vectors, power grid space data analysis result evaluation grades are obtained. The method and the system can improve the efficiency, precision, scientificity and robustness of the power grid space data analysis result evaluation.

Inventors

GAO DENGJUN
GAO YUAN
Bi Dongdan
SUN YANBEI
CUI LITING
YANG DONGMING
LI JUN
LIU JIANGLONG
WANG XINYU
GUO MENG

Assignees

内蒙古电力勘测设计院有限责任公司

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (10)

1. The utility model provides a power grid space data analysis result evaluation method which is characterized by comprising the following steps: Acquiring a target power grid space data analysis result set; According to the target power grid space data analysis result set, carrying out quantization calculation on single indexes in a power grid space data analysis result evaluation index system to obtain a quantization score vector; Randomly sampling the quantized score vector to obtain a virtual sample; performing inverse fact conflict injection on the virtual sample to obtain a conflict enhancement sample; carrying out causal effect processing on the virtual samples and the conflict enhancement samples to obtain an average causal effect intensity vector; according to the average causal effect intensity vector, carrying out linear processing on the initial weight to obtain an optimal weight vector; And obtaining the evaluation grade of the analysis result of the power grid space data according to the quantized score vector and the optimal weight vector.
2. The grid-space-data-analysis-result evaluation method according to claim 1, wherein the acquiring the target grid-space-data-analysis-result set includes: acquiring an initial power grid space data analysis result set; cleaning and de-duplication processing is carried out on the initial power grid space data analysis result set to obtain a first intermediate result; And carrying out standardization processing on the first intermediate result to obtain a target power grid space data analysis result set.
3. The grid space data analysis result evaluation method according to claim 1, wherein performing quantization calculation on a single index in a grid space data analysis result evaluation index system according to the target grid space data analysis result set to obtain a quantized score vector comprises: evaluating an index system according to the analysis result of the power grid space data to obtain a dimensionality quantization model; And according to the dimensionality quantization model, analyzing a result set by utilizing the space data of the target power grid to obtain a quantization score vector.
4. The grid space data analysis result evaluation method according to claim 1, wherein randomly sampling the quantized score vector to obtain a virtual sample comprises: layering according to the quantized score interval to obtain at least two layers; and according to the hierarchy, layering the quantized score vector, and randomly sampling without returning to obtain a virtual sample.
5. The grid space data analysis result evaluation method according to claim 1, wherein performing inverse fact conflict injection on the virtual sample to obtain a conflict enhancement sample comprises: setting conflict types and conflict intensities for the virtual samples to obtain intermediate samples; And performing inverse fact conflict injection on the intermediate samples to obtain conflict enhancement samples.
6. The grid space data analysis result evaluation method according to claim 1, wherein performing causal effect processing on the virtual samples and the conflict enhancement samples to obtain an average causal effect intensity vector comprises: Performing reliability prediction according to the virtual samples and the conflict enhancement samples to obtain a reference prediction result and an intervention prediction result; Obtaining an individual causal effect according to the reference prediction result and the intervention prediction result; and carrying out weighted average on the individual causal effect to obtain an average causal effect intensity vector.
7. The grid space data analysis result evaluation method according to claim 1, wherein the linear processing of the initial weights according to the average causal effect intensity vector to obtain an optimal weight vector comprises: Index sorting is carried out on the average causal effect intensity vector to obtain an allocation adjustment coefficient; According to the distribution adjustment coefficient, carrying out linear processing on the initial weight to obtain an intermediate weight vector; and carrying out normalization processing on the intermediate weight vector to obtain an optimal weight vector.
8. A grid space data analysis result evaluation device, characterized by comprising: The acquisition module is used for acquiring a target power grid space data analysis result set; The processing module is used for carrying out quantization calculation on single indexes in the power grid space data analysis result evaluation index system according to the target power grid space data analysis result set to obtain a quantization score vector, carrying out random sampling on the quantization score vector to obtain a virtual sample, carrying out anti-fact conflict injection on the virtual sample to obtain a conflict enhancement sample, carrying out causal effect processing on the virtual sample and the conflict enhancement sample to obtain an average causal effect intensity vector, carrying out linear processing on initial weights according to the average causal effect intensity vector to obtain an optimal weight vector, and obtaining a power grid space data analysis result evaluation grade according to the quantization score vector and the optimal weight vector.
9. A computing device, comprising: one or more processors; Storage means for storing one or more programs which when executed by the one or more processors cause the one or more processors to implement the method of any of claims 1 to 7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a program which, when executed by a processor, implements the method according to any of claims 1 to 7.

Description

Power grid space data analysis result evaluation method, device and equipment Technical Field The present invention relates to the field of data processing technologies, and in particular, to a method, an apparatus, and a device for evaluating a result of analysis of spatial data of a power grid. Background With the deep advancement of digital transformation of the power grid, the scale and complexity of the space data (such as geographic coordinate information, equipment attributes, topological connection relations and the like) of the power grid are increased sharply, and the data analysis result becomes a key decision basis of core services such as power grid planning, operation and maintenance scheduling and the like. In the prior art, the evaluation of the analysis result of the power grid space data is mostly dependent on expert experience or a simple statistical method, and a comprehensive score or grade is obtained after weighted calculation by constructing an index system and giving subjective or statistical correlation-based weight to each index. The scheme of the prior art has the following defects: The evaluation accuracy is insufficient, namely the existing method lacks consideration of the fusion characteristics of multi-source data, has a single index system, and is difficult to comprehensively cover the core quality dimensions such as the integrity, the accuracy and the consistency of the spatial data; The efficiency and the suitability are low, a unified evaluation resource management system is not available, an evaluation framework is required to be repeatedly built when facing different scenes (power grid planning, operation and maintenance and emergency), and the operation is tedious and time-consuming; the robustness is poor, abnormal conditions caused by data transmission interference, equipment errors and the like are not fully considered, the tolerance of data deviation and logic conflict is low, and the stability of an evaluation result is poor. Disclosure of Invention The invention aims to provide a power grid space data analysis result evaluation method, a device and equipment. The efficiency, the precision, the scientificity and the robustness of the power grid space data analysis result evaluation can be improved. In order to solve the technical problems, the technical scheme of the invention is as follows: A power grid space data analysis result evaluation method comprises the following steps: Acquiring a target power grid space data analysis result set; According to the target power grid space data analysis result set, carrying out quantization calculation on single indexes in a power grid space data analysis result evaluation index system to obtain a quantization score vector; Randomly sampling the quantized score vector to obtain a virtual sample; performing inverse fact conflict injection on the virtual sample to obtain a conflict enhancement sample; carrying out causal effect processing on the virtual samples and the conflict enhancement samples to obtain an average causal effect intensity vector; according to the average causal effect intensity vector, carrying out linear processing on the initial weight to obtain an optimal weight vector; And obtaining the evaluation grade of the analysis result of the power grid space data according to the quantized score vector and the optimal weight vector. Optionally, the obtaining the target grid space data analysis result set includes: acquiring an initial power grid space data analysis result set; cleaning and de-duplication processing is carried out on the initial power grid space data analysis result set to obtain a first intermediate result; And carrying out standardization processing on the first intermediate result to obtain a target power grid space data analysis result set. Optionally, according to the target power grid space data analysis result set, performing quantization calculation on a single index in a power grid space data analysis result evaluation index system to obtain a quantized score vector, including: evaluating an index system according to the analysis result of the power grid space data to obtain a dimensionality quantization model; And according to the dimensionality quantization model, analyzing a result set by utilizing the space data of the target power grid to obtain a quantization score vector. Optionally, performing random sampling on the quantized score vector to obtain a virtual sample, including: layering according to the quantized score interval to obtain at least two layers; and according to the hierarchy, layering the quantized score vector, and randomly sampling without returning to obtain a virtual sample. Optionally, performing inverse fact conflict injection on the virtual sample to obtain a conflict enhanced sample, including: setting conflict types and conflict intensities for the virtual samples to obtain intermediate samples; And performing inverse fact conflict injection on the interme