CN-121981584-A - Industrial power scenic index construction method, system, equipment and medium based on multidimensional data index

CN121981584ACN 121981584 ACN121981584 ACN 121981584ACN-121981584-A

Abstract

The invention discloses an industrial power scenic spot index construction method, system, equipment and medium based on multidimensional data indexes, which relate to the technical field of industrial power analysis and comprise the steps of obtaining a plurality of candidate indexes for representing industrial scenic spot changes; determining an economic cycle reference date serving as an index comparison reference, performing preprocessing operation on a history sequence of candidate indexes, dividing the indexes into a preceding index, a consistent index and a lagging index, generating a preceding diffusion index, a consistent diffusion index and a lagging diffusion index, weighting and summarizing various indexes according to a preset weight determination rule to respectively obtain a preceding synthesis index, a consistent synthesis index and a lagging synthesis index, and taking various diffusion indexes and various synthesis indexes as scenic assessment bases. The invention comprehensively utilizes the diffusion index method and the synthesis of three indexes of advance, coincidence and lag, not only can judge the economic period stage of the industry, but also can infer the future trend, and solves the problems of single model and insufficient accuracy in the prior art.

Inventors

YE XIAOCHEN
LIU YAN
ZHAO LIN
JI XING
CHENG MENGZENG
LIU GUANGSHUO
ZHAO JINGZHI

Assignees

国网辽宁省电力有限公司经济技术研究院

Dates

Publication Date: 20260505
Application Date: 20251125

Claims (10)

1. A method for constructing industrial power scenic index based on multidimensional data index is characterized by comprising the steps of, Acquiring a plurality of candidate indexes for representing industrial scenic spot changes from industrial power multidimensional data according to a preset index selection principle to form a scenic spot index pool; Analyzing the historical economic data, identifying the peak value and valley value positions of the economic cycle, and determining the economic cycle datum date serving as an index comparison reference; performing preprocessing operation on the historical sequence of each index in the scenic spot index pool to obtain a standardized processing sequence for scenic spot analysis; determining the degree of lead of each index relative to the economic cycle reference date based on the timing correlation between the reference index and other candidate indexes, thereby classifying the indexes into a leading index, a consistent index and a lagging index; The changing directions of the leading index group, the consistent index group and the lagging index group are respectively identified, and the leading diffusion index, the consistent diffusion index and the lagging diffusion index are generated according to the number proportion of the indexes in the expansion state in each group; Based on the change characteristics of various indexes, weighting and summarizing the various indexes according to a preset weight determining rule, and respectively obtaining a preceding synthesis index, a consistent synthesis index and a lagged synthesis index by combining a long-term trend adjusting mechanism; and taking various diffusion indexes and various synthesis indexes as scenic spot evaluation basis to form industrial electric scenic spot indexes for representing the industrial economic period stage, the operation trend and the scenic spot degree change characteristics.
2. The method for constructing an industrial power scenic spot index based on multi-dimensional data as set forth in claim 1, wherein said obtaining a plurality of candidate indicators for characterizing changes in industrial scenic spots comprises, Determining an overall data range from the multi-dimensional data as a candidate index source based on structural features of the industrial economic operation; In the total data range, comprehensively evaluating and screening various data items according to a preset index selection principle to form a candidate index set; And (3) performing scenic spot response characteristic analysis on the candidate index set, examining the expression rule of each data item in the history scenic spot change process, and finally determining candidate indexes for constructing a scenic spot index pool according to the representation capability of the industrial scenic spot change.
3. The method for constructing an industrial power scenic spot index based on a multi-dimensional data index according to claim 2, wherein said determining an economic cycle benchmark date as an index comparison reference comprises, Selecting a group of economic indexes reflecting the economic fluctuation, meeting the requirement of being consistent with the economic cycle fluctuation, and preliminarily determining the reference date of the economic fluctuation based on the economic indexes; And comparing the reference scenic spot circulation annual table with the preliminarily determined reference date, and finally selecting the economic cycle reference date.
4. The method of claim 3, wherein the step of performing a preprocessing operation on the historical sequence of each index in the pool of indices of the scenic spot comprises, Consistency checking is carried out on the historical sequences of all indexes in the scenic spot index pool, existing missing data are identified, and the missing values are complemented through a preset interpolation strategy; normalizing the completed history sequence; carrying out seasonal decomposition on the normalized historical sequence; trend decomposition is carried out on the history sequence after seasonal decomposition.
5. The method for constructing an industrial power scenic spot index based on a multi-dimensional data index according to claim 4, wherein said dividing the index into a leading index, a consistent index and a lagging index comprises, Comparing each candidate index in the foreground index pool with the historical change sequence of the reference index, observing the relative change mode in the time direction, and primarily judging the time response characteristic of the candidate index; Based on preliminary judgment, judging whether the trend is reflected before the economic activity changes or whether the trend is synchronously changed with the reference index or whether the corresponding trend is displayed after the reference index changes by identifying the response sequence of the candidate index to the trend change of the reference index in different time periods; Dividing the index with trend change before the turning point of the reference index into advance index, dividing the index with synchronous fluctuation rule on the change node with the reference index into consistent index, and dividing the index with trend change lagging behind the reference index into lagging index.
6. The method of claim 5, wherein generating the look-ahead diffusion index, the uniform diffusion index, and the late diffusion index comprises, Respectively counting the number of indexes of the leading index group, the consistent index group and the lagging index group in an expansion state at the same time point, and comparing the ratio of the expansion state to the total number of indexes in the groups to obtain expansion ratio information; and taking the expansion proportion information obtained through statistics as a depicting basis of the diffusion state to respectively form a preceding diffusion index, a consistent diffusion index and a lagging diffusion index.
7. The method of claim 6, wherein the step of obtaining the look-ahead synthesis index, the consistent synthesis index, and the lag synthesis index, respectively, Extracting change characteristics aiming at the leading index group, the consistent index group and the lagging index group to form standardized change expression; Weighting each index in the preceding index group, the consistent index group and the lagging index group according to a preset weight determination rule based on the standardized variation expression; And performing trend adjustment operation on the weighted results based on the trend of the economic cycle to form a preceding synthesis index, a consistent synthesis index and a lagging synthesis index respectively.
8. An industrial power scenic index construction system based on multi-dimensional data indexes, which is characterized by comprising an index pool generation module, a data processing module, an index classification module and an index generation module, wherein the industrial power scenic index construction method based on the multi-dimensional data indexes is applied to any one of claims 1-7; the index pool generation module acquires a plurality of candidate indexes for representing industrial scenic spot changes from industrial power multidimensional data according to a preset index selection principle to form a scenic spot index pool; The data processing module analyzes the historical economic data, identifies the peak value and the valley value positions of the economic cycle, determines the economic cycle datum date serving as an index comparison reference, and performs preprocessing operation on the historical sequence of each index in the scenic gas index pool to obtain a standardized processing sequence for scenic gas analysis; The index classification module judges the leading degree of each index relative to the economic cycle reference date based on the time sequence correlation between the reference index and other candidate indexes, and further classifies the indexes into a leading index, a consistent index and a lagging index; The index generation module respectively identifies the changing directions of the leading index group, the consistent index group and the lagging index group, generates the leading diffusion index, the consistent diffusion index and the lagging diffusion index according to the number proportion of the indexes in the expansion state in each group, carries out weighted summarization on each type of index according to a preset weight determination rule based on the changing characteristics of each type of index, and respectively obtains the leading synthesis index, the consistent synthesis index and the lagging synthesis index by combining a long-term trend adjustment mechanism, and forms industrial power scenic spot indexes representing the changing characteristics of industrial economic period stage, operation trend and scenic spot degree by taking each type of diffusion index and each type of synthesis index as scenic spot evaluation basis.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of a method for constructing an industrial power scenic index based on multidimensional data indexes as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of a method for constructing an industrial power scenic index based on multidimensional data indexes as claimed in any one of claims 1 to 7.

Description

Industrial power scenic index construction method, system, equipment and medium based on multidimensional data index Technical Field The invention relates to the technical field of industrial power analysis, in particular to an industrial power scenic index construction method, system, equipment and medium based on multidimensional data indexes. Background The power consumption is used as a 'sunny rain gauge' and a 'thermometer' operated by national economy, and the change of the power consumption is closely related to the development of macroscopic economy, in particular to the production and operation activities of various industries. Therefore, the method for scientifically and accurately constructing the industrial power scenic index has important significance for monitoring the industrial running condition, predicting the industrial power economic trend and assisting government and enterprise decisions in real time. In recent years, the traditional method only researches the overall scenery degree of the electric power, lacks the adaptation to the characteristics of industrial industry, has limited interpretation power of a general model, simultaneously, has single dimension from the electric power data selection of an early warning index pool, is difficult to comprehensively reflect the scenery connotation of the electric power, cannot distinguish electric quantity increase and is driven by benign factors such as production efficiency improvement, capacity expansion and the like, in addition, the consistency index can be used for describing the running track of the overall economy, determining the peak-valley position and the reference track of the economic activity, and the hysteresis index can analyze whether the current economic cycle is ended or is the current-stage circulating service, thereby playing the role of post verification. At present, the prior patent only carries out scenic index construction on the leading indicator, ignores economic fluctuation, is a process of gradually diffusing in a macroscopic economic system, analyzes important variables such as consistent index, hysteresis index and the like, and hardly comprehensively describes industrial economy and electric power prediction and early warning requirements under a high-quality development background. Disclosure of Invention In view of the above-mentioned problems, the present invention provides a method, a system, a device and a medium for constructing industrial power scenic index based on multidimensional data indexes. Therefore, the problem to be solved by the invention is that the prior scheme is to ignore economic fluctuation, which is a process of gradually diffusing in a macroscopic economic system, and analysis of important variables of a consistent index and a hysteresis index, so that it is difficult to comprehensively describe industrial economy and power prediction and early warning requirements under the background of high-quality development. In order to solve the technical problems, the invention provides a method for constructing an industrial electric scenic spot index based on multidimensional data indexes, which comprises the steps of acquiring a plurality of candidate indexes for representing industrial scenic spot changes from industrial electric multidimensional data according to a preset index selection principle to form a scenic spot index pool; the method comprises the steps of analyzing historical economic data, identifying peak value and valley value positions of an economic cycle, determining economic cycle reference date serving as index comparison reference, performing preprocessing operation on historical sequences of various indexes in a scenic gas index pool to obtain a standardized processing sequence for scenic gas analysis, judging the leading degree of the various indexes relative to the economic cycle reference date based on time sequence correlation between the reference indexes and other candidate indexes, further dividing the indexes into leading indexes, consistent indexes and lagging indexes, respectively identifying the changing directions of the leading index group, the consistent index group and the lagging index group, generating leading diffusion indexes, consistent diffusion indexes and lagging diffusion indexes according to the index quantity occupation ratio of the various indexes in an expanding state, weighting and summarizing the various indexes according to preset weight determining rules based on the changing characteristics of the various indexes, respectively obtaining the leading synthesis indexes, the consistent synthesis indexes and the lagging synthesis indexes by combining a long-term trend adjusting mechanism, and forming industrial power indexes representing the industrial economic cycle stage, the running trend and scenic gas degree change characteristics by taking the various diffusion indexes and the various synthesis indexes as scenic gas evaluation basis. The