A free trade zone development analysis system based on power big data and its construction method

Abstract

The present invention discloses a free trade zone development analysis system based on electric power big data and a construction method thereof. The system of the present invention comprises a data acquisition front end; an indicator data display unit; a basic object data display unit; a data preprocessing unit, a data feature extraction unit, a data feature dimension reduction unit; an indicator system data construction unit, an indicator system data evaluation unit; a data model construction unit, a data model training unit; a display back end; the data acquisition front end is connected to the input end of the indicator data display unit, the basic object data display unit, the data preprocessing unit, the data feature extraction unit, the data feature dimension reduction unit, the indicator system data construction unit, the indicator system data evaluation unit, the data model construction unit, and the data model training unit.

Description

Technical Field

[0001] The present invention relates to the field of information technology processing technology, and in particular to a free trade zone development analysis system based on power big data and a construction method thereof. Background Art

[0002] The establishment of the free trade zone is a new pattern of all-round opening up in the new era, benchmarking the international high standards of investment and trade liberalization and facilitation. The free trade zone accelerates the free flow of goods, services, talents, capital, information, etc., from the opening of factors to the opening of international trade and investment rules. The free trade zone mainly includes silicon-based new materials, bio-based new materials, new energy and other industries, as well as silicon-based and bio-based manufacturing centers.

[0003] Existing domestic and international research on the evaluation of free trade zone development is either qualitative, lacking scientificity and accuracy; or only focusing on the quantitative evaluation of a certain aspect of the free trade zone's functional data, but lacking a big data framework for the overall development evaluation of the free trade zone. Summary of the invention

[0004] The purpose of the present invention is to provide a free trade zone development analysis system based on power big data and a construction method thereof to solve the problems raised in the background technology.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: a free trade zone development analysis system based on power big data, including a data acquisition front end; an indicator data display unit; a basic object data display unit; a data preprocessing unit, a data feature extraction unit, a data feature dimension reduction unit; an indicator system data construction unit, an indicator system data evaluation unit; a data model construction unit, a data model training unit; a display back end;

[0006] The data acquisition front end is connected to the input end of the indicator data display unit, the basic object data display unit, the data preprocessing unit, the data feature extraction unit, the data feature dimension reduction unit, the indicator system data construction unit, the indicator system data evaluation unit, the data model construction unit, and the data model training unit;

[0007] The output ends of the indicator data display unit, basic object data display unit, data preprocessing unit, data feature extraction unit, data feature dimension reduction unit, indicator system data construction unit, indicator system data evaluation unit, data model construction unit, and data model training unit are connected to the display backend.

[0008] Furthermore, the indicator data display unit includes a module for displaying the electricity output value data within the free trade zone and a module for displaying the radiation efficiency data of the free trade zone area.

[0009] Furthermore, the basic object data display unit includes a free trade zone enterprise data collection and display module, a free trade zone enterprise directory data display module, a free trade zone enterprise output value data display module, a free trade zone enterprise electricity consumption data display module, a free trade zone enterprise electricity power data display module, a free trade zone enterprise electricity consumption change trend data display module and a free trade zone enterprise industry category data display module.

[0010] Furthermore, the basic object data display unit also includes an upper-level relevant policy display module.

[0011] Furthermore, the display backend includes a free trade zone monitoring and analysis database, and the free trade zone monitoring and analysis database is connected to the display screen unit of the display backend.

[0012] A method for constructing a free trade zone development analysis system based on power big data includes the following steps:

[0013] 1) Data collection and entry: Data collection of enterprises in the free trade zone, including the list of enterprises in the free trade zone, the output value of enterprises in the free trade zone, the relevant policy data of the superior, the electricity consumption data of enterprises in the free trade zone, the power consumption data of enterprises in the free trade zone, the trend data of electricity consumption changes of enterprises in the free trade zone, and the industry category data of enterprises in the free trade zone;

[0014] 2) Data preprocessing, data feature extraction and data feature dimensionality reduction: Aggregate and summarize the acquired raw data in the data center, use Python to clean the data, combine specific data features to perform transitional analysis such as discrete degree and central tendency, and calculate derived indicators, form a transformation from raw data to model wide table indicators, and perform feature scaling and mean normalization mapping on the cleaned data; and combine PCA, LDA, and SVD algorithms to reduce the feature dimensionality of model wide table indicators based on the actual distribution characteristics of the data;

[0015] 3) Construction and evaluation of the free trade zone development indicator data system: It is planned to use the power data of the free trade zone enterprises as the basis and combine it with external data, and use the hierarchical analysis method to determine the weights of various data indicators. At the same time, according to the designed two-layer indicator system, through the evaluation method, an evaluation matrix corresponding to the indicator system is constructed to evaluate and analyze the development data of the free trade zone;

[0016] 4) Model construction and training: K-shape, K-means, random forest, decision tree, and neural network algorithms are used to perform high-dimensional space clustering analysis and enterprise feature data classification on the power users of enterprises in the free trade zone, and to build an industrial data mining and analysis model from the perspective of power; the original data set is divided into a training set and a test set, and the holdout method and cross-validation method are used for verification, and the model is continuously optimized and improved;

[0017] 5) Get the data frame.

[0018] Furthermore, the steps 4) and 5) further include the following specific steps:

[0019] Industrial data analysis and model building from the perspective of electricity:

[0020] a1) Compare the data differences between the sub-industries provided by the superior and the sub-industries in the marketing business system and make records;

[0021] b1) Through the electricity consumption data of various industries in the industrial base, the industry capacity expansion data is counted year by year, and the industry development data and the industry's new and old capacity replacement data are analyzed through the capacity expansion analysis results;

[0022] c1) Combine power data with industry capacity data, analyze industry power consumption and capacity changes, mine the correlation data between industry power consumption and industry capacity, and build a capacity prediction model for key industries;

[0023] d1) Based on the power data analysis, the scale data of each enterprise in the industry is constructed to build an enterprise clustering model, clustering enterprises with similar scale in the same industry, so as to discover enterprises that deviate from the average output value and have high energy consumption;

[0024] e1) Analyze the development balance data of upstream and downstream industrial chains of key development industries in the free trade zone from the perspective of electricity, so as to obtain the ratio of the number, type and production capacity of upstream and downstream enterprises in the industrial chain;

[0025] Data perspective analysis of free trade zone development based on power big data

[0026] a2) Based on the marketing business applications and electricity consumption collection business system data accessed by the data center, combined with the external data of the free trade zone enterprise list, a free trade zone development perspective analysis model based on power big data is constructed using data mining and analysis algorithms;

[0027] b2) Based on the perspective analysis model of the free trade zone development, set the time for the free trade zone special monitoring and analysis, export the special monitoring data report, import the online monitoring report automatic generation tool, and automatically generate the free trade zone monitoring report;

[0028] c2) Based on the results of the special monitoring and analysis of the free trade zones, a free trade zone monitoring and analysis database is designed and produced based on the free trade zone scale data, the free trade zone electricity consumption data on the same period last year, and the free trade zone electricity consumption development trend data.

[0029] Furthermore, the step 4) also includes the acquisition of external enterprise data: the external free trade zone enterprise data is output from the superior department website and industry data, and data use application is made or data is crawled through a web crawler.

[0030] A method for constructing a free trade zone development analysis system based on power big data includes the following steps:

[0031] 1) Data preparation:

[0032] Provide data support based on the power data center, use the data cleaning, data conversion, and data integration of the data center to build the corresponding data model to form the indicator data, business data, and detailed database required for the scenario;

[0033] 2) Data processing:

[0034] Reconstruct scattered big data resources through data cleaning, integration, conversion, and reduction processes, including missing value and outlier processing, feature autocorrelation screening, character encoding processing, and normalization processing;

[0035] 3) Index system data construction and evaluation:

[0036] From the perspective of power, the analytic hierarchy process is used to determine the weights of various indicators, evaluate and analyze the development of the free trade zone, and decompose the free trade zone development evaluation indicator data system into four primary indicator data areas: the free trade zone construction heat data area, the industrial structure analysis data area within the free trade zone, the industrial agglomeration degree data area, and the industrial chain balanced development data area. The secondary indicator data areas are further subdivided under each primary indicator data area;

[0037] According to the designed two-layer indicator system, the evaluation matrix corresponding to the indicator data system is constructed through the evaluation method; then the evaluation matrix is ​​tested for consistency. If the consistency test fails, it is re-evaluated; the weights of indicators at each layer are determined through the arithmetic mean method, geometric mean method, and eigenvalue method; finally, the evaluation results of the free trade zone development are obtained through the accumulation and summation of the quantified indicator values ​​and weights;

[0038] 4) Model construction;

[0039] 5) Application verification:

[0040] To verify the application of the existing model, cross-validation and self-service sampling verification methods are used to test and verify the results of the model by using some non-training set user collected data as test set data;

[0041] Based on the verification results, the model is improved by re-implementing feature engineering, extracting new features based on actual results, selecting different data mining algorithms, and debugging algorithm parameters to improve the accuracy of the model;

[0042] 7) Results display:

[0043] Based on the analysis results of free trade zones and industrial development from the perspective of electricity, a free trade zone development analysis model is constructed, and time-specific monitoring and analysis of free trade zones and industries is set. At the same time, analysis report templates and prototypes are designed, and the report content is filled in according to the template format. The analysis result data is displayed on the prototype page. Finally, the report and prototype are used to display data information on changes in the number of free trade zones and industrial enterprises, changes in enterprise scale, industrial agglomeration, balance of the industrial chain, changes in enterprise production capacity, electricity consumption trends, and year-on-year and month-on-month changes in enterprise electricity consumption.

[0044] The technical effect of the present invention is that the present invention analyzes the development data of the free trade zone from the perspective of electricity, analyzes the industrial development with the silicon-based industry as the object, establishes and displays the corresponding database, and uses the corresponding database to provide a reference basis for the execution center or the control center to formulate and optimize the local free trade zone policy, so as to better promote the development of free trade. That is, the present invention combines the positioning of the development of the free trade zone with the enterprise data in the free trade zone, the needs of operating the free trade zone, and the electricity consumption characteristic data of each enterprise in the free trade zone, and constructs an analysis system that conforms to the development of the free trade zone from the perspective of electricity. Through the construction of the free trade zone development indicator data system, the hierarchical analysis method is used to assign corresponding weights to different data. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a simplified diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0046] The present invention conducts a data analysis of enterprise development in the free trade zone from the perspective of electricity, intuitively reflects the operating status of enterprises in the free trade zone through electricity data, and takes the silicon-based industry as an example to conduct a data analysis of silicon-based industry development. Finally, the development status of the free trade zone is evaluated based on the analysis of indicator system data, free trade zone enterprise operating status data and silicon-based industry development status data.

[0047] First, establish a free trade zone development indicator data system based on power big data, and reflect the overall development data of the free trade zone through power data analysis, including electricity output value data within the free trade zone, superior efficiency data, free trade zone regional radiation efficiency data, industrial agglomeration and industrial chain balanced development data, etc.

[0048] Second, from the perspective of electricity, we will analyze the economic operation data of enterprises in the key development areas of the free trade zone. Taking the silicon-based industry in the area as an example, we will analyze the electricity consumption data of key industries, the replacement data of old and new production capacity, the output value and energy consumption data, the industrial chain development data, etc., to provide decision-making support for higher-level departments.

[0049] Third, establish a perspective analysis of the free trade zone development based on power big data to achieve thematic analysis and presentation of data.

[0050] A free trade zone development analysis system based on power big data of the present invention comprises a data collection front end 1; an indicator data display unit 3; a basic object data display unit 4; a data preprocessing unit 51, a data feature extraction unit 52, a data feature dimension reduction unit 53; an indicator system data construction unit 61, an indicator system data evaluation unit 62; a data model construction unit 63, a data model training unit 64; a display back end 2;

[0051] The data acquisition front end 1 is connected to the input end of the indicator data display unit 3, the basic object data display unit 4, the data preprocessing unit 51, the data feature extraction unit 52, the data feature dimension reduction unit 53, the indicator system data construction unit 61, the indicator system data evaluation unit 62, the data model construction unit 63, and the data model training unit 64;

[0052] The output ends of the indicator data display unit 3, basic object data display unit 4, data preprocessing unit 51, data feature extraction unit 52, data feature dimension reduction unit 53, indicator system data construction unit 61, indicator system data evaluation unit 62, data model construction unit 63, and data model training unit 64 are connected to the display backend 2.

[0053] Furthermore, the index data display unit 3 includes a free trade zone electricity output value data display module 31 and a free trade zone regional radiation efficiency data display module 32.

[0054] Furthermore, the basic object data display unit 4 includes a free trade zone enterprise data collection and display module 41, a free trade zone enterprise directory data display module 42, a free trade zone enterprise output value data display module 43, a free trade zone enterprise electricity consumption data display module 44, a free trade zone enterprise electricity power data display module 45, a free trade zone enterprise electricity consumption change trend data display module 46 and a free trade zone enterprise industry category data display module 47.

[0055] Furthermore, the basic object data display unit 4 also includes an upper-level relevant policy display module 48.

[0056] Furthermore, the display backend 2 includes a free trade zone monitoring and analysis database 21 , and the free trade zone monitoring and analysis database 21 is connected to the display screen unit 22 of the display backend 2 .

[0057] Specifically: the system of the present invention includes the following steps:

[0058] 1) Data collection and entry: Data collection of free trade zone enterprises, including free trade zone enterprise directory data, free trade zone enterprise output value data, superior relevant policy data, free trade zone enterprise electricity consumption data, free trade zone enterprise electricity power data, free trade zone enterprise electricity consumption trend data, free trade zone enterprise industry category data, etc.

[0059] 2) Data preprocessing, data feature extraction and (data) feature dimensionality reduction: The acquired raw data is aggregated and summarized in the data center, and data cleaning is performed using Python. Based on the specific data features, a transitional analysis of the degree of dispersion and central tendency and calculation of derived indicators are performed to form a transformation from raw data to model wide table indicators. The cleaned data is then subjected to feature scaling and mean normalization mapping. Based on the actual distribution characteristics of the data, PCA, LDA, and SVD algorithms are used to perform feature dimensionality reduction on the model wide table indicators.

[0060] 3) Construction and evaluation of the free trade zone development indicator data system: It is planned to use the power data of the free trade zone enterprises as the basis and combine it with external data, and use the hierarchical analysis method to determine the weights of various data indicators. At the same time, according to the designed two-layer indicator system, through the evaluation method, an evaluation matrix corresponding to the indicator system is constructed to evaluate and analyze the development data of the free trade zone, so as to provide big data reference for the subsequent development and construction of the free trade zone.

[0061] 4) Data model construction and training: K-shape, K-means, random forest, decision tree, and neural network algorithms are used to conduct high-dimensional space clustering analysis and enterprise characteristic data classification on corporate electricity users in the free trade zone, and to build an industrial data mining and analysis model from the perspective of electricity; the original data set is divided into a training set and a test set, and the holdout method and cross-validation method are used for verification, and the model is continuously optimized and improved.

[0062] 5) Get the data frame.

[0063] Specific embodiment 1, taking the data of silicon-based new materials industry in the free trade zone as an example, the process of obtaining the data framework is as follows:

[0064] This paper analyzes the development data of the free trade zone and silicon-based enterprises from the perspective of electricity. The data application construction content mainly includes:

[0065] Data analysis and model building of silicon-based industry from the perspective of electricity:

[0066] a1) Compare the data differences between the silicon-based new material industry segments provided by the superior and the industry segments in the marketing business system and make records;

[0067] b1) Through the electricity consumption data of various industries in the silicon-based industrial base, the industry capacity expansion data is counted year by year, and the industry development data and the industry's new and old capacity replacement data are analyzed through the capacity expansion analysis results;

[0068] c1) Combine power data with silicon-based industry capacity data, analyze industry power consumption and capacity changes, and mine the correlation data between industry power consumption and industry capacity, so as to achieve the purpose of using power data to characterize industry output value. At the same time, build a capacity prediction model for key industries to predict the capacity change trend of key silicon-based industries in the future;

[0069] d1) Based on the power data analysis, the scale data of each enterprise in the industry is constructed to build an enterprise clustering model, clustering enterprises with similar scale in the same industry, so as to discover enterprises that deviate from the average output value and have high energy consumption;

[0070] e1) Based on the power perspective, the data on the development balance of the upstream and downstream industrial chains of the key development industries of silicon-based new materials in the free trade zone are analyzed to obtain the ratio of the number, type and production capacity of upstream and downstream enterprises in the industrial chain.

[0071] Data perspective analysis of free trade zone development based on power big data

[0072] a2) Based on the marketing business applications and electricity consumption collection business system data accessed by the data middle platform, combined with the external data of the free trade zone enterprise list, a free trade zone development perspective analysis model based on power big data is constructed using data mining and analysis algorithms to achieve parallel monitoring and analysis functions in both time and industry dimensions, effectively supporting the special monitoring and analysis work of the free trade zone.

[0073] b2) Based on the perspective analysis model of the free trade zone development, set the time for special monitoring and analysis of the free trade zone, export the special monitoring data report, import the online monitoring report automatic generation tool, and automatically generate the free trade zone monitoring report.

[0074] c2) Based on the results of special monitoring and analysis of the free trade zones, a free trade zone monitoring and analysis database will be designed and produced based on the free trade zone scale data, the free trade zone electricity consumption data on the same period last year, and the free trade zone electricity consumption development trend data, to centrally display and publicize the free trade zone construction and development data.

[0075] Furthermore, it also includes the acquisition of external enterprise data:

[0076] The external free trade zone enterprise data is output from the superior department website and silicon-based industry data, and data use applications are made or data is crawled through web crawlers.

[0077] Specific embodiment 2, still taking the silicon-based new materials industry as an example, the silicon-based industry situation analysis and model construction based on power data:

[0078] Based on the mining and analysis of electricity data, taking the silicon-based industry in the area as an example, the electricity data is combined with the silicon-based industry's production capacity data, and the changing trends of the industry's electricity consumption and production capacity data are analyzed. The relationship between the industry's electricity consumption and industry production capacity data is mined to achieve the purpose of using electricity data to represent the industry's output value data. At the same time, a key industry production capacity prediction model is constructed to predict the future production capacity change trends of key industries in the silicon-based industry, and provide data support for superior departments and enterprises in terms of the replacement of old and new production capacities, output value energy consumption, and industrial chain development.

[0079] The system includes the following steps:

[0080] 1) Data preparation

[0081] Based on the data support provided by the power data middle platform, the data cleaning, data conversion and data integration of the data middle platform are used to build the corresponding data model, form the indicator data, business data and detailed database required for the scenario, and ensure the timeliness, accuracy and completeness of the scenario application data.

[0082] 2) Data processing:

[0083] The power system data may be interfered by noise during the collection and transmission process, resulting in low data quality. Therefore, corresponding data preprocessing is required before the data is processed.

[0084] Data preprocessing is to reconstruct the scattered big data resources (according to the given business theme) through data cleaning, integration, conversion and reduction, so as to provide basic data support for subsequent analysis and mining. Data preprocessing mainly includes missing value and outlier processing, feature autocorrelation screening, character encoding processing and normalization processing.

[0085] 3) Index system data construction and evaluation:

[0086] From the perspective of electricity, the hierarchical analysis method is used to determine the weights of various indicators, evaluate and analyze the development of the free trade zone, and provide a reference for the subsequent development and construction of the free trade zone. The free trade zone development evaluation index data system is decomposed into four first-level indicator data areas: the free trade zone construction heat data area, the industrial structure analysis data area within the free trade zone, the industrial agglomeration degree data area, and the industrial chain balanced development data area. The second-level indicator data area is further subdivided under each first-level indicator data area.

[0087] According to the designed two-layer indicator system, the evaluation matrix corresponding to the indicator data system is constructed through the evaluation method; then the evaluation matrix is ​​tested for consistency. If the consistency test fails, it is re-evaluated; the weights of indicators at each layer are determined through the arithmetic mean method, geometric mean method, and eigenvalue method; finally, the development evaluation results of the free trade zone are obtained by accumulating and summing the quantified indicator values ​​and weights.

[0088] Specific:

[0089] 3.1) Data analysis on the popularity of free trade zone construction

[0090] The construction enthusiasm in the free trade zone is analyzed by analyzing the number of newly registered enterprises in the free trade zone, the change data of enterprise electricity consumption, the load rate data, the number of new users and electricity data of the construction industry in the free trade zone, the number of new customers applying for temporary electricity use, the electricity consumption, and the data on the conversion of electricity consumption from temporary electricity use to formal electricity use;

[0091] 3.2) Analysis of industrial structure within the FTZ

[0092] Statistics are collected on the number of enterprises, industry categories, electricity consumption and output value in the free trade zone, and the changing trends of the number of enterprises, production capacity and electricity consumption in the free trade zone are analyzed. The trend of the number of enterprises, production capacity and electricity consumption is used to comprehensively represent the level of industrial development in the free trade zone. At the same time, the industrial combination status of the free trade zone is analyzed based on the current status of enterprises in the free trade zone, and finally the correlation between the industrial development level and the industrial combination status is explored to realize the industrial structure analysis in the free trade zone;

[0093] 3.3) Construction of capacity forecasting model for key industries

[0094] Combine the industry capacity data provided by the superior departments with the industry electricity consumption data, deeply explore the relationship between industry capacity and industry electricity consumption, use electricity data to characterize industry capacity, use time series and other algorithms to build an industry capacity prediction model, and use this model to predict the capacity change trend of key industries in the future silicon-based industry;

[0095] 3.4) Analysis of industrial concentration in the free trade zone

[0096] The development of silicon-based industry, bio-based industry and new energy industry is the key industry of the free trade zone. The concentration of the three major industries in the free trade zone can be used as the evaluation basis for the development of the free trade zone. The location entropy algorithm can reflect the level of industrial agglomeration at the regional level to a certain extent. The location entropy algorithm is used to combine the data of the number of enterprises, industry categories, enterprise production capacity and enterprise electricity consumption in the region and the free trade zone to comprehensively analyze the level of industrial agglomeration in the free trade zone.

[0097] 3.5) Analysis of enterprise output value deviation

[0098] There must be differences in the production capacity of enterprises of different sizes. In order to deeply explore the enterprises in the same industry that deviate greatly from the average output value (electricity consumption), we first cluster the enterprises of similar scale in the same industry to achieve the analysis effect of the only variable, explore the electricity consumption characteristics of enterprises of different scales, characterize the scale of enterprises with electricity consumption characteristics, use clustering algorithms to build an enterprise clustering model, and scientifically group enterprises in the same industry to explore enterprises that deviate greatly from the mean.

[0099] 4) Model construction:

[0100] 4.1) Analysis of differences between internal and external industry details

[0101] Compare the list of silicon-based enterprises and corresponding industry categories provided by the superior department with the industry category data in the power marketing business system. If there are any differences in industry naming, record the differences to facilitate the improvement of industry category naming in the power marketing business system;

[0102] 4.2) Analysis of the characteristics of capacity replacement in the silicon-based industry

[0103] As the industry's technological capabilities improve, the industry will also improve accordingly. Use mathematical statistics methods to find the characteristics of industry capacity updates, comprehensively analyze the changing patterns of electricity consumption data under the characteristics of capacity replacement, and explore the electricity consumption characteristics during capacity updates, so as to find the industry capacity replacement characteristics from the perspective of electricity, and characterize the capacity replacement characteristics through electricity data characteristics;

[0104] 4.3) Analysis of the balance of silicon-based industry chain

[0105] First, based on the silicon-based industrial chain relationship network, explore the configuration data of upstream and downstream enterprises in the industrial chain, and analyze whether there are chain breakpoints in the industrial chain; second, explore the correlation between the production capacity and electricity of various industries in the silicon-based industrial chain, use electricity to represent the industry capacity, and use mathematical statistics and big data technology to analyze the relationship between the production capacity and consumption of various industries upstream and downstream of the industrial chain. This is used as the basis for judging the capacity balance of upstream and downstream industries in the silicon-based industrial chain. According to the capacity balance standard of the silicon-based industrial chain industry, the weak links in the chain are explored, and the enterprise configuration data of various industries in the weak links of the chain are deeply analyzed from top to bottom to achieve accurate positioning of the weak links in the industrial chain.

[0106] 5) Application Verification: Carry out application verification on the existing model, adopt cross-validation and self-service sampling verification methods to test and verify the results of the model by using some non-training set user collected data as test set data.

[0107] Based on the verification results, the model is improved by re-implementing feature engineering, extracting new features based on actual results, selecting different data mining algorithms, debugging algorithm parameters, and other methods to improve the accuracy of the model.

[0108] 6) Results display:

[0109] Based on the analysis results of the development of the free trade zones and silicon-based industries from the perspective of electricity, a free trade zone development analysis model is constructed, and a time-specific monitoring and analysis of the free trade zones and silicon-based industries is set. At the same time, in order to intuitively display the development of the free trade zones and silicon-based industries, analysis report templates and prototypes are designed. The report content is filled according to the template format, and the analysis result data is displayed on the prototype page. Finally, the report and prototype are used to display data information such as changes in the number of free trade zone and silicon-based industry enterprises, changes in enterprise scale, industrial agglomeration, balance of the industrial chain, changes in enterprise production capacity, electricity consumption trends, and year-on-year and month-on-month changes in enterprise electricity consumption.

[0110] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A free trade zone development analysis system based on power big data, characterized in that: It includes a data collection front end (1); an indicator data display unit (3); a basic object data display unit (4); a data preprocessing unit (51), a data feature extraction unit (52), and a data feature dimension reduction unit (53); an indicator system data construction unit (61), an indicator system data evaluation unit (62); a data model construction unit (63), and a data model training unit (64); and a display back end (2); The data collection front end (1) is connected to the input ends of the indicator data display unit (3), the basic object data display unit (4), the data preprocessing unit (51), the data feature extraction unit (52), the data feature dimension reduction unit (53), the indicator system data construction unit (61), the indicator system data evaluation unit (62), the data model construction unit (63), and the data model training unit (64); The output ends of the indicator data display unit (3), the basic object data display unit (4), the data preprocessing unit (51), the data feature extraction unit (52), the data feature dimension reduction unit (53), the indicator system data construction unit (61), the indicator system data evaluation unit (62), the data model construction unit (63), and the data model training unit (64) are connected to the display backend (2); The index data display unit (3) comprises a free trade zone electricity output value data display module (31) and a free trade zone regional radiation efficiency data display module (32); The basic object data display unit (4) includes a free trade zone enterprise data collection display module (41), a free trade zone enterprise directory data display module (42), a free trade zone enterprise output value data display module (43), a free trade zone enterprise electricity consumption data display module (44), a free trade zone enterprise power consumption data display module (45), a free trade zone enterprise electricity consumption change trend data display module (46) and a free trade zone enterprise industry category data display module (47); The method for constructing a free trade zone development analysis system based on power big data comprises the following steps: 1) Data collection and entry: Data collection of enterprises in the free trade zone, including the list of enterprises in the free trade zone, the output value of enterprises in the free trade zone, the relevant policy data of the superior, the electricity consumption data of enterprises in the free trade zone, the power consumption data of enterprises in the free trade zone, the trend data of electricity consumption changes of enterprises in the free trade zone, and the industry category data of enterprises in the free trade zone; 2) Data preprocessing, data feature extraction and data feature dimensionality reduction: Aggregate and summarize the acquired raw data in the data center, use Python to clean the data, conduct discrete degree and central tendency transition analysis and derivative index calculation based on specific data features, form a transformation from raw data to model wide table indicators, and perform feature scaling and mean normalization mapping on the cleaned data; and perform feature dimensionality reduction on model wide table indicators based on the actual distribution characteristics of the data in combination with PCA, LDA or SVD algorithms; 3) Construction and evaluation of the free trade zone development indicator data system: It is planned to use the power data of enterprises in the free trade zone as the basis and combine it with external data, and use the hierarchical analysis method to determine the weights of various data indicators. At the same time, according to the designed two-layer indicator system, through the evaluation method, an evaluation matrix corresponding to the indicator system is constructed to evaluate and analyze the development data of the free trade zone; 4) Model construction and training: Use K-shape or K-means or random forest or decision tree or neural network algorithm to perform high-dimensional space cluster analysis and enterprise feature data classification on the power users of enterprises in the free trade zone, and build an industrial data mining and analysis model from the perspective of power; divide the original data set into training set and test set, and use the holdout method and cross-validation method to verify and continuously optimize and improve the model; Industrial data analysis and model building from the perspective of electricity: a1) Compare the data differences between the industry segments provided by the superior and the industry segments in the marketing business system and make records; b1) Through the electricity consumption data of various industries in the industrial base, the industry capacity expansion data is counted year by year, and the industry development data and the industry's new and old capacity replacement data are analyzed through the capacity expansion analysis results; c1) Combine power data with industry capacity data, analyze changes in industry power consumption and capacity, mine the correlation data between industry power consumption and industry capacity, and build a capacity prediction model for key industries; d1) Based on the scale data of each enterprise in the power data analysis industry, an enterprise clustering model is constructed to cluster enterprises with similar scale in the same industry, so as to discover enterprises that deviate from the average output value and have high energy consumption; e1) Analyze the development balance data of upstream and downstream industrial chains of key development industries in the free trade zone from the perspective of electricity, so as to obtain the ratio of the number, type and production capacity of upstream and downstream enterprises in the industrial chain; Data perspective analysis of free trade zone development based on power big data a2) Based on the marketing business applications and electricity consumption collection business system data accessed by the data center, combined with the external data of the free trade zone enterprise list, a free trade zone development perspective analysis model based on power big data is constructed using data mining and analysis algorithms; b2) Based on the perspective analysis model of the free trade zone development, set the time for the free trade zone special monitoring and analysis, export the special monitoring data report, import the online monitoring report automatic generation tool, and automatically generate the free trade zone monitoring report; c2) Based on the results of the FTZ special monitoring and analysis, design and produce the FTZ monitoring and analysis database from the FTZ scale data, the FTZ electricity consumption data of the same period last year, and the FTZ electricity consumption development trend data; 5) Get the data frame.

2. According to the free trade zone development analysis system based on power big data according to claim 1, it is characterized in that: The basic object data display unit (4) also includes a superior-related policy display module (48).

3. According to the free trade zone development analysis system based on power big data according to claim 1, it is characterized in that: The display backend (2) comprises a free trade zone monitoring and analysis database (21), and the free trade zone monitoring and analysis database (21) is connected to a display screen unit (22) of the display backend (2).

4. According to the free trade zone development analysis system based on power big data according to claim 1, it is characterized in that: The step 4) also includes the acquisition of external enterprise data: the external free trade zone enterprise data is output from the superior department website and industry data, and data use application is made or data is crawled through a web crawler.

5. A method for constructing a free trade zone development analysis system based on power big data according to claim 1, characterized in that: The steps include: 1) Data preparation: Provide data support based on the power data center, use the data cleaning, data conversion, and data integration of the data center to build the corresponding data model to form the indicator data, business data, and detailed database required for the scenario; 2) Data processing: Reconstruct scattered big data resources through data cleaning, integration, conversion, and reduction processes, including missing value and outlier processing, feature autocorrelation screening, character encoding processing, and normalization processing; 3) Index system data construction and evaluation: From the perspective of power, the analytic hierarchy process is used to determine the weights of various indicators, evaluate and analyze the development of the free trade zone, and decompose the free trade zone development evaluation indicator data system into four primary indicator data areas: the free trade zone construction heat data area, the industrial structure analysis data area within the free trade zone, the industrial agglomeration degree data area, and the industrial chain balanced development data area. The secondary indicator data areas are further subdivided under each primary indicator data area; According to the designed two-layer indicator system, the evaluation matrix corresponding to the indicator data system is constructed through the evaluation method; then the evaluation matrix is ​​checked for consistency. If the consistency check fails, it is re-evaluated; The weights of indicators at each level are determined by arithmetic mean, geometric mean and eigenvalue methods; finally, the evaluation results of the free trade zone development are obtained by accumulating and summing the quantitative indicator values ​​and weights; 4) Model construction; 5) Application Verification: To verify the application of the existing model, cross-validation and self-service sampling verification methods are used to test and verify the results of the model by using some non-training set user collected data as test set data; Based on the verification results, the model is improved by re-implementing feature engineering, extracting new features based on actual results, selecting different data mining algorithms, and debugging algorithm parameters to improve the accuracy of the model; 7) Results display: Based on the analysis results of free trade zones and industrial development from the perspective of electricity, a free trade zone development analysis model is constructed, and time-specific monitoring and analysis of free trade zones and industries is set. At the same time, analysis report templates and prototypes are designed, and the report content is filled in according to the template format. The analysis result data is displayed on the prototype page. Finally, the report and prototype are used to display data information on changes in the number of free trade zones and industrial enterprises, changes in enterprise scale, industrial agglomeration, balance of the industrial chain, changes in enterprise production capacity, electricity consumption trends, and year-on-year and month-on-month changes in enterprise electricity consumption.

Images