CN113191802A - Clustering algorithm-based natural gas customer analysis method and system - Google Patents
Clustering algorithm-based natural gas customer analysis method and system Download PDFInfo
- Publication number
- CN113191802A CN113191802A CN202110460875.XA CN202110460875A CN113191802A CN 113191802 A CN113191802 A CN 113191802A CN 202110460875 A CN202110460875 A CN 202110460875A CN 113191802 A CN113191802 A CN 113191802A
- Authority
- CN
- China
- Prior art keywords
- customer
- natural gas
- analysis
- data
- analysis method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 138
- 238000004458 analytical method Methods 0.000 title claims abstract description 70
- 239000003345 natural gas Substances 0.000 title claims abstract description 69
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 27
- 238000007621 cluster analysis Methods 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 26
- 239000007789 gas Substances 0.000 claims description 161
- 230000009467 reduction Effects 0.000 claims description 21
- 230000001932 seasonal effect Effects 0.000 claims description 13
- 238000013461 design Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000000556 factor analysis Methods 0.000 claims description 4
- 238000003064 k means clustering Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 8
- 238000010438 heat treatment Methods 0.000 description 5
- 239000003949 liquefied natural gas Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000003889 chemical engineering Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000003337 fertilizer Substances 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a natural gas customer analysis method and a system based on a clustering algorithm, wherein the analysis method comprises the following steps: collecting customer data of a natural gas customer; carrying out cluster analysis on the customer data by using a cluster analysis method, and dividing the natural gas customers into different groups of customer groups; and analyzing the value of a plurality of groups of the customer groups. According to the invention, the customers are divided into different value customer groups through a clustering algorithm, so that different customer strategies can be set conveniently aiming at different customer characteristics, the limitation of developing customer analysis by means of experience and traditional statistical means is solved, the analysis is more comprehensive, the analysis effect is more accurate, and different strategies can be set conveniently aiming at different customer characteristics.
Description
Technical Field
The invention belongs to the field of big data analysis, and particularly relates to a natural gas customer analysis method and system based on a clustering algorithm.
Background
For natural gas customer analysis, the following three aspects are mainly considered: firstly, customer insights are effective means and basic guarantee for improving customer service quality and realizing natural gas marketing and efficiency improvement. Customer appeal to natural gas supply presents three main aspects: firstly, the supply is sufficient and stable, and the gas supply safety is ensured; secondly, the price has certain advantages and the cost is controllable; and thirdly, certain adjustment service can be provided in operation. In order to continuously meet customer requirements and simultaneously improve sales benefits, differentiated services and marketing strategies need to be implemented on the basis of scientific insight on gas use characteristics and demand changes of different customers, demand price elasticity and gas purchase decision behaviors. Secondly, researching a customer analysis method is the need of accurately describing the characteristics of the multi-element coupling gas consumption of customers. The natural gas consumption has the characteristics of various related industries, remarkable regional characteristics and the like. The customer gas consumption characteristics are the comprehensive reflection of the gas consumption fluctuation of different terminal users, and the multivariate coupling cross relationship exists among external influence factors influencing the gas consumption fluctuation of the users. The method is characterized in that the characteristics of the gas consumption of a customer are related to the correlation analysis of hundreds of external factor mass data, and quantitative characterization needs to be carried out by means of mass data analysis. And thirdly, the accurate grasp of the characteristics of the client is an urgent need for improving the operation safety of the natural gas industry chain of the company and expanding the sales volume. The fluctuation of the natural gas customer demand is closely related to the industrial characteristics of the customer, and the change of external factors such as economic environment, major events, energy price and the like has great influence on the customer gas demand. At present, a customer analysis technology based on a cluster analysis method is not available in the natural gas industry, customers can be analyzed only by experience or statistics, and the method has certain limitations, incomplete analysis and inaccurate analysis structure.
Disclosure of Invention
In order to solve the problems, the invention provides a natural gas customer analysis method and system based on a clustering algorithm, which are convenient for setting different strategies according to different customer characteristics.
In order to achieve the purpose, the invention adopts the technical scheme that: a natural gas customer analysis method based on a clustering algorithm, the analysis method comprising the steps of:
collecting customer data of a natural gas customer;
carrying out cluster analysis on the customer data by using a cluster analysis method, and dividing the natural gas customers into different groups of customer groups;
and analyzing the value of a plurality of groups of the customer groups.
Optionally, before the step of performing cluster analysis on the natural gas customer, the method further comprises the steps of:
and designing a data wide table field for the collected customer data.
Optionally, after the step of designing the data wide table field and before the step of performing cluster analysis on the natural gas customer, the method further comprises the steps of:
and performing dimension reduction processing on the data wide table field.
Optionally, in the step of performing dimension reduction processing on the data wide table fields, the dimension reduction processing is performed by using a factor analysis method, the number of the data wide table fields with loads higher than a predetermined threshold is reduced, and the cluster analysis is performed by using the reduced number of the data wide table fields.
Optionally, in the step of designing the data wide table field for the collected customer data, the step of performing feature classification on the customer data is further included.
Optionally, the categories of the feature classification at least comprise basic features of the customer, recent gas usage rules, seasonal preferences, planned gas usage conditions, historical trend changes, pressure reduction execution conditions, financial credit indicators and customer status.
Optionally, in the step of dividing the natural gas clients into different client groups by using a cluster analysis method, the natural gas clients are divided into a plurality of groups of client groups according to the cluster result characteristics, the contour values and the cluster map by using a K-Means cluster analysis method.
Optionally, after the step of analyzing the value of the customer base, the method further comprises the steps of:
and setting different customer strategies aiming at different customer groups according to the value analysis result.
And, a natural gas customer analysis system based on clustering algorithm, comprising:
the data acquisition unit is used for acquiring customer data of a natural gas customer;
the cluster analysis unit is used for carrying out cluster analysis on the customer data by using a cluster analysis method and dividing the natural gas customers into different groups of customer groups;
a value analysis unit for performing a value analysis on the customer base.
Optionally, the analysis system further includes a field design unit, and the field design unit is configured to design a data wide table field for the collected customer data.
Optionally, the analysis system further includes a data dimension reduction unit, where the data dimension reduction unit is configured to perform dimension reduction processing on the data wide table field.
Optionally, the cluster analysis unit performs cluster analysis by using a K-Means cluster analysis method.
Due to the adoption of the technical scheme, the invention has the following beneficial effects: through the clustering algorithm, the clients are divided into different value client groups, different client strategies can be set conveniently according to different client characteristics, the limitation that client analysis is carried out by means of experience and traditional statistics is solved, analysis is more comprehensive, analysis effect is more accurate, and different strategies can be set conveniently according to different client characteristics.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 shows a flow chart of steps of a method for natural gas customer analysis based on a clustering algorithm according to an embodiment of the present invention.
Fig. 2 shows a system block diagram of a natural gas customer analysis system based on a clustering algorithm according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the method for analyzing a natural gas customer based on a clustering algorithm in an embodiment of the present invention includes the following steps:
s1: customer data of a natural gas customer is collected.
In fact, different customer groups have different and significant impact factors on them. For example, some urban gas customers use natural gas for heating in winter, the seasonality is particularly obvious, the planned gas consumption rate of some customers is very high, the natural gas configuration, scheduling and control are facilitated, and the like.
S2: and designing a data wide table field for the collected customer data.
In this embodiment, the data wide table field corresponding to the collected customer data includes, but is not limited to: customer ID, customer name, province, industry type, annual gas usage, monthly average gas usage, winter gas usage duty, weekend gas usage duty, and the like. Specifically, the following tables 1 to 8 show.
In order to improve the analysis effect and accuracy, the collected customer data comprises a plurality of category factors, such as the categories of basic characteristics, recent gas usage rules, seasonal preferences, planned gas usage conditions, historical trend changes, pressure reduction execution conditions, financial credit indexes, customer states and the like of each customer are observed, and the multi-index cross analysis is performed to meet the cognitive mode of overall behavior evaluation of the customer. Therefore, the customer data of a plurality of natural gas customers is feature-classified according to the above-described features.
For example, the client ID, the client name, the province, the industry type, etc. are attributed to the client basic characteristics; the recent gas utilization rules include annual gas utilization, monthly gas utilization, winter gas utilization, weekend gas utilization and the like. The following are shown in the following table 1-table 8:
wide table field | Feature classification |
Customer ID | Basic characteristics of customer |
MDM coding | Basic characteristics of customer |
Name of customer | Basic characteristics of customer |
Client abbreviation | Basic characteristics of customer |
Province of labor | Basic characteristics of customer |
Company of the affiliated area | Basic characteristics of customer |
Company of affiliated group | Basic characteristics of customer |
Type of air supply | Basic characteristics of customer |
Whether it can be interrupted | Basic characteristics of customer |
Dependence of gas source | Basic characteristics of customer |
Whether or not the interior directly belongs to the refinery | Basic characteristics of customer |
Type of industry | Basic characteristics of customer |
TABLE 1 customer base characteristics Categories
Wide table field | Field interpretation | Feature classification |
Annual gas consumption | Natural gas purchase amount of user in one year | Law of recent gas use |
Proportion of annual gas consumption to total gas consumption | Annual gas usage/total gas usage of all customers per year | Law of recent gas use |
Monthly average gas consumption | The natural gas purchase amount of the user in one year/purchased in several months | Law of recent gas use |
Winter gas consumption ratio | Gas consumption/annual gas consumption in different regional heating seasons | Law of recent gas use |
Proportion of gas consumption in non-winter | Gas consumption/annual gas consumption in non-heating seasons of different regions | Law of recent gas use |
Total gas consumption in working day | Calculating the total gas consumption of working days | Law of recent gas use |
Total gas consumption at weekend | Calculating the total gas consumption of weekends | Law of recent gas use |
Working daily gas ratio | Daily gas consumption total/annual gas consumption | Law of recent gas use |
Gas consumption ratio at weekend | Gas consumption of weekend/annual gas consumption | Law of recent gas use |
Maximum monthly gas consumption in the year | Calculating the gas consumption of different months and finding out the maximum value of the gas consumption of the months | Law of recent gas use |
Month with maximum gas consumption in the year | Calculating the gas consumption of different months and finding out the month with the maximum gas consumption | Law of recent gas use |
Minimum monthly gas consumption in the year | Calculating the gas consumption of different months and finding out the minimum value of the gas consumption of the months | Law of recent gas use |
Month with lowest gas consumption in the year | Calculating the gas consumption of different months and finding out the month with the minimum gas consumption | Law of recent gas use |
Selling price | Actual selling price | Law of recent gas use |
TABLE 2 recent usage patterns
Wide table field | Field interpretation | Feature classification |
Gas consumption in the first quarter of the year | Total amount of gas used for 1 to 3 months | Seasonal preferences |
Gas consumption ratio in the first quarter of the year | Total gas consumption of 1-3 months/annual gas consumption | Seasonal preferences |
Gas consumption in the second quarter of the year | Total amount of gas used for 4 to 6 months | Seasonal preferences |
The ratio of gas consumption in the second quarter of the year | Total gas consumption of 4-6 months/annual gas consumption | Seasonal preferences |
Gas consumption in the third quarter of the year | Total amount of gas used for 7 to 9 months | Seasonal preferences |
Gas consumption ratio in the third quarter of the year | Total gas consumption of 7-9 months/annual gas consumption | Seasonal preferences |
Gas consumption in the fourth quarter of the year | Total amount of gas used for 10 to 12 months | Seasonal preferences |
The ratio of gas consumption in the fourth quarter of the year | Total monthly gas consumption of 10-12/annual gas consumption | Seasonal preferences |
Gas characteristics | Calculating the gas consumption in winter/the gas consumption not in winter, | seasonal preferences |
TABLE 3 season preference categories
Wide table field | Field interpretation | Feature classification |
Total planned gas usage on weekends | Calculating the total planned gas consumption of weekends | Planned gas usage |
Daily planned air consumption ratio | Daily planned gas usage amount total/annual planned gas usage amount | Planned gas usage |
Ratio of planned gas consumption on weekend | Weekend planned gas usage total/annual planned gas usage | Planned gas usage |
Annual planned gas consumption | Monthly plan volume summation | Planned gas usage |
Monthly average planned gas consumption | Monthly planning volume sum/monthly number of months planned | Planned gas usage |
Planned air consumption ratio in winter | Planned gas usage/annual planned gas usage for heating seasons in different regions | Planned gas usage |
Proportion of planned gas consumption in non-winter | Planned gas usage/annual planned gas usage in non-heating seasons of different regions | Planned gas usage |
Total of planned gas consumption on weekdays | Calculating the total planned gas usage on a weekday | Planned gas usage |
Annual planned gas consumption | Monthly plan volume summation | Planned gas usage |
TABLE 4 planned gas usage categories
TABLE 5 plan complete Categories
TABLE 6 historical trend change categories
TABLE 7 Depression Enforcement categories
TABLE 8 financial credit index and client status categories
In addition, when collecting customer data of a natural gas customer, data with empty data records such as customer ID and daily sales volume need to be removed, and multiple industries of one customer are defined as customer industries according to the industry with the highest sales volume.
S3: and (3) performing dimension reduction on the data wide table fields, performing dimension reduction by adopting a factor analysis method, reducing the number of fields with loads higher than a preset threshold value (wherein the loads represent the number of the data wide table fields), and performing cluster analysis by using fewer fields to reduce excessive field dimensions.
In the invention, because the data volume is large, the number of derived fields is large, the correlation between partial fields is strong, and the clustering effect is poor if all the fields are used for clustering, the dimensionality reduction is carried out by adopting a factor analysis method. For example, the median prepaid balance and the average prepaid balance in table 8 above, the correlation between the two sets of data is strong. And clustering is performed by using fewer fields, so that the clustering analysis effect is improved.
In order to eliminate the correlation among indexes, the principal component analysis is mainly adopted for reducing the dimension, for the extracted factors, firstly, data wide table fields which have large influence on different factors are selected, and secondly, screening is carried out in all the selected data wide table fields again, wherein the screening principle is as follows: and judging one of the fields with stronger correlation based on a field calculation mode, ensuring the number of the fields selected from each factor to be more balanced as much as possible, and if the selected fields are concentrated in a certain factor, generating the condition of larger distribution difference of clustering groups.
In the step of determining the field having a relatively strong correlation, the determination is mainly made based on the data calculation method of the client data.
For example, as shown in table 9, for the extracted factors, firstly, the fields with large influence on different factors (greater than 0.7) are selected, the fields framed by the digital parts in table 9 are selected, then, the fields are screened again in all the selected fields, and the screening result is the data wide table fields framed by the digital parts in table 9.
TABLE 9 analysis overview
S4: and carrying out cluster analysis on the customer data by using a cluster analysis method, and dividing the natural gas customers into different groups of customer groups.
In this embodiment, since most of the data indexes of the customer data are continuous, the K-Means cluster analysis method is adopted. And dividing the natural gas customers into a plurality of groups of customer groups according to the clustering result characteristics, the contour values and the clustering chart. The contour value combines the aggregation degree and the separation degree of the clusters and is used for evaluating the clustering effect, the value is between-1 and 1, and the larger the value is, the better the clustering effect is. The cluster map can be analyzed by using a sector map with better contrast effect. And then the optimal value of the actual classification number of the customer group is found out according to the comparison of the cluster map. For example, in the analysis, if the contour value of the class 4 is the largest, but it can be seen from the formed cluster map that the distribution of the class 4 is obviously uneven, two classes are too large, and the features in the classes may not be very close, so that the most suitable classification number is obtained by comparing the cluster map with other contour values.
S5: and analyzing the values of the plurality of groups of customer groups.
In the embodiment, the natural gas customers are mainly classified into the following eight types of customer groups according to the collected customer data of a plurality of natural gas customers:
cluster-1 (546): the average monthly gas consumption is high, the same-ratio growth rate of the average monthly gas consumption is low, the average selling price is low, the gas consumption in four seasons is balanced, the gas consumption in winter is medium, the gas consumption in winter is low in the same-ratio growth rate, the plan coincidence rate is low, and the average value of the prepayment balance is low; (high value and high potential)
Cluster-2 (325 people): the average monthly gas consumption is medium, the same-ratio growth rate of the average monthly gas consumption is high, the average selling price is medium, the gas consumption in four seasons is balanced, the gas consumption in winter is medium, the same-ratio growth rate of the gas consumption in winter is high, the plan coincidence rate is medium, and the average value of the prepayment balance is medium; (high-value growth)
Cluster-3 (17 people): the average monthly air consumption is low, the same-ratio growth rate of the average monthly air consumption is low, in the average selling price, the imbalance of the air consumption of four seasons is mainly concentrated in the second quarter, the winter air consumption is low and is a negative value compared with the same-ratio growth rate, the plan coincidence rate is low, and the average value of the prepayment balance is high; (Low value summer type)
Cluster-4 (148 people): the average monthly gas consumption is low, the same-ratio growth rate of the average monthly gas consumption is low, the average selling price is high, the imbalance of the gas consumption in four seasons is mainly concentrated in the fourth quarter, the gas consumption in winter is high, the gas consumption in winter is medium in the same-ratio growth rate, the plan compliance rate is medium, and the average value of the prepaid balance is low; (Low value winter type)
Cluster-5 (513 people): the average monthly gas consumption is high, the same-ratio growth rate of the average monthly gas consumption is low, the average selling price is medium, the gas consumption in four seasons is balanced, the gas consumption in winter is medium, the same-ratio growth rate of the gas consumption in winter is medium, the plan coincidence rate is low, and the average value of the prepayment balance is high; (high value stability)
Cluster-7 (574 people): the average monthly gas consumption is medium, the same-ratio increase rate of the average monthly gas consumption is high, the average selling price is low, the gas consumption in four seasons is balanced, the gas consumption in winter is medium, the gas consumption in winter is low in the same-ratio increase rate, the plan coincidence rate is high, and the average value of the prepayment balance is medium; (high potential of moderate value)
Cluster-8 (191 people): the average monthly gas consumption is medium, the same-ratio growth rate of the average monthly gas consumption is medium, the average selling price is low, the imbalance of the gas consumption in four seasons is mainly concentrated in the first quarter, the gas consumption in winter is high, the same-ratio growth rate of the gas consumption in winter is medium, the plan compliance rate is medium, and the average value of the prepaid balance is high; (Medium value winter type)
Wherein: cluster-6 has only one client and some features have large differences and are not analyzed as cluster groups.
The distribution of the seven customer groups in the industry is analyzed to obtain table 10:
number of customers | CNG | LNG factory | LNG loading | City gas | Power generation | Industrial fuel | Chemical fertilizer | Chemical engineering | Total of |
Cluster-1 | 86 | 370 | 9 | 55 | 11 | 15 | 546 | ||
Cluster-2 | 30 | 12 | 80 | 108 | 18 | 52 | 12 | 13 | 325 |
Cluster-3 | 1 | 8 | 7 | 1 | 17 | ||||
Cluster-4 | 8 | 6 | 43 | 69 | 6 | 15 | 1 | 148 | |
Cluster-5 | 22 | 3 | 68 | 341 | 13 | 51 | 9 | 6 | 513 |
Cluster-7 | 61 | 9 | 392 | 9 | 62 | 14 | 27 | 574 | |
Cluster-8 | 7 | 3 | 16 | 120 | 38 | 4 | 3 | 191 | |
Total of | 214 | 33 | 208 | 1408 | 55 | 280 | 51 | 65 | 2314 |
TABLE 10 customer base industry profiles
As can be understood from table 10 (in table 10, CNG indicates compressed natural gas and LNG indicates liquefied natural gas), the distribution of the industry in each customer group is not significantly different, indicating that the classification of the customer group is industry-independent.
S6: and setting different client strategies aiming at different client groups according to the value analysis result.
Where different client policies are set, for example: and the lost client is judged to be a lost client aiming at the clients who do not use gas for a period of time in the analysis result, and the lost client retrieval work can be pertinently developed.
The method also comprises the following steps before collecting the customer data of the natural gas customer: collecting a gas consumption data table, financial invoice data and financial data of a natural gas client; the method takes the recorded data of the gas consumption data table, the financial invoice data and the financial data table of the natural gas customer as the basic analysis data of the early-stage model development of the natural gas customer analysis method.
The natural gas customer analysis system based on the clustering algorithm of the embodiment of the invention, as shown in fig. 2, comprises a data acquisition unit, a clustering analysis unit and a value analysis unit. The data acquisition unit is used for acquiring customer data of a natural gas customer; the cluster analysis unit is used for carrying out cluster analysis on the customer data by using a cluster analysis method and dividing the natural gas customers into different groups of customer groups; the value analysis unit is used for carrying out value analysis on the customer base.
Further, the analysis system further includes a field design unit, and in combination with step S2 in the natural gas customer analysis method based on the clustering algorithm according to the embodiment of the present invention, the field design unit performs design of a data wide table field on the collected customer data and performs feature classification, where in this embodiment, the feature classification at least includes categories such as customer basic features, recent gas usage rules, seasonal preferences, planned gas usage conditions, historical trend changes, pressure reduction execution conditions, financial credit indicators, and customer statuses.
Furthermore, the analysis system further includes a data dimension reduction unit, and in combination with step S3 in the natural gas customer analysis method based on the clustering algorithm according to the embodiment of the present invention, the data dimension reduction unit performs dimension reduction on the data wide table fields.
The clustering analysis unit divides the natural gas customers into a plurality of groups of customer groups according to the clustering result characteristics, the contour values and the clustering chart by adopting a K-Means clustering analysis method.
Although the present invention has been described with reference to a preferred embodiment, it should be understood that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (12)
1. A natural gas customer analysis method based on a clustering algorithm is characterized by comprising the following steps:
collecting customer data of a natural gas customer;
carrying out cluster analysis on the customer data by using a cluster analysis method, and dividing the natural gas customers into different groups of customer groups;
and analyzing the value of a plurality of groups of the customer groups.
2. The natural gas customer analysis method based on clustering algorithm according to claim 1, characterized in that before the step of clustering analysis of the natural gas customer, further comprising the steps of:
and designing a data wide table field for the collected customer data.
3. The natural gas customer analysis method based on clustering algorithm according to claim 2, characterized in that after the step of designing data wide table fields and before the step of clustering analysis of the natural gas customers, further comprising the steps of:
and performing dimension reduction processing on the data wide table field.
4. The natural gas customer analysis method based on clustering algorithm according to claim 3, wherein in the step of performing dimension reduction processing on the data wide table fields, the dimension reduction processing is performed by using a factor analysis method, the number of data wide table fields with loads higher than a predetermined threshold value is reduced, and clustering analysis is performed using the reduced number of data wide table fields.
5. The natural gas customer analysis method based on clustering algorithm as claimed in claim 2, wherein in the step of designing data wide table field for the collected customer data, further comprising performing feature classification for the customer data.
6. The natural gas customer analysis method based on clustering algorithm according to claim 5, characterized in that the categories of the feature classification comprise at least customer basic features, recent gas usage rules, seasonal preferences, planned gas usage situations, historical trend changes, pressure reduction execution situations, financial credit indicators, customer status.
7. The natural gas customer analysis method based on the clustering algorithm according to claim 1, wherein in the step of dividing the natural gas customers into different customer groups by using the clustering analysis method, the natural gas customers are divided into a plurality of groups of customer groups according to the clustering result characteristics, the contour values and the clustering chart by using a K-Means clustering analysis method.
8. The natural gas customer analysis method based on clustering algorithm according to claim 1, characterized in that after the step of performing value analysis on the customer base, further comprising the steps of:
and setting different customer strategies aiming at different customer groups according to the value analysis result.
9. A natural gas customer analysis system based on a clustering algorithm, comprising:
the data acquisition unit is used for acquiring customer data of a natural gas customer;
the cluster analysis unit is used for carrying out cluster analysis on the customer data by using a cluster analysis method and dividing the natural gas customers into different groups of customer groups;
a value analysis unit for performing a value analysis on the customer base.
10. The natural gas customer analysis system based on clustering algorithm of claim 9, characterized in that the analysis system further comprises a field design unit for data wide table field design of the collected customer data.
11. The natural gas customer analysis system based on clustering algorithm of claim 10, characterized in that the analysis system further comprises a data dimension reduction unit for performing dimension reduction processing on the data wide table field.
12. The clustering algorithm-based natural gas customer analysis system of claim 9, wherein the cluster analysis unit performs cluster analysis using a K-Means cluster analysis method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110460875.XA CN113191802A (en) | 2021-04-27 | 2021-04-27 | Clustering algorithm-based natural gas customer analysis method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110460875.XA CN113191802A (en) | 2021-04-27 | 2021-04-27 | Clustering algorithm-based natural gas customer analysis method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113191802A true CN113191802A (en) | 2021-07-30 |
Family
ID=76979522
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110460875.XA Pending CN113191802A (en) | 2021-04-27 | 2021-04-27 | Clustering algorithm-based natural gas customer analysis method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113191802A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107133652A (en) * | 2017-05-17 | 2017-09-05 | 国网山东省电力公司烟台供电公司 | Electricity customers Valuation Method and system based on K means clustering algorithms |
US20170364795A1 (en) * | 2016-06-15 | 2017-12-21 | Akw Analytics Inc. | Petroleum analytics learning machine system with machine learning analytics applications for upstream and midstream oil and gas industry |
CN109685581A (en) * | 2019-01-03 | 2019-04-26 | 江苏方天电力技术有限公司 | A kind of large power customers electricity consumption behavior analysis method based on label clustering technology |
CN111159258A (en) * | 2019-12-31 | 2020-05-15 | 科技谷(厦门)信息技术有限公司 | Customer clustering implementation method based on cluster analysis |
CN111553720A (en) * | 2019-12-10 | 2020-08-18 | 北京理工大学 | User electricity utilization behavior analysis method based on improved k-means algorithm |
-
2021
- 2021-04-27 CN CN202110460875.XA patent/CN113191802A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170364795A1 (en) * | 2016-06-15 | 2017-12-21 | Akw Analytics Inc. | Petroleum analytics learning machine system with machine learning analytics applications for upstream and midstream oil and gas industry |
CN107133652A (en) * | 2017-05-17 | 2017-09-05 | 国网山东省电力公司烟台供电公司 | Electricity customers Valuation Method and system based on K means clustering algorithms |
CN109685581A (en) * | 2019-01-03 | 2019-04-26 | 江苏方天电力技术有限公司 | A kind of large power customers electricity consumption behavior analysis method based on label clustering technology |
CN111553720A (en) * | 2019-12-10 | 2020-08-18 | 北京理工大学 | User electricity utilization behavior analysis method based on improved k-means algorithm |
CN111159258A (en) * | 2019-12-31 | 2020-05-15 | 科技谷(厦门)信息技术有限公司 | Customer clustering implementation method based on cluster analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | A hybrid FCW-EMD and KF-BA-SVM based model for short-term load forecasting | |
CN110110881B (en) | Power customer demand prediction analysis method and system | |
CN108388955A (en) | Customer service strategies formulating method, device based on random forest and logistic regression | |
CN107784518A (en) | A kind of power customer divided method based on multidimensional index | |
CN112561730A (en) | Power supply service analysis method based on double-layer clustering and fuzzy comprehensive evaluation | |
CN108364191A (en) | Top-tier customer Optimum Identification Method and device based on random forest and logistic regression | |
CN112711616A (en) | User power consumption characteristic analysis method based on clustering algorithm | |
Zhong et al. | Research on electricity consumption behavior of electric power users based on tag technology and clustering algorithm | |
CN111178957B (en) | Method for early warning sudden increase of electric quantity of electricity consumption customer | |
CN112686473A (en) | Classification algorithm-based power consumption prediction method and prediction system | |
CN111798333A (en) | Energy utilization evaluation and electricity utilization safety analysis method and system | |
CN114565293A (en) | Evaluation method for providing long-period demand response capability by industrial load | |
CN113191802A (en) | Clustering algorithm-based natural gas customer analysis method and system | |
CN113344449A (en) | Method for predicting monthly-scale industrial water demand | |
CN110795610B (en) | Clustering-based power load analysis method | |
CN116701965A (en) | BIRCH clustering algorithm-based panoramic carbon representation method for enterprise users | |
CN116362584A (en) | Economic analysis method based on user side energy storage capacity configuration | |
CN114841832A (en) | Power consumer portrait label establishing method based on secondary clustering of power loads | |
Bazzazan et al. | Effects of targeting energy subsidies on domestic electricity demand in Iran | |
CN114444934A (en) | Enterprise sales periodic evaluation algorithm and tool application thereof | |
CN112001551B (en) | Ground and commercial power grid sales electricity quantity prediction method based on large-user electricity quantity information | |
CN113627821A (en) | Method and system for identifying abnormal electricity utilization based on electricity utilization behavior characteristics | |
CN114202403A (en) | Credit evaluation method and system based on electric power data | |
CN112100246A (en) | Customer electricity value mining method based on multi-dimensional graph code label | |
CN112381559A (en) | Tobacco retailer segmentation method based on unsupervised machine learning algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |