CN115759664A - User carbon portrait method based on electric energy use behaviors - Google Patents
User carbon portrait method based on electric energy use behaviors Download PDFInfo
- Publication number
- CN115759664A CN115759664A CN202211484481.9A CN202211484481A CN115759664A CN 115759664 A CN115759664 A CN 115759664A CN 202211484481 A CN202211484481 A CN 202211484481A CN 115759664 A CN115759664 A CN 115759664A
- Authority
- CN
- China
- Prior art keywords
- user
- electricity
- carbon
- label
- electric energy
- 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
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a user carbon portrait method based on electric energy use behaviors, which comprises the following steps: preprocessing user data; constructing a multi-dimensional user feature label model, and establishing a multi-source feature label system of a user from three dimensions of the user electricity utilization characteristic, the user low-carbon electricity utilization behavior and the user electricity utilization and carbon elimination characteristic; according to the time intervals, calculating various characteristic index comprehensive indexes; and (4) visually presenting the carbon portrait result according to the cluster of the user, and using the portrait result as a basis for a power company to set a user demand side response policy and low-carbon regulation and control. According to the invention, through analyzing the refined characteristics of the user, namely researching the carbon portrait of the user, the dominance of the implicit characteristics of the user is realized, and the electric power company is helped to make reasonable low-carbon regulation measures from the perspective of the production and elimination characteristics of the electric energy carbon emission of the user according to the low-carbon electricity consumption behavior characteristics of different residential users.
Description
Technical Field
The invention relates to a user carbon portrait method based on electric energy use behaviors, and belongs to the technical field of power supply and distribution.
Background
With the double-carbon construction target of 'carbon peak reaching and carbon neutralization' proposed in China, energy saving and carbon reduction become the current hot point problem. In recent years, the proportion of electricity consumption of residents in the whole society in China is continuously increased, and carbon dioxide generated by using electric energy in daily life of the residents cannot be ignored. By analyzing the electric energy using behaviors of the resident users, the adjustment of the electric energy using modes of the residents is the most direct and effective mode for reducing the carbon emission of the resident electricity consumption.
At present, with the popularization and function updating of smart power grids and the development of big data technology, the collection of multidimensional data of users for user behavior portrayal can be realized, wherein the carbon portrayal is portrayal based on the comprehensive carbon characteristics of the electric energy use behaviors of the users. The electric power company can optimize the demand side response model of the resident user by adjusting the marketing mode and the electric charge policy, and realize accurate regulation and control of the electric energy using behavior of the user, so that the purposes of energy conservation and emission reduction are achieved.
Disclosure of Invention
The invention aims to provide a user carbon portrait method based on electric energy use behaviors, aims to help an electric power company to set reasonable low-carbon regulation and control measures, performs tagging abstraction on attributes of a user from three aspects of power utilization characteristics of the user, low-carbon power utilization behaviors of the user and electric energy generation and carbon elimination characteristics of the user (the electric energy generation and carbon elimination characteristics of the user are comprehensive characteristics of regulation and control of carbon emission potential in the generation and consumption processes of the user), provides attribute tags for calculating users of various dimensions, and improves a k-means clustering algorithm to be applied to a user comprehensive carbon portrait method.
The purpose of the invention is realized by the following technical scheme:
a user carbon portrait method based on electric energy use behaviors comprises the following steps:
s1, user data preprocessing, comprising: collecting and screening the multidimensional behavior data of the user, and eliminating error values and wrong value user data;
s2, constructing a multi-dimensional user feature tag model, and establishing a multi-source feature tag system of the user from three dimensions of the electricity utilization characteristic of the user, the low-carbon electricity utilization behavior of the user and the carbon elimination characteristic of the electricity utilization of the user;
s3, calculating various characteristic index comprehensive indexes according to the time periods, wherein the various characteristic index comprehensive indexes comprise: the system comprises a user electricity utilization characteristic tag, a user low-carbon electricity utilization behavior tag and a user electricity utilization carbon elimination characteristic tag, wherein different residential users are divided into clusters with different attributes according to the user electricity utilization characteristic tag, the user low-carbon electricity utilization behavior tag and the user electricity utilization carbon elimination characteristic tag;
and S4, visually presenting the carbon portrait result according to the user cluster, and using the portrait result as a basis for setting a user demand side response policy and low-carbon regulation and control of the power company.
The present invention further includes the following preferred embodiments.
In the foregoing method for carbon portrait of user based on electric energy usage behavior, in step 1, the multidimensional behavior data of the user includes:
based on user multidimensional electric energy use behavior data collected by a non-user terminal, electricity utilization data of various loads identified by the non-user terminal are acquired by taking 15 minutes as one-time sampling time and taking days as a unit, wherein 96-point electricity utilization data of various loads of a user are acquired;
the load of the user home collected by the non-user-entry terminal comprises the following steps: refrigerator, long-time electric heating, short-time electric heating, variable frequency air conditioner, non-variable frequency air conditioner, electromagnetic oven, electric cooker, electric water heater, air conditioner, microwave oven, washing machine;
based on the user's charges of electricity data that marketing system gathered to day unit, include: the total electric charge of the user, the electric charge generated in the peak time period and the valley time period of the user;
the user network behavior statistical data obtained based on the online business hall, the mobile phone APP and the power grid customer service telephone comprise: the method comprises the steps that a user sets the number of times that the household air conditioner runs to an energy-saving temperature, the number of times that the user checks electricity charge using conditions, the number of times that the user checks electricity charge by logging in a mobile phone APP, the number of times that the user checks electricity charge by logging in a webpage edition of an online business hall and the number of times that the user checks electricity charge by dialing a power supply service hotline 95598;
in the method for describing the carbon of the user based on the electric energy use behavior, in step 2, the multi-dimensional user feature tags comprise a user electricity utilization characteristic tag, a user low-carbon electricity utilization behavior tag and a user electric energy generation carbon elimination characteristic tag.
In the step 2, the user electricity utilization characteristic label aims at representing the electricity utilization condition of the user, and the regulation and control potential of the user is mined by researching the load characteristic and the electricity utilization condition of the user; the user electricity characteristic tag comprises: the electric quantity characteristic label, the load characteristic label and the electric appliance composition characteristic label; wherein:
the electric quantity characteristic label L E Defined as the ratio of the amount of electricity used by the user to the average amount of electricity used by the user in the calculation period, as shown in the following equation:
in the formula, W i For the amount of power consumed by the user during the calculation period, W av The average power consumption of all users in the statistical range is obtained, and n is the number of the users in the statistical range;
the load characteristic label is defined as the maximum load power, the average load rate, the maximum load utilization hours and the peak-valley difference rate in the analysis time period, and is respectively expressed as:
maximum load power:
L max-Ti =P max-Ti (2)
average load factor:
maximum load utilization hours:
peak-to-valley difference rate:
in the formula, P Ti For type i adjustable load power, P max-Ti For maximum power in the statistical Ti time period, P av-Ti Is the average power, P, over a statistical Ti period max To count the maximum power in time, P min The minimum power in the statistical time is obtained;
the electric appliance constitutes a characteristic label, which is defined as the ratio of the total power of the load which can respond on the demand side to the reference power, and is shown as the following formula:
L app =(∑k i P Ti +∑k j P zj )/1000 (6)
in the formula, P Ti For type i adjustable load power, P zj For the class j reducible loads, k denotes the number of classes, k i For class i the number of loads, k, can be adjusted j The number of reducible loads for class j; the adjustable load is a load which does not stop running but can adjust power in the using process of a user, and the reducible load is a load which can interrupt running in the using process of the user;
in the step 2, the user low-carbon electricity consumption behavior tag aims to evaluate the interaction capacity of the user and the power grid from a subjective angle, so that the difficulty level of achieving the purpose of low carbon emission reduction by the power company through regulating and controlling the electricity consumption behavior of the user is reflected. The low-carbon electricity consumption behavior label for the user comprises: the system comprises a peak-valley electricity price sensitivity estimation label, an electricity utilization tendency estimation label and a user low-carbon electric energy use consciousness estimation label; wherein
The peak-valley electricity price sensitivity estimation label is defined as the ratio of the peak-valley electricity price and the electricity quantity change at the turning moment of the peak-valley electricity price, and is respectively expressed as:
peak-to-valley electricity charge ratio:
in the formula, C f Representing the peak time electricity charge, C, in the calculation period g Representing the off-hour electricity charge in the calculation period;
electric quantity change at the peak-valley electricity price turning moment:
in the formula, W i (t n + 1) represents the electricity quantity one hour after the transition of the peak-to-valley electricity price at night of day i, W i (t n -1) represents the electricity quantity one hour before the turning of the morning peak-valley electricity price on the ith day, T is the calculation period, L cs The sensitivity of the user to the electricity price is reflected;
the power utilization tendency estimation label is defined as a valley power coefficient and is expressed as:
in the formula, L v Indicating the amount of electricity used by the user during the valley period, L z The total power consumption of the user is represented, and the valley power coefficient reflects the power consumption tendency of the user in the valley period;
user low carbon electric energy uses consciousness to estimate, carries out data acquisition and analysis with other relevant network systems through electric power system marketing department, sets up the number of times of energy-conserving temperature operation with domestic air conditioner including the user, and the user looks over the number of times of charges of electricity in service behavior, and the user looks over the number of times of charges of electricity through logging on cell-phone APP, and the user is through the number of times of looking over charges of electricity of logging on business office webpage version on the net, and the user looks over the number of times of charges of electricity through dialling 95598:
L es =N esc +N ce (10)
N ce =N App +N net +N 598 (11)
in the formula, N esc Number of times of operation of setting home air conditioner to energy saving temperature for user, N ce Number of times of checking electricity charge usage for user, N App Number of times, N, that the user checks the electricity charge by logging in to the mobile phone APP net Number of times of checking electric charge for user by logging in web page version of online business hall, N 598 Number of times the user views the electricity charge, L, by dialing 95598 es Indicating a low carbon power usage awareness estimate, L es The larger the numerical value is, the more attention the user pays to the low-carbon electricity consumption behavior of the user, and the better the low-carbon electricity consumption awareness is.
In step 2, the user electric energy production and carbon consumption characteristic label aims to represent the potential size of reducing carbon dioxide emission by reasonably distributing and using electric energy in response to the relevant electric energy saving and carbon reduction policy formulated by the power grid in terms of using electric energy by residential users, and the user electric energy production and carbon consumption characteristic label comprises: the system comprises a user electric energy clean energy proportion label, a green electric appliance energy consumption proportion label, a user carbon emission characteristic label, a user carbon emission regulation and control coefficient label and a user effective substitute electric quantity label; wherein
The user electric energy clean energy ratio label L cl Defined as the percentage of clean energy in the user's electric energy used in the calculation cycle, as shown in the following formula:
in the formula, L p The amount of photovoltaic energy used in the electrical energy for the user, L t Usage of fossil power in electric energy for users, L cl The proportion of clean energy in the electric energy daily used by the user is reflected, and the value of the proportion is positively correlated with the carbon reduction potential of the user;
the green electric applianceEnergy consumption ratio label L g Defined as the ratio of the power consumption of the green household appliance in the appliances used by the user to the total power consumption of the user in the calculation period, as shown in the following formula:
in the formula, L ge Monthly consumption of consumer green appliances, L t Consumption of fossil power in electric energy for users, L p The energy consumption ratio of the green electric appliances of the user can reflect the potential of reducing carbon of the electric energy of the user.
The label L for the carbon emission characteristic of the electricity used by the user ce Defined as the ratio of the carbon emissions generated by the personal electricity of the user to the average carbon emissions generated by the user using the electric energy within the statistical sample range during the calculation period, as shown in the following formula:
L tc =L t ×a t (15)
L avc =L av ×a t (16)
in the formula, L tc Carbon emissions for consumer electricity utilization, L av Carbon emission, L, for all users in the community using electricity t Usage of fossil power in electric energy for users, L av Is the per-capita power consumption of users in the sample cell, a t The carbon emission coefficient corresponding to the electric power energy is 0.96kg/kwh; the user electricity consumption carbon emission characteristic label is the most direct expression of the potential of reducing carbon dioxide emission through electric energy regulation and control;
the user carbon emission regulation coefficient label L rce The calculation cycle is defined as the proportion of the remaining photovoltaic power in the power used by the user after the base load is satisfied to occupy the demand-responsive load, as shown in the following formula:
in the formula, L b Is the base load of the user, P Ti For adjustable load power of class i, P zj For class j power capable of load shedding, K i For the number of adjustable loads of class i, K j For the j-class quantity of reducible loads, the user carbon emission regulation coefficient reflects the potential that the clean energy used by the user can be used for participating in demand side response;
the user effectively replaces the electric quantity label E es Defined as the effective replacement electric quantity to be replaced by electric energy in the calculation period, as shown in the following formula:
in the formula, alpha rem For calculating the proportion of the clean energy consumption of the user in the period, M es Number of electric energy-substituting devices per household, E esm The method has the advantages that the replacement quantity of the electric energy replacement equipment in the calculation period is effectively replaced by the user in the family energy structure represented by the electric quantity label, the conventional replacement quantity of primary energy such as coal, petroleum and natural gas is replaced by the convenient, safe and low-carbon clean electric energy, and the low-carbon emission reduction potential of the user is objectively reflected.
In the foregoing method for carbon representation of a user based on electric energy usage behavior, in step 3, the method for calculating various comprehensive indicators of characteristic indexes includes: data standardization processing, various comprehensive index value analysis, a K-means clustering algorithm and user clustering analysis;
the data standardization processing comprises the following steps:
(1) Establishing a comprehensive regulation and control matrix at the same time period, wherein the comprehensive regulation and control matrix comprises three types of 15-dimensional labels: 6-dimensional L of user electricity characteristic label 1 -L 6 User low-carbon electricity consumption behavior label 4-dimensional L 7 -L 10 User electric energy production carbon elimination characteristic label 5-dimensional L 11 -L 15 I.e. the data samples for N users are represented as:
(2) Data of the same category for all users is represented as a column vector:
{L i (1),L i (2),…L i (15)}
(3) And (3) cleaning data, and performing data standardization treatment by adopting minimum-maximum standardization:
in the formula, L (j) min Denotes the minimum value of L (j), L (j) max Represents the maximum value of L (j);
the various types of comprehensive index value analysis comprise:
(1) User electricity characteristic label L rp
The user electricity consumption characteristic tag comprises a user electricity consumption characteristic tag L 1 -L 6 Total 6-dimensional data, and the integrated tag value L of the power consumption characteristics is obtained by the following equation rp In the formula w j The weight of each dimension of data:
(2) Electric energy regulation willingness label L rw
Electric energy regulation and control willingness label comprises user electricity utilization characteristic label L 7 -L 10 4-dimensional data, and the integrated tag value L of the power consumption characteristics is obtained by the following formula rw In the formula w j The weight of each dimension of data:
(3) Electric energy carbon-reducing potential labelL crp
The label of the electric energy carbon-reducing potential comprises a label L of the electric characteristics of the user 11 -L 15 5D data, the comprehensive label value L of the electricity utilization characteristic is obtained by the following formula crp In the formula w j The weight of each dimension of data:
wherein the weight value of each dimension is determined by an entropy weight method, L 1 -L 6 The weight of the weight is: 0.23, 0.25, 0.05, 0.06, 0.03, 0.39 7 -L 10 The weight of the weight is: 0.26, 0.14, 0.46 11 -L 15 The weight of the weight is: 0.01, 0.19, 0.09, 0.01, 0.70.
Analyzing and determining the cluster of the user based on a K-Means clustering algorithm, wherein:
characteristic label L for user electricity rp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m1 classes, and clustering central points L of the classes are respectively obtained rp (k);
Label L for electric energy regulation willingness rw (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m2 classes, and clustering center points L of the classes are respectively obtained rw (k);
Tag L for reducing carbon potential of electric energy crp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m3 classes, and clustering center points L of the classes are respectively obtained crp (k);
According to L rp (i)、L rw (i)、L crp (i) Combining the clustering center points of the three types of data to form m1 multiplied by m2 multiplied by m3 three-dimensional plane center points which respectively correspond to m1 multiplied by m2 multiplied by m3 clusters;
wherein m1= m2= m3=3.
In the method for visually presenting the carbon portrait result based on the electric energy use behavior, the step four, wherein the step of visually presenting the carbon portrait result refers to presenting the user carbon portrait result visually according to the user electricity consumption characteristic label, the user low-carbon electricity consumption behavior label, the user low-carbon electric energy production carbon elimination characteristic label and the user cluster determined by k-means clustering.
Compared with the prior art, the invention has the following beneficial technical effects:
according to the invention, through analyzing the refined characteristics of the user, namely researching the carbon portrait of the user, the dominance of the implicit characteristics of the user is realized, and the electric power company is helped to make reasonable low-carbon regulation measures from the perspective of the production and elimination characteristics of the electric energy carbon emission of the user according to the low-carbon electricity consumption behavior characteristics of different residential users.
Drawings
FIG. 1 is a basic flow diagram of a method for user carbon portrayal based on power usage behavior in accordance with the present invention;
FIG. 2 is a flow chart of the overall regulatory analysis of the present invention;
FIG. 3 is a visual presentation diagram of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific examples.
The basic flow of the user carbon representation method is shown in FIG. 1.
Firstly, collecting and screening electric energy use behavior data of a user, wherein the collected data comprises the following steps: based on multidimensional electric energy use behavior data of the user collected by the non-user-entry terminal, electricity data of various loads identified by the non-user-entry terminal are acquired by taking 15 minutes as one-time sampling time and taking days as a unit, wherein the electricity data of 96 points of various loads of the user are acquired;
the load of the user home collected by the non-user-entry terminal comprises the following steps: refrigerator, long-time electric heating, short-time electric heating, variable frequency air conditioner, non-variable frequency air conditioner, electromagnetic oven, electric cooker, electric water heater, air conditioner, microwave oven, washing machine;
based on the user's charges of electricity data that marketing system gathered to day unit, include: the total electric charge of the user, the electric charge generated in the peak time period and the valley time period of the user;
the user network behavior statistical data obtained based on the online business hall, the mobile phone APP and the power grid customer service telephone comprise: the method comprises the steps that a user sets the number of times that the household air conditioner runs to an energy-saving temperature, the number of times that the user checks electricity charge using conditions, the number of times that the user checks electricity charge by logging in a mobile phone APP, the number of times that the user checks electricity charge by logging in a webpage edition of an online business hall and the number of times that the user checks electricity charge by dialing a power supply service hotline 95598;
according to the existing user electric energy use data, the purpose of portrait and the user electric energy use attribute are combined, a label system of the user carbon portrait is designed, and the label system comprises: the method comprises the steps of designing a user electricity utilization characteristic label, a user low-carbon electricity utilization behavior label and a user electric energy generation carbon elimination characteristic label, and designing a specific calculation method of each dimension label.
The user electricity characteristic tag comprises: the electric quantity characteristic label, the load characteristic label and the electric appliance composition characteristic label; wherein the electric quantity characteristic label L B Defined as the ratio of the user electricity consumption to the average user electricity consumption in the calculation period, as shown in the following formula:
in the formula, W i For the amount of power consumed by the user during the calculation period, W av The average power consumption of all users in the statistical range is obtained, and n is the number of the users in the statistical range;
the load characteristic label is defined as the maximum load power, the average load rate, the maximum load utilization hours and the peak-valley difference rate in the analysis time period, and is respectively expressed as:
maximum load power:
L max-Ti =P max-Ti (2)
average load factor:
maximum load utilization hours:
peak-to-valley difference rate:
in the formula, P max-Ti For maximum power in the statistical Ti time period, P av-Ti Is the average power, P, over a statistical Ti period max To count the maximum power in time, P min The minimum power in the statistical time is obtained;
the electric appliance constitutes a characteristic label, which is defined as the ratio of the total power of the load which can respond on the demand side to the reference power, and is shown as the following formula:
L app =(∑k i P Ti +∑k j P zj )/1000 (6)
in the formula, P Ti For type i adjustable load power, P zj For the class j reducible loads, k denotes the number of classes, k i For class i the number of loads, k, can be adjusted j The number of loads can be shed for class j. The adjustable load refers to a load which does not stop running but can adjust power in the using process of a user, and the reducible load refers to a load which can interrupt running in the using process of the user.
The peak-valley electricity price sensitivity estimation label is defined as the ratio of the peak-valley electricity price and the electricity quantity change at the turning moment of the peak-valley electricity price, and is respectively expressed as:
peak-to-valley electricity charge ratio:
in the formula, C f Representing the peak time electricity charge, C, in the calculation period g Representing the off-hour electricity charge in the calculation period;
electric quantity change at the time of peak-valley electricity price turning:
in the formula, W i (t n + 1) represents the electricity quantity one hour after the transition of the peak-to-valley electricity price at night of day i, W i (t n -1) represents the electricity quantity one hour before the turning of the morning peak-valley electricity price on the ith day, T is the calculation period, L cs Reflecting the user's sensitivity to electricity prices.
The electric propensity estimation tag is defined as the valley power coefficient, expressed as:
in the formula, L v Indicating the amount of electricity used by the user during the valley period, L z The total power consumption of the user is represented, and the valley power coefficient reflects the power utilization tendency of the user in the valley period.
The user low-carbon electric energy use awareness estimation label comprises the following components: the user sets up domestic air conditioner to the number of times of energy-conserving temperature operation, and the user looks over the number of times of charges of electricity in service behavior, and the user looks over the number of times of charges of electricity through logging in cell-phone APP, and the user looks over the number of times of charges of electricity through logging in online business office webpage edition, and the user looks over the number of times of charges of electricity through dialling 95598:
L es =N esc +N ce (10)
N ce =N App +N net +N 598 (11)
in the formula, N esc Number of times of operation of setting home air conditioner to energy saving temperature for user, N ce Number of times of checking electricity charge usage for user, N App Number of times, N, that the user checks the electricity charge by logging in to the mobile phone APP net For the number of times of checking the electricity charge of the user by logging in the web page version of the online business hall, N 598 The number of times the electricity charge is viewed for the user by dialing 95598. L is es Indicating a low carbon power usage awareness estimate, L es The larger the numerical value is, the more attention the user pays to the low-carbon electricity consumption behavior of the user, and the better the low-carbon electricity consumption awareness is.
The label for the carbon elimination characteristic of the user electric energy production comprises the following components: the system comprises a user electric energy clean energy proportion label, a green electric appliance energy consumption proportion label, a user carbon emission characteristic label, a user carbon emission regulation and control coefficient label and a user effective substitute electric quantity label; wherein
User electric energy clean energy ratio label L cl Defined as the percentage of clean energy in the user's electric energy used in the calculation cycle, as shown in the following formula:
in the formula, L p The amount of photovoltaic energy used in the electrical energy for the user, L t Usage of fossil power in electric energy for users, L cl The proportion of clean energy in the electric energy daily used by the user is reflected, and the value of the proportion is positively correlated with the carbon reduction potential of the user.
Green electric appliance energy consumption ratio label L g Defined as the ratio of the power consumption of the green household appliance in the appliances used by the user to the total power consumption of the user in the calculation period, as shown in the following formula:
in the formula, L ge Monthly consumption of green appliances for the user, L t Consumption of fossil power in electric energy for users, L p For use by a userThe utilization amount of photovoltaic energy in the electric energy and the energy consumption proportion of green electric appliances of users can reflect the potential of reducing carbon of the electric energy of the users.
Consumer carbon emission characteristic label L ce Defined as the ratio of the carbon emissions generated by the personal electricity of the user to the average carbon emissions generated by the user using the electric energy within the statistical sample range during the calculation period, as shown in the following formula:
L tc =L t ×a t (15)
L avc =L av ×a t (16)
in the formula, L tc Carbon emissions for consumer electricity utilization, L av Carbon emission, L, for all users in the community using electricity t Usage of fossil power in electric energy for users, L av Is the electric energy consumption per capita of the users in the sample cell, a t The carbon emission coefficient corresponding to the electric power energy source is 0.96kg/kwh. The user's electricity consumption carbon emission characteristic label is the most direct manifestation of the potential of users to reduce carbon dioxide emission through electric energy regulation.
User carbon emission regulation coefficient label L rce The remaining energy of the photovoltaic energy used by the user in the calculation cycle after the base load is met occupies the proportion of the demand-responsive load, which is defined as follows:
in the formula, L b Is the base load of the user, P Ti For adjustable load power of class i, P zj For class j power at which load is curtailed, K i For the number of adjustable loads of class i, K j For class j curtailable load numbers, the user carbon emission control factor reflects the potential that clean energy used by the user can be used to participate in demand side responses.
User effectively replaces electric quantity label E es Defined as the effective replacement electric quantity to be replaced by electric energy in the calculation period, as shown in the following formula:
in the formula, alpha rem For calculating the proportion of the clean energy consumption of the user in the period, M es Number of electric energy-substituting devices for the user's home esm The method has the advantages that the replacement quantity of the electric energy replacement equipment in the calculation period is effectively replaced by the user in the family energy structure represented by the electric quantity label, the conventional replacement quantity of primary energy such as coal, petroleum and natural gas is replaced by the convenient, safe and low-carbon clean electric energy, and the low-carbon emission reduction potential of the user is objectively reflected.
After the specific calculation method of each dimension label system is determined, data processing is performed on the sub-labels in each label system to obtain each dimension label as shown in fig. 2, and then the carbon portrait is comprehensively presented according to the steps shown in fig. 2.
Firstly, standardizing data of each dimension sub-label, expressing the data of each dimension label obtained by calculation by using a data sample matrix, establishing a comprehensive regulation and control matrix in the same time period, expressing the same data by using vectors, then cleaning the data, removing the data which does not meet the requirements, and finally standardizing the cleaned data.
(1) Establishing a comprehensive regulation and control matrix at the same time period, wherein the comprehensive regulation and control matrix comprises three types of 14-dimensional labels: 6-dimensional L of user electricity characteristic label 1 -L 6 User low-carbon electricity consumption behavior label 4-dimensional L 7 -L 10 User electric energy production carbon elimination characteristic label 5-dimensional L 11 -L 15 I.e. the data samples for N users are represented as:
(2) Data of the same category for all users is represented as a column vector:
{L i (1),L i (2),…L i (15)}
(3) And (3) cleaning data, and performing data standardization treatment by adopting minimum-maximum standardization:
in the formula, L (j) min Denotes the minimum value of L (j), L (j) max Represents the maximum value of L (j).
After the data standardization processing is completed, various comprehensive index analyses need to be performed on each dimension sub-label, and the various comprehensive index analyses include:
(1) User electricity characteristic label L rp
The user electricity characteristic label comprises a user electricity characteristic label L 1 -L 6 Total 6-dimensional data, and the integrated tag value L of the power consumption characteristics is obtained by the following equation rp In the formula w j Is the weight of each dimension of data.
(2) Electric energy regulation willingness label L rw
Electric energy regulation and control willingness label comprises user electricity utilization characteristic label L 7 -L 10 4-dimensional data, and a comprehensive tag value L of the power consumption characteristics is obtained by the following formula rw In the formula w j Is the weight of each dimension of data.
(3) Electric energy carbon-reducing potential label L crp
Electric energy regulation and control willingness label comprises user electricity utilization characteristic label L 11 -L 15 5D data, the comprehensive label value L of the electricity utilization characteristic is obtained by the following formula crp In the formula w j For data of various dimensionsThe weight of (2).
Wherein the weight value of each dimension is determined by an entropy weight method, L 1 -L 6 The weight of the weight is: 0.23, 0.25, 0.05, 0.06, 0.03, 0.39 7 -L 10 The weight of the weight is: 0.26, 0.14, 0.46 11 -L 15 The weight of (A) is as follows: 0.01, 0.19, 0.09, 0.01, 0.70.
After the analysis of various comprehensive indexes is finished, determining the cluster of the user by using a k-means clustering algorithm, firstly clustering data according to a classification requirement by using Euclidean distance and combining with a k-means clustering idea, then solving the clustering center points of various data, and judging the cluster of the user according to each clustering center point. The specific process is as follows:
tag L for electricity characteristics of user rp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m1 classes, and clustering center points L of the various classes are respectively obtained rp (k);
Label L for electric energy regulation willingness rw (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m2 classes, and clustering center points L of the classes are respectively obtained rw (k);
Tag L for reducing carbon potential of electric energy crp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m3 classes, and clustering central points L of the classes are respectively obtained crp (k);
According to L rp (i)、L rw (i)、L crp (i) Combining the clustering central points of the three types of data to form m1 x m2 x m3 three-dimensional plane central points which respectively correspond to m1 x m2 x m3 clusters;
after the cluster determination of the user is completed through the k-means clustering algorithm, the user cluster analysis is needed to finally determine the cluster of the user.
Firstly, the three-dimensional labels are divided into three categories of low, medium and high according to the three-dimensional labels, then the clustering sequence is determined, the characteristics of the user with higher clustering are more beneficial to carbon reduction regulation, and finally the user carbon image is comprehensively presented after the clustering of the user is determined according to the clustering sequence as shown in fig. 3.
When the user electricity consumption characteristic label is constructed, the method divides each day into four periods of early peak, noon peak, late peak and night valley according to typical power grid load, and respectively corresponds to T1 period: 6-00-10, period T2: 10-00, period T3: 16-00, period T4: 21:00-6:00. The method is characterized in that 27 clusters are divided according to different users according to the electricity utilization characteristics, the low-carbon electricity utilization behavior characteristics and the electric energy generation and carbon elimination characteristics of the users, three-dimensional comprehensive presentation is carried out on various users, and the positions of the various users in a three-dimensional coordinate system, the number of the various users and the approximate proportion of the various users can be clearly seen through portrayal. The power company can reasonably make a policy to regulate and control the user behavior according to the user characteristics displayed by the portrait, and the purposes of energy conservation and emission reduction are achieved.
The invention maps the behaviors of various users to the two-dimensional plane of the electric energy generation and carbon elimination characteristics of the users and the electric power utilization characteristics of the users, the two-dimensional plane of the electric energy generation and carbon elimination characteristics of the users and the two-dimensional plane of the low-carbon electric power utilization characteristics of the users, so that the positions of the users in the planes can be clearly seen, the control variable analysis of the characteristics of the users in all aspects is facilitated for an electric power company, and the carbon reduction regulation and control strategy is refined.
The invention can help the power company to objectively and subjectively master the carbon characteristics and the regulation potential of each user in the use of the electric energy and the interaction and regulation capability of the user and the power grid in each typical time period, thereby being capable of more reasonably formulating the electric energy regulation and control scheme at the user side, regulating and guiding the carbon dioxide discharged by the electric energy daily used by residents in the production and use processes, and finally achieving the purpose of reducing the carbon dioxide discharge in the use of the electric energy of the residents.
In addition to the above embodiments, the present invention may have other embodiments, and any technical solutions formed by equivalent substitutions or equivalent transformations fall within the scope of the claims of the present invention.
Claims (10)
1. A user carbon portrait method based on electric energy use behaviors is characterized by comprising the following steps:
s1, preprocessing user data, including: collecting and screening the multidimensional behavior data of the user, and eliminating error values and wrong value user data;
s2, constructing a multi-dimensional user feature tag model, and establishing a multi-source feature tag system of the user from three dimensions of the electricity utilization characteristic of the user, the low-carbon electricity utilization behavior of the user and the carbon elimination characteristic of the electricity utilization of the user;
s3, calculating various characteristic index comprehensive indexes according to the time periods, wherein the various characteristic index comprehensive indexes comprise: the system comprises a user electricity utilization characteristic tag, a user low-carbon electricity utilization behavior tag and a user electricity utilization carbon elimination characteristic tag, wherein different residential users are divided into clusters with different attributes according to the user electricity utilization characteristic tag, the user low-carbon electricity utilization behavior tag and the user electricity utilization carbon elimination characteristic tag;
and S4, visually presenting the carbon portrait result according to the user cluster, and using the portrait result as a basis for setting a user demand side response policy and low-carbon regulation and control of the power company.
2. The method as claimed in claim 1, wherein the step S1 of using the multidimensional behavior data of the user comprises: based on user multidimensional electric energy use behavior data collected by a non-user terminal, electricity utilization data of various loads identified by the non-user terminal are acquired by taking 15 minutes as one-time sampling time and taking days as a unit, wherein 96-point electricity utilization data of various loads of a user are acquired; based on the user's charges of electricity data that marketing system gathered to day unit, include: the total electricity rate of the user, the electricity rate generated during the peak period and the valley period of the user.
3. The method as claimed in claim 1, wherein in step S2, the user electricity consumption feature tag includes: the electric quantity characteristic label, the load characteristic label and the electric appliance composition characteristic label; wherein:
the electric quantity characteristic label L E Defined as the ratio of the user electricity consumption to the average user electricity consumption in the calculation period, as shown in the following formula:
in the formula, W i For the amount of power consumed by the user during the calculation period, W av The average power consumption of all users in the statistical range is obtained, and n is the number of the users in the statistical range;
the load characteristic label is defined as the maximum load power, the average load rate, the maximum load utilization hours and the peak-valley difference rate in the analysis time period, and is respectively expressed as:
maximum load power:
L max-Ti =P max-Ti (2)
average load factor:
maximum load utilization hours:
peak-to-valley difference rate:
in the formula, P Ti For type i adjustable load power, P max-Ti For maximum power in the statistical Ti time period, P av-Ti Is the average power, P, over a statistical Ti period max To count the maximum power in time, P min The minimum power in the statistical time is obtained;
the electric appliance constitutes a characteristic label, which is defined as the ratio of the total power of the load which can respond on the demand side to the reference power, and is shown as the following formula:
L app =(∑k i P TI +∑k j P zj )/1000 (6)
in the formula, P Ti For type i adjustable load power, P zj For the class j reducible loads, k denotes the number of classes, k i For class i the number of loads, k, can be adjusted j The number of reducible loads for class j; the adjustable load is a load which does not stop running but can adjust power in the using process of a user, and the reducible load is a load which can interrupt running in the using process of the user.
4. The method for representing a carbon image of a user based on electric energy usage behaviors as claimed in claim 1, wherein the step S2 is performed by using a low carbon electricity usage behavior tag of the user, comprising: the system comprises a peak-valley electricity price sensitivity estimation label, an electricity utilization tendency estimation label and a user low-carbon electric energy use consciousness estimation label; wherein:
the peak-valley electricity price sensitivity estimation label is defined as the ratio of the peak-valley electricity price and the electricity quantity change at the turning moment of the peak-valley electricity price, and is respectively expressed as:
peak-to-valley electricity charge ratio:
in the formula, C f To representCalculating the peak time electricity charge, C, in the period g Representing the off-hour electricity charge in the calculation period;
electric quantity change at the time of peak-valley electricity price turning:
in the formula, W t (t n + 1) represents the electricity quantity one hour after the transition of the peak-to-valley electricity price at night of day i, W i (t n -1) represents the electricity quantity one hour before the turning of the morning peak-valley electricity price on the ith day, T is the calculation period, L cs The sensitivity of the user to the electricity price is reflected;
the power utilization tendency estimation label is defined as a valley power coefficient and is expressed as:
in the formula, L v Indicating the amount of electricity used by the user during the valley period, L z The total power consumption of the user is represented, and the valley power coefficient reflects the power consumption tendency of the user in the valley period;
user low carbon electric energy uses consciousness to estimate, carries out data acquisition and analysis with other relevant network systems through electric power system marketing department, sets up the number of times of energy-conserving temperature operation with domestic air conditioner including the user, and the user looks over the number of times of charges of electricity in service behavior, and the user looks over the number of times of charges of electricity through logging on cell-phone APP, and the user is through the number of times of looking over charges of electricity of logging on business office webpage version on the net, and the user looks over the number of times of charges of electricity through dialling 95598:
L es =N esc +N ce (10)
N ce =N App +N net +N 598 (11)
in the formula, N esc Number of times of operation of setting home air conditioner to energy saving temperature for user, N ce Number of times of checking electricity charge usage for user, N App Checking electricity for user by logging in mobile phone APPNumber of charges, N net For the number of times of checking the electricity charge of the user by logging in the web page version of the online business hall, N 598 Number of times the user views the electricity charge, L, by dialing 95598 es Indicating a low carbon power usage awareness estimate, L es The larger the numerical value is, the more attention the user pays to the low-carbon electricity consumption behavior of the user, and the better the low-carbon electricity consumption awareness is.
5. The method for user carbon portrayal based on electric energy use behaviors as claimed in claim 1, wherein in step S2, a user electric energy clean energy proportion label, a green electric appliance energy consumption proportion label, a user electric carbon emission characteristic label, a user carbon emission control coefficient label and a user effective substitute electric quantity label are included; wherein:
the user electric energy clean energy ratio label L cl Defined as the percentage of clean energy in the user's electric energy used in the calculation cycle, as shown in the following formula:
in the formula, L p Amount of photovoltaic energy used in the electric energy for the user, L t Usage of fossil power in electric energy for users, L cl The proportion of clean energy in the electric energy daily used by the user is reflected, and the value of the proportion is positively correlated with the carbon reduction potential of the user;
the green electric appliance energy consumption proportion label L g Defined as the ratio of the power consumption of the green household appliance in the appliances used by the user to the total power consumption of the user in the calculation period, as shown in the following formula:
in the formula, L ge Monthly consumption of consumer green appliances, L t Usage of fossil power in electric energy for users, L p In electric energy for consumer useThe consumption of the energy of the photovoltaic energy and the energy consumption proportion of the green electric appliances of the user reflect the potential of reducing carbon of the electric energy of the user;
the label L for the carbon emission characteristic of the electricity used by the user ce Defined as the ratio of the carbon emissions generated by the personal electricity of the user to the average carbon emissions generated by the user using the electric energy within the statistical sample range during the calculation period, as shown in the following formula:
L tc =L t ×a t (15)
L avc =L av ×a t (16)
in the formula, L tc Carbon emissions for consumer electricity utilization, L av Carbon emission, L, for all users in the community using electricity t Usage of fossil power in electric energy for users, L av Is the electric energy consumption per capita of the users in the sample cell, a t Carbon emission coefficient corresponding to the electric power energy; the user electricity consumption carbon emission characteristic label is the most direct expression of the potential of reducing carbon dioxide emission through electric energy regulation and control;
the user carbon emission regulation coefficient label L rce The calculation cycle is defined as the proportion of the remaining photovoltaic power in the power used by the user after the base load is satisfied to occupy the demand-responsive load, as shown in the following formula:
in the formula, L b Is the base load of the user, P Ti For adjustable load power of class i, P zj For class j power at which load is curtailed, K i For number of adjustable loads of class i, K j For the j-type reducible load number, the user carbon emission regulation coefficient reflects the potential that clean energy used by the user can be used for participating in demand side response;
the user effectively replaces the electric quantity label E es Defined as the effective replacement electric quantity to be replaced by electric energy in the calculation period, as shown in the following formula:
in the formula, alpha rem For calculating the proportion of the clean energy consumption of the user in the period, M es Number of electric energy-substituting devices for the user's home esm The method has the advantages that the replacement quantity of the electric energy replacement equipment in the calculation period is effectively replaced by the user in the family energy structure represented by the electric quantity label, the conventional replacement quantity of primary energy such as coal, petroleum and natural gas is replaced by the convenient, safe and low-carbon clean electric energy, and the low-carbon emission reduction potential of the user is objectively reflected.
6. The method as claimed in claim 1, wherein the step S3 of calculating the comprehensive index of each type of characteristic index includes: data standardization processing, various comprehensive index value analysis, a K-means clustering algorithm and user clustering analysis.
7. The method as claimed in claim 6, wherein the data normalization process comprises:
(1) Establishing a comprehensive regulation and control matrix at the same time period, wherein the comprehensive regulation and control matrix comprises three types of 15-dimensional labels: 6-dimensional L of user electricity characteristic label 1 —L 6 User low-carbon electricity consumption behavior label 4-dimensional L 7 —L 10 User electric energy production carbon elimination characteristic label 5-dimensional L 11 —L 15 I.e. the data samples for N users are represented as:
(2) Data of the same category for all users is represented as a column vector:
{L i (1),L i (2),…L i (15)}
(3) And (3) cleaning data, and performing data standardization treatment by adopting minimum-maximum standardization:
in the formula, L (j) min Denotes the minimum value of L (j), L (j) max Represents the maximum value of L (j).
8. The method as claimed in claim 6, wherein the analysis of the various types of comprehensive index values includes:
(1) User electricity characteristic label L rp
The user electricity characteristic label comprises a user electricity characteristic label L 1 —L 6 Total 6-dimensional data, and the comprehensive tag value L of the power consumption characteristics is obtained by the following formula rp In the formula w j The weight of each dimension of data:
(2) Electric energy regulation willingness label L rw
Electric energy regulation and control willingness label comprises user electricity utilization characteristic label L 7 —L 10 4-dimensional data, and the integrated tag value L of the power consumption characteristics is obtained by the following formula rw In the formula w j The weight of each dimension of data:
(3) Electric energy carbon-reducing potential label L crp
The label of the electric energy carbon-reducing potential comprises a label L of the electric characteristics of the user 11 —L 15 5D data, the comprehensive label value L of the electricity utilization characteristic is obtained by the following formula crp In the formula w j The weight of each dimension of data:
wherein the weight value of each dimension is determined by an entropy weight method, L 1 —L 6 The weight of (A) is as follows: 0.23, 0.25, 0.05, 0.06, 0.03, 0.39 7 —L 10 The weight of the weight is: 0.26, 0.14, 0.46 11 —L 15 The weight of the weight is: 0.01, 0.19, 0.09, 0.01, 0.70.
9. The method as claimed in claim 6, wherein the K-Means based clustering algorithm analyzes and determines the cluster to which the user belongs, and wherein:
tag L for electricity characteristics of user rp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m1 classes, and clustering center points L of the various classes are respectively obtained rp (k);
Label L for electric energy regulation willingness rw (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m2 classes, and clustering center points L of the classes are respectively obtained rw (k);
Tag L for reducing carbon potential of electric energy crp (i) Clustering thought of a K-means clustering algorithm based on Euclidean distance is clustered into m3 classes, and clustering center points L of the classes are respectively obtained crp (i);
According to L rp (i)、L rw (i)、L crp (i) Combining the clustering center points of the three types of data to form m1 multiplied by m2 multiplied by m3 three-dimensional plane center points which respectively correspond to m1 multiplied by m2 multiplied by m3 clusters;
wherein m1= m2= m3=3.
10. The method as claimed in claim 1, wherein the step S4 of presenting the user carbon representation is to present the user electricity consumption characteristic label, the user low carbon electricity consumption behavior label, the user electricity generation and carbon elimination characteristic label, and the determined cluster to which each user belongs, which are calculated according to the collected user electricity consumption behavior data, in a visual manner.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211484481.9A CN115759664A (en) | 2022-11-24 | 2022-11-24 | User carbon portrait method based on electric energy use behaviors |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211484481.9A CN115759664A (en) | 2022-11-24 | 2022-11-24 | User carbon portrait method based on electric energy use behaviors |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115759664A true CN115759664A (en) | 2023-03-07 |
Family
ID=85337203
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211484481.9A Pending CN115759664A (en) | 2022-11-24 | 2022-11-24 | User carbon portrait method based on electric energy use behaviors |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115759664A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118152830A (en) * | 2024-05-09 | 2024-06-07 | 国网山东省电力公司营销服务中心(计量中心) | User carbon emission characteristic image drawing method and system based on mean value clustering algorithm |
-
2022
- 2022-11-24 CN CN202211484481.9A patent/CN115759664A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118152830A (en) * | 2024-05-09 | 2024-06-07 | 国网山东省电力公司营销服务中心(计量中心) | User carbon emission characteristic image drawing method and system based on mean value clustering algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Association rule mining based quantitative analysis approach of household characteristics impacts on residential electricity consumption patterns | |
Luo et al. | Social information filtering-based electricity retail plan recommender system for smart grid end users | |
Yang et al. | A framework of customizing electricity retail prices | |
Ozawa et al. | Determining the relationship between a household’s lifestyle and its electricity consumption in Japan by analyzing measured electric load profiles | |
Fu et al. | Clustering-based short-term load forecasting for residential electricity under the increasing-block pricing tariffs in China | |
CN106815677A (en) | The recognition methods of non-intrusion type load and device | |
CN108776939A (en) | The analysis method and system of user power utilization behavior | |
CN109345409A (en) | A kind of resident's comprehensive energy efficiency management method based on bandwidth carrier | |
Klingler et al. | Residential photovoltaic self-consumption: Identifying representative household groups based on a cluster analysis of hourly smart-meter data | |
CN115759664A (en) | User carbon portrait method based on electric energy use behaviors | |
CN110944413A (en) | Electric heat subdivision method based on historical load identification data under cloud edge cooperative architecture | |
Xu et al. | Peak shaving potential analysis of distributed load virtual power plants | |
Waczowicz et al. | Demand response clustering-how do dynamic prices affect household electricity consumption? | |
CN113935856A (en) | Method and device for controlling electric energy, processor and computer readable storage medium | |
CN113362103A (en) | Method for customizing peak-valley time-of-use electricity price of power grid through user electricity consumption behavior characteristics | |
Gong et al. | Analysis of electricity consumption behavior of air conditioning based on the perspective of power demand response | |
Chen et al. | Research on residents' electricity behaviour analysis and control strategy optimisation | |
Lang et al. | Analysis of residents differentiated power consumption behavior based on load classification | |
Liu et al. | Recognition model of air conditioning electricity consumption behaviors for demand response based on k-means clustering | |
CN111860574A (en) | Power grid user grouping method and device, storage medium and electronic equipment | |
Chin et al. | Household electricity load disaggregation based on low resolution smart meter data | |
Chen et al. | Review on Smart Meter Data Clustering and Demand Response Analytics | |
Jangid et al. | Load Profile Segmentation of Various Load Categories using Clustering | |
CN113822343B (en) | Group renting room identification method based on fine-grained energy consumption data | |
Shen et al. | Classification and Potential Evaluation of Residential Users in Demand Response Based on NILM Data |
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 |