A power retail package pushing method and system based on customer portrait

Abstract

The application discloses a kind of based on customer portrait's electric power retail package push method and system, it is related to electric power market technical field.The method includes: constructing the customer peak clipping and valley filling portrait model based on DDPM, the historical load of customer is mined, the peak clipping and valley filling characteristics of customer are extracted, and peak clipping and valley filling time difference model and peak clipping and valley filling electric quantity difference model are constructed in combination with grid peak value characteristics, valley value characteristics, according to the height of peak clipping and valley filling time difference and the height of peak clipping and valley filling electric quantity difference, four kinds of electric power retail package are formulated respectively;Construct the electric power retail package judgment model based on GPT, judge the peak clipping and valley filling time difference and peak clipping and valley filling electric quantity difference of customer, according to the judgment result of GPT selects to push one of package one to package four to customer.The application can realize according to the peak clipping and valley filling characteristics of customer, and the appropriate electric power retail package is customized and pushed to user, improves the accuracy of push, improves the enthusiasm of user participation electric power package activity.

Description

A method and system for recommending electricity retail packages based on customer profiles Technical Field

[0001] This invention relates to the field of electricity market technology, specifically to a method and system for pushing electricity retail packages based on customer profiles. Background Technology

[0002] With the gradual opening of the electricity market and the application of intelligent technologies, the electricity retail industry is facing increasingly fierce market competition. To improve the service quality and customer satisfaction of power companies, personalized electricity package recommendations based on user profiles have become an important development trend. Through accurate user profiling, electricity retailers can provide more personalized package options based on customers' electricity needs and habits, enhancing user stickiness and satisfaction. However, existing electricity retail package recommendation methods still have a number of shortcomings and urgently need improvement.

[0003] First, traditional user profiling methods rely heavily on simple statistical information, such as basic user information and electricity consumption. However, customer electricity consumption behavior is typically volatile, and this simple statistical information cannot fully reflect the actual electricity demand of customers at different times of day (such as peak and off-peak periods). It is difficult to capture the deeper characteristics of users in peak shaving and valley filling, making it difficult to accurately match the recommended electricity packages with the actual needs of users.

[0004] Secondly, existing pricing strategies fail to adequately consider the varying electricity demands of customers during specific time periods. For instance, customers may face higher electricity prices during peak demand periods and lower prices during off-peak periods. Existing methods often ignore these temporal differences and cannot dynamically adjust based on customers' peak-shaving and valley-filling characteristics. Consequently, recommended pricing strategies fail to optimize electricity consumption and may even increase users' electricity costs.

[0005] Furthermore, traditional methods of recommending electricity packages lack targeting and personalization. Some existing technologies only recommend packages based on basic user information (such as monthly electricity consumption and electricity costs), failing to deeply analyze changes in customer electricity demand at different times, and failing to effectively match customer load fluctuations with grid load. Consequently, they cannot provide truly personalized and optimized electricity package recommendations, resulting in a poor user experience.

[0006] Furthermore, current electricity package recommendation methods lack dynamic adjustment and real-time optimization capabilities. With changes in the electricity market and rapid fluctuations in user demand, traditional methods cannot respond quickly to these changes. Power companies struggle to adjust their package recommendation strategies based on real-time data, resulting in package recommendations lagging behind market and customer needs, thus impacting the optimal utilization of electricity resources and customer satisfaction.

[0007] Therefore, a new method and system for pushing electricity retail packages is needed to deeply explore users' electricity demand patterns, including peak shaving and valley filling characteristics, to provide electricity retailers with more refined user profiles, better meet the grid's peak shaving and valley filling needs, and improve the attractiveness and conversion rate of the push. Summary of the Invention

[0008] To address the problems of overly simplistic user profiles, neglect of time-time differences, lack of personalized recommendations, and lack of dynamic adjustment functions in existing technologies, this invention proposes a method and system for pushing electricity retail packages based on customer profiles. By conducting in-depth analysis of customers' historical load, extracting peak-shaving and valley-filling features of customers' electricity consumption behavior based on DDPM technology, and constructing different electricity package models in conjunction with the load characteristics of the power grid, customized electricity packages are pushed through GPT judgment.

[0009] To achieve the above objectives, the present invention employs the following technical solution:

[0010] On the one hand, this invention proposes a method for recommending electricity retail packages based on customer profiles, the method comprising:

[0011] Construct a customer peak shaving and valley filling profile model based on DDPM, including a customer peak value feature profile model and a customer valley value feature profile model, to mine the customer's historical load, extract the customer's peak shaving and valley filling features, and realize the profile of the customer's peak shaving and valley filling features.

[0012] Based on the peak and valley characteristics of the power grid and the aforementioned customer peak-shaving and valley-filling profile model, a peak-shaving and valley-filling time difference model and a peak-shaving and valley-filling electricity difference model are constructed. Based on the magnitude of the peak-shaving and valley-filling time difference and the magnitude of the peak-shaving and valley-filling electricity difference, four electricity retail packages are formulated, specifically:

[0013] Package 1: Electricity retail package when the difference between peak shaving and valley filling time is high, and the difference between peak shaving and valley filling electricity volume is also high;

[0014] Package Two: Electricity retail package when the peak shaving and valley filling time difference is high and the peak shaving and valley filling electricity difference is low;

[0015] Package 3: Electricity retail package when the peak shaving and valley filling time difference is low and the peak shaving and valley filling electricity difference is high;

[0016] Package Four: Electricity retail packages with low peak-shaving and valley-filling time differences and low peak-shaving and valley-filling electricity differences;

[0017] A GPT-based electricity retail package judgment model is constructed, including a peak shaving and valley filling time difference judgment model and a peak shaving and valley filling electricity difference judgment model. The peak shaving and valley filling time difference and peak shaving and valley filling electricity difference of customers are judged, and one of the packages from package one to package four is selected to be pushed to customers based on the judgment results of GPT.

[0018] As a preferred embodiment of the present invention, the customer peak feature profiling model is based on the DDPM algorithm to mine the customer's historical load and extract the customer peak features, which are expressed as follows:

[0019]

[0020] The peak customer profile is represented as follows:

[0021]

[0022] In the formula, For customer peak characteristics; This represents a customer peak feature profile model based on DDPM; P L (t) represents the customer's historical load; For the start time of customer load peak, For the end time of customer load peak, For extracting peak customer load;

[0023] The customer valley feature profiling model is based on the DDPM algorithm to mine the customer's historical workload and extract the customer valley features, represented as follows:

[0024]

[0025] Customer peak value characteristic profile is represented as follows:

[0026]

[0027] In the formula, For customer valley value characteristics; This represents a customer valley feature profile model based on DDPM; The start time of the customer's load trough. The end time of the customer's load trough. This is for extracting customer load valley values.

[0028] In a preferred embodiment of the present invention, the power grid peak characteristics include the start time of the power grid peak, the end time of the power grid peak, and the power grid peak power, expressed as:

[0029]

[0030] In the formula, Indicates the peak characteristics of the power grid; This refers to the start time of peak grid power. This refers to the end time of peak power in the power grid. This represents the peak power of the power grid.

[0031] The grid valley characteristics include the start time of the grid power valley, the end time of the grid power valley, and the grid valley power, expressed as:

[0032]

[0033] In the formula, Indicates the valley value characteristics of the power grid; This refers to the start time of the power grid valley. This refers to the end time of the power grid valley. This represents the off-peak power of the power grid.

[0034] As a preferred embodiment of the present invention, the peak shaving and valley filling time difference model includes a peak time difference model and a valley time difference model;

[0035] The peak time difference model specifically calculates the peak time difference based on the start time of the grid power peak, the end time of the grid power peak, and the start time and end time of the customer load peak. The expression is as follows:

[0036]

[0037] The valley time difference model specifically calculates the valley time difference based on the start time of the grid power valley, the end time of the grid power valley, and the start time and end time of the customer load valley. The expression is as follows:

[0038]

[0039] The peak shaving and valley filling time difference model is expressed as follows:

[0040] In the formula, This represents the peak time difference. The time difference between the valley values; T C This represents the time difference for peak shaving and valley filling.

[0041] As a preferred embodiment of the present invention, the peak shaving and valley filling power difference model includes a peak power difference model and a valley power difference model;

[0042] The peak power difference model specifically calculates the peak power difference based on the cumulative peak power of the power grid and the cumulative peak power of the customer load. The expression is as follows:

[0043]

[0044] The valley-peak electricity difference model specifically calculates the valley-peak electricity difference based on the cumulative valley-peak electricity of the power grid and the cumulative valley-peak electricity of the customer load. The expression is as follows:

[0045]

[0046] The peak-shaving and valley-filling electricity difference model is expressed as follows:

[0047] In the formula, This represents the peak power difference. Q represents the difference in electricity consumption during off-peak hours. C The difference in electricity consumption during peak shaving and valley filling; t represents time.

[0048] As a preferred embodiment of the present invention, the four electricity retail packages are respectively represented as follows:

[0049]

[0050] In the formula, Y1 represents Package One, R1 is the basic cost of Package One, and M1 is the included electricity consumption of Package One; Y2 represents Package Two, R2 is the basic cost of Package Two, and M2 is the included electricity consumption of Package Two; Y3 represents Package Three, R3 is the basic cost of Package Three, and M3 is the included electricity consumption of Package Three; Y4 represents Package Four, R4 is the basic cost of Package Four, and M4 is the included electricity consumption of Package Four; T C T represents the time difference for peak shaving and valley filling. θ The threshold for the time difference between peak shaving and valley filling; Q C Q represents the difference in electricity consumption for peak shaving and valley filling. θ The threshold for peak-shaving and valley-filling electricity difference;

[0051] The basic fees and included electricity charges for the four electricity retail packages are calculated using the following formulas:

[0052]

[0053]

[0054] In the formula, R0 is the basic cost of the electricity retail package; α1, α2, α3, and α4 are the time difference cost coefficients corresponding to packages one, two, three, and four, respectively; β1, β2, β3, and β4 are the electricity difference cost coefficients corresponding to packages one, two, three, and four, respectively; M0 is the basic included electricity volume of the electricity retail package; δ1, δ2, δ3, and δ4 are the time difference electricity coefficients corresponding to packages one, two, three, and four, respectively; and ε1, ε2, ε3, and ε4 are the electricity difference electricity coefficients corresponding to packages one, two, three, and four, respectively.

[0055] As a preferred embodiment of the present invention, the peak shaving and valley filling time difference judgment model is based on the historical load P of customer i. L,i (t), grid load P G (t), Customer i's historical electricity consumption basic cost R i Customer i's historical electricity consumption Q i and peak shaving and valley filling time difference threshold T θ As input, the result T is determined by the peak-shaving and valley-filling time difference of customer i. δ,i For output, the expression is:

[0056]

[0057] When T δ,i A value of 0 indicates the peak shaving and valley filling time difference T for customer i. C,i Less than or equal to the peak-shaving and valley-filling time difference threshold T θ When T δ,i A value of 1 indicates the peak shaving and valley filling time difference T for customer i. C,i Greater than the peak-shaving and valley-filling time difference threshold T θ The expression is:

[0058]

[0059] The peak shaving and valley filling power difference judgment model is based on customer i's historical load P. L,i (t), grid load P G (t), Customer i's historical electricity consumption basic cost R i Customer i's historical electricity consumption Q i The threshold Q for peak shaving and valley filling power difference θ As input, the result Q is determined by the difference in peak-shaving and valley-filling electricity consumption for customer i. δ,i For output, the expression is:

[0060]

[0061] When Q δ,i A value of 0 indicates the peak-shaving and valley-filling power difference Q for customer i. C,iLess than or equal to the peak-shaving and valley-filling power difference threshold Q θ When Q δ,i A value of 1 indicates the peak-shaving and valley-filling power difference Q for customer i. C,i The difference in electricity consumption between peak and valley filling is greater than the threshold Q. θ The expression is:

[0062]

[0063] In the formula, This represents a peak-shaving and valley-filling time difference judgment model based on GPT. This represents the peak-shaving and valley-filling power difference judgment model based on GPT; t represents time.

[0064] As a preferred embodiment of the present invention, the step of selecting one of Package One to Package Four to push to the customer based on the judgment result of GPT specifically includes:

[0065] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i Take 1, and determine the peak-shaving and valley-filling power difference judgment result Q for customer i. δ,i When selecting option 1, push Package 1 to customer i;

[0066] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i Take 1, and determine the peak-shaving and valley-filling power difference judgment result Q for customer i. δ,i When the threshold is 0, push Package Two to customer i;

[0067] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i The result Q is set to 0, and the peak-shaving and valley-filling power difference judgment result for customer i is... δ,i When taking option 1, push package 3 to customer i;

[0068] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i The result Q is set to 0, and the peak-shaving and valley-filling power difference judgment result for customer i is... δ,i When the value is 0, push Package 4 to customer i.

[0069] On the other hand, the present invention also proposes a customer profile-based electricity retail package push system, based on the customer profile-based electricity retail package push method described above, the system comprising:

[0070] The customer peak shaving and valley filling profiling module is used to build a customer peak shaving and valley filling profiling model, including a customer peak feature profiling model and a customer valley feature profiling model. Based on the DDPM algorithm, the module mines the customer's historical load and extracts the customer's peak features and valley features respectively to realize the profiling of the customer's peak shaving and valley filling features.

[0071] A power grid feature extraction module is used to extract power grid peak value features and power grid valley value features. The power grid peak value features include the start time of the power grid peak, the end time of the power grid peak, and the peak power of the power grid. The power grid valley value features include the start time of the power grid valley, the end time of the power grid valley, and the valley power of the power grid.

[0072] The peak shaving and valley filling time difference module is used to construct a peak shaving and valley filling time difference model to calculate the peak shaving and valley filling time difference. It includes a peak time difference model and a valley time difference model, which are used to calculate the peak time difference and valley time difference, respectively.

[0073] The peak shaving and valley filling power difference module is used to build a peak shaving and valley filling power difference model to calculate the peak shaving and valley filling power difference, including a peak power difference model and a valley power difference model, which calculate the peak power difference and valley power difference respectively;

[0074] The electricity retail package creation module is used to create four different electricity retail packages based on the difference between peak shaving and valley filling time and the difference between peak shaving and valley filling electricity volume.

[0075] The electricity retail package judgment module is used to build an electricity retail package judgment model based on GPT, including a peak shaving and valley filling time difference judgment model and a peak shaving and valley filling electricity difference judgment model, to judge the customer's peak shaving and valley filling time difference and peak shaving and valley filling electricity difference.

[0076] The electricity retail package push module is used to select one of the packages from package one to package four to push to the customer based on the judgment result of GPT.

[0077] Furthermore, the present invention also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a customer profile-based electricity retail package push method as described above.

[0078] Compared with existing technologies, the beneficial effects of this invention are as follows: By modeling the customer's historical load using the DDPM algorithm and extracting the customer's peak shaving and valley filling characteristics, the invention achieves accurate characterization of the customer's electricity consumption patterns, enabling electricity retail packages to be precisely customized according to the customer's actual needs, thereby improving user satisfaction; by combining grid load characteristics and customer peak shaving and valley filling characteristics to construct peak shaving and valley filling time difference and electricity difference models, four typical personalized electricity retail packages are designed to adapt to the needs of different users and reduce user electricity costs while reducing waste; through a GPT-based intelligent judgment model, the most suitable package is automatically pushed according to the customer's electricity consumption characteristics, improving the accuracy of electricity retail package push, optimizing power resource allocation, reducing customer electricity costs, and significantly improving service efficiency and user experience. Attached Figure Description

[0079] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0080] in:

[0081] Figure 1 is a flowchart of the method of the present invention;

[0082] Figure 2 is a schematic diagram of the modular structure of the system of the present invention. Detailed Implementation

[0083] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0084] As shown in Figure 1, this is an embodiment of the present invention, which provides a method for pushing electricity retail packages based on customer profiles, including the following steps:

[0085] S1: Construct a customer peak shaving and valley filling profile model based on DDPM, including a customer peak value feature profile model and a customer valley value feature profile model, to mine the customer's historical load, extract the customer's peak shaving and valley filling features, and realize the profile of the customer's peak shaving and valley filling features (representing the customer's overall electricity consumption fluctuation pattern).

[0086] DDPM (Denoising Diffusion Probabilistic Model) is a probabilistic model based on a diffusion process. Its core idea is to gradually transform data into noise through a forward diffusion process, and then gradually transform the noise back into data through a reverse process. This reverse process is achieved by learning a conditional probability distribution that describes how to recover data from noise given the current state and the target state.

[0087] Historical electricity consumption data (historical load) represents a customer's actual electricity load over a past period, recording their power consumption at different times. The DDPM algorithm is used to model and analyze this historical load, extracting peak-shaving and valley-filling characteristics—specifically, features during peak and off-peak electricity demand periods. Peak load characteristics refer to various features extracted from the customer's historical load data for the "peak load" period (such as peak start time, peak end time, peak power, and peak duration). These features represent the customer's electricity consumption behavior during peak demand periods and can be used to describe their electricity consumption patterns during high-load periods. Valley load characteristics are similar to peak load characteristics, referring to the customer's electricity consumption characteristics during off-peak periods (such as valley start time, valley end time, valley power, and valley duration). These features describe the customer's electricity consumption patterns during off-peak electricity demand periods.

[0088] The term "customer profile" typically refers to a comprehensive description of a customer's electricity consumption behavior, formed by integrating multiple features. For example, a customer peak characteristic profile includes not only the customer's peak characteristics but also the patterns and trends obtained by analyzing these characteristics (e.g., the frequency of peak occurrences, the intensity of each peak, etc.). It can be understood as a profile of a customer's electricity consumption during peak electricity periods, including but not limited to a single peak characteristic.

[0089] In one embodiment, the customer peak feature profiling model extracts customer peak features based on the DDPM algorithm by mining historical customer workload, as shown below:

[0090]

[0091] The peak customer profile is represented as follows:

[0092]

[0093] In the formula, Peak characteristics for customers; This represents a customer peak feature profile model based on DDPM; P L (t) represents the customer's historical load; For the start time of customer load peak, For the end time of customer load peak, For extracting peak customer load;

[0094] The customer valley load feature profiling model is based on the DDPM algorithm to mine the customer's historical workload and extract the customer valley load feature, represented as:

[0095]

[0096] Customer peak value characteristic profile is represented as follows:

[0097]

[0098] In the formula, For customer valley value characteristics; This represents a customer valley feature profile model based on DDPM; The start time of the customer's load trough. The end time of the customer's load trough. This is for extracting customer load valley values.

[0099] S2: Based on the peak and valley characteristics of the power grid and the customer peak-shaving and valley-filling profile model, a peak-shaving and valley-filling time difference model and a peak-shaving and valley-filling electricity difference model are constructed. Based on the magnitude of the peak-shaving and valley-filling time difference and the magnitude of the peak-shaving and valley-filling electricity difference, four electricity retail packages are formulated, specifically:

[0100] Package 1: Electricity retail package when the difference between peak shaving and valley filling time is high, and the difference between peak shaving and valley filling electricity volume is also high;

[0101] Package Two: Electricity retail package when the peak shaving and valley filling time difference is high and the peak shaving and valley filling electricity difference is low;

[0102] Package 3: Electricity retail package when the peak shaving and valley filling time difference is low and the peak shaving and valley filling electricity difference is high;

[0103] Package 4: Electricity retail package with low peak shaving and valley filling time difference and low peak shaving and valley filling electricity difference.

[0104] Furthermore, the peak characteristics of the power grid include the start time of the peak power generation, the end time of the peak power generation, and the peak power, expressed as:

[0105]

[0106] In the formula, Indicates the peak characteristics of the power grid; This refers to the start time of peak grid power. This refers to the end time of peak power in the power grid. This represents the peak power of the power grid.

[0107] The characteristics of power grid valleys include the start time of the power valley, the end time of the power valley, and the power valley value, expressed as:

[0108]

[0109] In the formula, Indicates the valley value characteristics of the power grid; This refers to the start time of the power grid valley. This refers to the end time of the power grid valley. This represents the off-peak power of the power grid.

[0110] In one embodiment, the peak shaving and valley filling time difference model includes a peak time difference model and a valley time difference model;

[0111] The peak time difference model specifically calculates the peak time difference based on the start time of the grid power peak, the end time of the grid power peak, and the start time and end time of the customer load peak. The expression is as follows:

[0112]

[0113] The valley time difference model specifically calculates the valley time difference based on the start time of the grid power valley, the end time of the grid power valley, and the start time and end time of the customer load valley. The expression is as follows:

[0114]

[0115] Therefore, the peak shaving and valley filling time difference model can be expressed as:

[0116] In the formula, This represents the peak time difference. The time difference between the trough values; T C This represents the time difference for peak shaving and valley filling.

[0117] The peak-shaving and valley-filling power difference model includes the peak power difference model and the valley power difference model;

[0118] The peak power difference model specifically calculates the peak power difference based on the cumulative peak power of the power grid and the cumulative peak power of the customer load. The expression is as follows:

[0119]

[0120] The off-peak electricity difference model specifically calculates the off-peak electricity difference based on the cumulative off-peak electricity consumption of the power grid and the cumulative off-peak electricity consumption of the customer load. The expression is as follows:

[0121]

[0122] Therefore, the peak-shaving and valley-filling electricity difference model can be expressed as:

[0123] In the formula, This represents the peak power difference. Q represents the difference in electricity consumption during off-peak hours. C The difference in electricity consumption during peak shaving and valley filling; t represents time.

[0124] The four electricity retail packages are represented as follows:

[0125]

[0126] In the formula, Y1 represents Package One, R1 is the basic cost of Package One, and M1 is the included electricity consumption of Package One; Y2 represents Package Two, R2 is the basic cost of Package Two, and M2 is the included electricity consumption of Package Two; Y3 represents Package Three, R3 is the basic cost of Package Three, and M3 is the included electricity consumption of Package Three; Y4 represents Package Four, R4 is the basic cost of Package Four, and M4 is the included electricity consumption of Package Four; T C T represents the time difference for peak shaving and valley filling. θ The threshold for the time difference between peak shaving and valley filling; Q C Q represents the difference in electricity consumption for peak shaving and valley filling. θ The threshold for peak-shaving and valley-filling electricity difference;

[0127] The basic fees and included electricity charges for the four electricity retail packages are calculated using the following formulas:

[0128]

[0129]

[0130] In the formula, R0 is the basic cost of the electricity retail package; α1, α2, α3, and α4 are the time difference cost coefficients corresponding to packages one, two, three, and four, respectively; β1, β2, β3, and β4 are the electricity difference cost coefficients corresponding to packages one, two, three, and four, respectively; M0 is the basic included electricity volume of the electricity retail package; δ1, δ2, δ3, and δ4 are the time difference electricity coefficients corresponding to packages one, two, three, and four, respectively; and ε1, ε2, ε3, and ε4 are the electricity difference electricity coefficients corresponding to packages one, two, three, and four, respectively.

[0131] S3: Construct a GPT-based electricity retail package judgment model, including a peak shaving and valley filling time difference judgment model and a peak shaving and valley filling electricity difference judgment model. Judge the peak shaving and valley filling time difference and peak shaving and valley filling electricity difference of customers, and select one of the packages from package one to package four to push to customers based on the judgment results of GPT.

[0132] GPT (Generative Pre-trained Transformer) is a language model based on artificial intelligence technology. GPT technology is based on the Transformer architecture in deep learning, and is pre-trained through unsupervised learning and fine-tuned on specific tasks to achieve efficient and accurate language processing.

[0133] In one specific embodiment, the peak shaving and valley filling time difference judgment model uses the historical load P of customer i. L,i (t), grid load P G (t), Customer i's historical electricity consumption basic cost R i Customer i's historical electricity consumption Q i and peak shaving and valley filling time difference threshold T θ As input, the result T is determined by the peak-shaving and valley-filling time difference of customer i. δ,i For output, the expression is:

[0134]

[0135] When T δ,i A value of 0 indicates the peak shaving and valley filling time difference T for customer i. C,i Less than or equal to the peak-shaving and valley-filling time difference threshold T θ When T δ,i A value of 1 indicates the peak shaving and valley filling time difference T for customer i. C,i Greater than the peak-shaving and valley-filling time difference threshold T θ The expression is:

[0136]

[0137] Peak shaving and valley filling power difference judgment model based on customer i's historical load P L,i (t), grid load P G (t), Customer i's historical electricity consumption basic cost R i Customer i's historical electricity consumption Q i The threshold Q for peak shaving and valley filling power difference θ As input, the result Q is determined by the difference in peak-shaving and valley-filling electricity consumption for customer i. δ,i For output, the expression is:

[0138]

[0139] When Q δ,i A value of 0 indicates the peak-shaving and valley-filling power difference Q for customer i. C,i Less than or equal to the peak-shaving and valley-filling power difference threshold Q θ When Q δ,iA value of 1 indicates the peak-shaving and valley-filling power difference Q for customer i. C,i The difference in electricity consumption between peak and valley filling is greater than the threshold Q. θ The expression is:

[0140]

[0141] In the formula, This represents a peak-shaving and valley-filling time difference judgment model based on GPT. This represents the peak-shaving and valley-filling power difference judgment model based on GPT; t represents time.

[0142] Furthermore, based on the GPT's assessment results, one of Package One through Package Four will be selected and pushed to the customer, specifically:

[0143] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i Take 1, and the result Q of the peak shaving and valley filling power difference judgment for customer i. δ,i When selecting option 1, push Package 1 to customer i;

[0144] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i Take 1, and the result Q of the peak shaving and valley filling power difference judgment for customer i. δ,i When the threshold is 0, push Package Two to customer i;

[0145] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i The result Q is set to 0, and the peak-shaving and valley-filling power difference judgment result for customer i is... δ,i When taking option 1, push package 3 to customer i;

[0146] When the peak shaving and valley filling time difference judgment result of customer i is T δ,i The result Q is set to 0, and the peak-shaving and valley-filling power difference judgment result for customer i is... δ,i When the threshold is 0, Package 4 is pushed to customer i.

[0147] Step S1 uses the DDPM algorithm to obtain peak and valley characteristic profiles of customers, serving as the foundational data for subsequently developing electricity retail packages. Step S2 combines these characteristics with grid load characteristics to create different electricity retail packages. Step S3 then pushes matching electricity retail packages to customers based on the judgment of the GPT model. In this way, the entire process, through multi-layered data input-output relationships, ultimately achieves personalized electricity retail package delivery, improving the efficiency of electricity resource utilization and customer satisfaction.

[0148] As shown in Figure 2, another embodiment of the present invention provides a customer profile-based electricity retail package push system, based on the customer profile-based electricity retail package push method described above, including:

[0149] The customer peak shaving and valley filling profiling module is used to build a customer peak shaving and valley filling profiling model, including a customer peak feature profiling model and a customer valley feature profiling model. Based on the DDPM algorithm, the module mines the customer's historical load and extracts the customer's peak features and valley features respectively to realize the profiling of the customer's peak shaving and valley filling features.

[0150] The power grid feature extraction module is used to extract the peak power features and valley power features of the power grid. The peak power features include the start time of the peak power, the end time of the peak power, and the peak power. The valley power features include the start time of the valley power, the end time of the valley power, and the valley power.

[0151] The peak shaving and valley filling time difference module is used to construct a peak shaving and valley filling time difference model to calculate the peak shaving and valley filling time difference. It includes a peak time difference model and a valley time difference model, which are used to calculate the peak time difference and valley time difference, respectively.

[0152] The peak shaving and valley filling power difference module is used to build a peak shaving and valley filling power difference model to calculate the peak shaving and valley filling power difference, including a peak power difference model and a valley power difference model, which calculate the peak power difference and valley power difference respectively;

[0153] The electricity retail package creation module is used to create four different electricity retail packages based on the difference between peak shaving and valley filling time and the difference between peak shaving and valley filling electricity volume.

[0154] The electricity retail package judgment module is used to build an electricity retail package judgment model based on GPT, including a peak shaving and valley filling time difference judgment model and a peak shaving and valley filling electricity difference judgment model, to judge the customer's peak shaving and valley filling time difference and peak shaving and valley filling electricity difference.

[0155] The electricity retail package push module is used to select one of the packages from package one to package four to push to the customer based on the judgment result of GPT.

[0156] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any other combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0157] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0158] Therefore, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a customer profile-based electricity retail package push method as described above.

[0159] In summary, this invention achieves accurate extraction of customer peak-shaving and valley-filling characteristics through in-depth analysis of customer historical load. By constructing customer peak and valley characteristic models, it successfully characterizes customer electricity demand characteristics during peak and off-peak periods, providing personalized data support for subsequent electricity package development. Through accurate profiling of customer peak-shaving and valley-filling characteristics, it can effectively identify customer electricity consumption patterns, thereby providing tailored electricity packages based on the differences in electricity consumption behavior among different customers.

[0160] In terms of package design, this invention combines the characteristics of the power grid load to construct a time difference and power difference model. By comparing the peak shaving and valley filling characteristics of customers and the power grid, four types of electricity retail packages are designed. These packages are pushed according to different combinations of peak shaving and valley filling time differences and power differences of customers, ensuring that each customer can obtain the electricity package that best meets their actual needs, thereby optimizing the allocation of power resources and reducing the electricity costs of customers.

[0161] Furthermore, this invention employs a GPT-based intelligent judgment model to automatically select and push the most suitable electricity package to customers based on their peak-shaving and valley-filling characteristics. This intelligent push method further improves the accuracy and personalization of package selection, reduces manual intervention, and enhances customer satisfaction. Overall, this invention not only improves the efficiency of electricity resource utilization but also enhances the personalized experience for customers, promoting the rationalization and economy of electricity consumption.

[0162] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for recommending electricity retail packages based on customer profiles, characterized in that, The method includes: constructing a customer peak-shaving and valley-filling profile model based on DDPM, including a customer peak feature profile model and a customer valley feature profile model, used to mine historical load data of customers, extract peak-shaving and valley-filling features of customers, and realize a profile of customer peak-shaving and valley-filling features; the customer peak feature profile model is based on the DDPM algorithm to mine historical load data of customers and extract customer peak features, represented as: The peak customer profile is represented as follows: In the formula, For customer peak characteristics; This represents a customer peak feature profile model based on DDPM; For customer historical load; For the start time of customer load peak, For the end time of customer load peak, The extracted customer load peak; the customer valley feature profiling model is based on the DDPM algorithm to mine the customer's historical load and extract the customer valley features, represented as: Customer peak value characteristic profile is represented as follows: In the formula, For customer valley value characteristics; This represents a customer valley feature profile model based on DDPM; The start time of the customer's load trough. The end time of the customer's load trough. To extract customer load valley values; based on grid peak value characteristics, grid valley value characteristics, and the customer peak-shaving and valley-filling profile model, a peak-shaving and valley-filling time difference model and a peak-shaving and valley-filling electricity difference model are constructed. Based on the magnitude of the peak-shaving and valley-filling time difference and the magnitude of the peak-shaving and valley-filling electricity difference, four electricity retail packages are formulated: Package 1: Electricity retail package with both high peak-shaving and valley-filling time difference and high peak-shaving and valley-filling electricity difference; Package 2: Electricity retail package with both high peak-shaving and valley-filling time difference and low peak-shaving and valley-filling electricity difference. Electricity retail packages are offered in two ways: Package 3: a package where the peak-shaving and valley-filling time difference is low and the peak-shaving and valley-filling electricity difference is high; Package 4: a package where the peak-shaving and valley-filling time difference is low and the peak-shaving and valley-filling electricity difference is low. A GPT-based electricity retail package judgment model is constructed, including a peak-shaving and valley-filling time difference judgment model and a peak-shaving and valley-filling electricity difference judgment model. The peak-shaving and valley-filling time difference and peak-shaving and valley-filling electricity difference of customers are judged, and one of Package 1 to Package 4 is selected to be pushed to customers based on the judgment results of GPT.

2. The method for recommending electricity retail packages based on customer profiles as described in claim 1, characterized in that, The peak power characteristics of the power grid include the start time of the peak power grid power, the end time of the peak power grid power, and the peak power grid power, expressed as follows: In the formula, Indicates the peak characteristics of the power grid; This refers to the start time of peak grid power. This refers to the end time of peak power in the power grid. The peak power of the power grid; the valley power characteristics include the start time of the valley power, the end time of the valley power, and the valley power, expressed as: In the formula, Indicates the valley value characteristics of the power grid; This refers to the start time of the power grid valley. This refers to the end time of the power grid valley. This represents the off-peak power of the power grid.

3. The method for recommending electricity retail packages based on customer profiles as described in claim 2, characterized in that, The peak shaving and valley filling time difference model includes a peak time difference model and a valley time difference model; specifically, the peak time difference model calculates the peak time difference based on the start time of the grid power peak, the end time of the grid power peak, and the start time and end time of the customer load peak, as expressed in the following expression: The valley time difference model specifically calculates the valley time difference based on the start time of the grid power valley, the end time of the grid power valley, and the start time and end time of the customer load valley. The expression is as follows: The peak shaving and valley filling time difference model is expressed as follows: In the formula, This represents the peak time difference. This represents the time difference between the valley values. This represents the time difference for peak shaving and valley filling.

4. The method for recommending electricity retail packages based on customer profiles as described in claim 2, characterized in that, The peak-shaving and valley-filling power difference model includes a peak power difference model and a valley power difference model; the peak power difference model specifically calculates the peak power difference based on the cumulative peak power of the power grid and the cumulative peak power of the customer load, as expressed in the following expression: The valley power difference model specifically calculates the valley power difference based on the cumulative valley power of the power grid and the cumulative valley power of the customer load, expressed as: The peak-shaving and valley-filling power difference model is expressed as follows: In the formula, This represents the peak power difference. This represents the difference in electricity consumption during off-peak hours. This represents the difference in electricity consumption for peak shaving and valley filling. Indicates time.

5. The method for recommending electricity retail packages based on customer profiles as described in claim 1, characterized in that, The four electricity retail packages are respectively represented as follows: In the formula, This refers to Package One. This is the base fee for Package One. The included battery capacity is for Package 1. This indicates Package Two. This is the base fee for Package Two. The included battery capacity is for Package Two; This indicates Package Three. This is the base fee for Package Three. The included battery capacity is for Package 3; This indicates Package Four. This is the base fee for Package Four. The included battery capacity is for Package 4; This represents the time difference for peak shaving and valley filling. The threshold for the time difference between peak shaving and valley filling; The difference in electricity consumption for peak shaving and valley filling. The threshold for peak-shaving and valley-filling electricity difference; among them, the calculation formulas for the basic fee and included electricity volume of the four electricity retail packages are as follows: ； In the formula, This is the base fee for electricity retail packages; 、 、 、 These are the time difference cost coefficients for packages one, two, three, and four, respectively. 、 、 、 These are the electricity difference fee coefficients for Package 1, 2, 3, and 4, respectively. Electricity is included in the basic electricity retail package; 、 、 、 These are the time difference power consumption coefficients corresponding to Packages One, Two, Three, and Four, respectively. 、 、 、 These are the power difference coefficients for Package 1, 2, 3, and 4, respectively.

6. The method for recommending electricity retail packages based on customer profiles as described in claim 1, characterized in that, The peak shaving and valley filling time difference judgment model is based on customer i's historical load. Power grid load Customer i's historical electricity consumption base fee Customer i's historical electricity consumption Peak shaving and valley filling time difference threshold The result is determined by the peak-shaving and valley-filling time difference of customer i, with the input being the peak-shaving and valley-filling time difference. For output, the expression is: ； when A value of 0 indicates the time difference for peak shaving and valley filling for customer i. Less than or equal to the peak-shaving and valley-filling time difference threshold ,when A value of 1 indicates the peak shaving and valley filling time difference for customer i. Greater than the peak-shaving and valley-filling time difference threshold The expression is: The peak-shaving and valley-filling power difference judgment model is based on the historical load of customer i. Power grid load Customer i's historical electricity consumption base fee Customer i's historical electricity consumption Peak shaving and valley filling power difference threshold The input is the peak-shaving and valley-filling electricity difference of customer i, and the result is judged accordingly. For output, the expression is: ； when A value of 0 indicates the difference in peak-shaving and valley-filling electricity consumption for customer i. Less than or equal to the peak-shaving and valley-filling power difference threshold ,when A value of 1 indicates the difference in peak-shaving and valley-filling electricity consumption for customer i. Greater than the peak-shaving and valley-filling power difference threshold The expression is: In the formula, This represents a peak-shaving and valley-filling time difference judgment model based on GPT. This represents a peak-shaving and valley-filling power difference judgment model based on GPT. Indicates time.

7. The method for recommending electricity retail packages based on customer profiles as described in claim 6, characterized in that, The step of selecting one of Package 1 to Package 4 to push to the customer based on the judgment result of GPT is as follows: when the peak shaving and valley filling time difference judgment result of customer i The result of determining the peak shaving and valley filling power difference for customer i is 1. When taking option 1, push Package 1 to customer i; when customer i's peak shaving and valley filling time difference judgment result... The result of determining the peak shaving and valley filling power difference for customer i is 1. When the value is 0, Package Two is pushed to customer i; when the peak shaving and valley filling time difference judgment result for customer i is... The result of determining the peak shaving and valley filling power difference for customer i is set to 0. When taking option 1, push Package 3 to customer i; when customer i's peak shaving and valley filling time difference judgment result... The result of determining the peak shaving and valley filling power difference for customer i is set to 0. When the threshold is 0, Package 4 is pushed to customer i.

8. A customer profile-based electricity retail package recommendation system, based on the customer profile-based electricity retail package recommendation method as described in any one of claims 1-7, characterized in that, The system includes: a customer peak shaving and valley filling profiling module, used to construct a customer peak shaving and valley filling profiling model, including a customer peak characteristic profiling model and a customer valley characteristic profiling model. Based on the DDPM algorithm, historical load data of customers is mined to extract customer peak characteristics and valley characteristics, respectively, thus profiling the customer's peak shaving and valley filling characteristics; a power grid feature extraction module, used to extract power grid peak characteristics and power grid valley characteristics. The power grid peak characteristics include the start time, end time, and peak power of the power grid peak; the power grid valley characteristics include the start time, end time, and valley power of the power grid valley; and a peak shaving and valley filling time difference module, used to construct a peak shaving and valley filling time difference model to calculate the peak shaving and valley filling time difference, including a peak time difference model and a valley time difference model. The system comprises four modules: a peak-shaving and valley-shaving electricity difference module, a peak-shaving and valley-shaving electricity difference module, and a power retail package formulation module. The peak-shaving and valley-shaving electricity difference module is used to construct a peak-shaving and valley-shaving electricity difference model to calculate the peak-shaving and valley-shaving electricity difference, including a peak electricity difference model and a valley electricity difference model. The power retail package judgment module is used to construct a GPT-based power retail package judgment model, including a peak-shaving and valley-shaving time difference judgment model and a peak-shaving and valley-shaving electricity difference judgment model, to judge the customer's peak-shaving and valley-shaving time difference and peak-shaving and valley-shaving electricity difference. The power retail package push module is used to select one of the four packages (package one through four) to push to the customer based on the GPT judgment results.

9. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of a method for pushing electricity retail packages based on customer profiles as described in any one of claims 1 to 7.

Images