A user energy storage auxiliary settlement method and system

Abstract

The application discloses a user energy storage auxiliary settlement method and system, and relates to the field of electric power markets and energy management.The method comprises the following steps: obtaining and preprocessing user side data; based on user power consumption data, user power consumption habit index data and energy storage equipment use time length data, calculating the user load fluctuation degree, the user power consumption fluctuation degree and the weekday and weekend use time length ratio; obtaining power consumption time data including each time period, and based on the power consumption time data, the user load fluctuation degree, the user power consumption fluctuation degree and the weekday and weekend use time length ratio, combining a linear regression model and Bayes theorem, outputting and correcting the power consumption prediction value of each time period; based on the corrected power consumption prediction value of each time period and the electricity price of each time period, calculating the user energy storage settlement fee.The application can more accurately predict the final electricity fee of the user.

Description

Technical Field

[0001] This invention relates to the fields of electricity markets and energy management, and more specifically, to a user energy storage auxiliary settlement method and system. Background Technology

[0002] With the continuous development of renewable energy and the gradual opening of the electricity market, user-side energy storage systems have played an important role in regulating grid load and improving energy utilization efficiency. As a result, a user energy storage auxiliary settlement method and system have emerged. One existing user energy storage auxiliary settlement method and system can collect various data on the user side, extract multiple feature data from user electricity consumption habit data to build a model for prediction, obtain the user's charging or discharging amount at different time periods based on the predicted values, and calculate the cost and revenue for each time period based on the user's charging or discharging amount and electricity price. Finally, a comprehensive calculation is performed to obtain the actual settlement cost of the energy storage device.

[0003] While existing user energy storage auxiliary settlement methods and systems have achieved user energy storage auxiliary settlement to a certain extent, they suffer from problems such as inaccurate data and complex settlement in the process of electricity and electricity bill settlement. It is difficult to extract multiple feature data from user electricity consumption habit index data to build a model, and it lacks the ability to obtain the discharge volume during peak hours and the charging volume during normal and off-peak hours based on the constructed model. It is also difficult to comprehensively calculate the actual settlement cost of energy storage equipment by combining the costs and benefits obtained at different times.

[0004] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0005] In view of the problems in related technologies, this invention proposes a user energy storage auxiliary settlement method and system to overcome the above-mentioned technical problems existing in the existing related technologies.

[0006] Therefore, the specific technical solution adopted by the present invention is as follows:

[0007] According to one aspect of the present invention, a user energy storage auxiliary billing method is provided, the method comprising:

[0008] S1. Acquire and preprocess user-side data, including user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data.

[0009] S2. Based on user electricity consumption data, user electricity consumption habit index data and energy storage device usage time data, calculate the user load fluctuation degree, user electricity consumption fluctuation degree and weekday to weekend usage time ratio;

[0010] S3. Obtain electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation, user electricity consumption fluctuation and weekday to weekend usage time ratio, and combined with linear regression model and Bayes' theorem, output and correct the predicted electricity consumption value for each time period.

[0011] S4. Based on the corrected electricity consumption forecasts for each time period and the electricity price for each time period, calculate the user's energy storage settlement fee.

[0012] Furthermore, based on user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data, the following are calculated: user load fluctuation level, user electricity consumption fluctuation level, and weekday / weekend usage time ratio:

[0013] Extract user power load data within a period from user electricity consumption habit index data, and the user power load data within a period includes the highest peak value, the lowest valley value and the average value;

[0014] The degree of fluctuation in user load is calculated based on the highest peak, lowest valley, and average value of user power load data within the period.

[0015] The degree of fluctuation in user electricity consumption is calculated based on user electricity consumption data;

[0016] Based on the power consumption data of the energy storage equipment, and combined with the number of weekdays and weekends, the ratio of weekday to weekend usage time is calculated.

[0017] Furthermore, based on the highest peak, lowest trough, and average user power load data within the period, the degree of user load fluctuation is calculated, including:

[0018] The degree of load fluctuation of a user is calculated by using the user's highest load peak value, lowest load trough value, average load value, and daily load value within a period.

[0019] Furthermore, based on user electricity consumption data, the degree of fluctuation in user electricity consumption is calculated, including:

[0020] The degree of fluctuation in a user's electricity consumption is calculated by using the user's daily electricity consumption within a period and the user's average electricity consumption within the period.

[0021] Furthermore, based on the electricity consumption data of the energy storage devices, and combined with the number of weekdays and weekends, the ratio of weekday to weekend usage time was calculated as follows:

[0022] Acquire daily electricity consumption data for users within a period, and extract features from user electricity consumption data and electricity consumption duration data to obtain average daily electricity consumption and standard deviation of electricity consumption;

[0023] Based on the average daily electricity consumption and the standard deviation of electricity consumption, and combined with machine learning algorithms, a classification model is trained, and the attributes of each day are output using the classification model.

[0024] Record the number of times each day is marked as a workday to obtain the total number of workdays within the period. Then, calculate the total number of weekend days within the period based on the difference between the total number of days within the period and the total number of workdays within the period.

[0025] The ratio of weekday to weekend usage time is calculated by using the total number of working days in the cycle, the usage time of energy storage equipment on each working day, the total number of weekend days in the cycle, and the usage time of energy storage equipment on each weekend.

[0026] Furthermore, electricity consumption time data for each time period is acquired, and based on this data, user load fluctuations, user electricity consumption fluctuations, and the ratio of weekday to weekend usage time, combined with a linear regression model and Bayes' theorem, the predicted electricity consumption values ​​for each time period are output and revised, including:

[0027] Collect electricity consumption data for different time periods, including peak hours, normal hours, and off-peak hours;

[0028] Using user load fluctuation, user electricity consumption fluctuation, weekday vs. weekend usage time ratio, and electricity usage time data as inputs and electricity consumption as output, the linear regression model is trained to obtain the prediction model.

[0029] The predicted electricity consumption values ​​for each time period are output using the prediction model, and the predicted values ​​of the prediction model are corrected based on the Bayesian algorithm.

[0030] Furthermore, the predicted electricity consumption values ​​for each time period are output using the prediction model, including:

[0031] In the predicted electricity consumption output by the prediction model, the maximum, median and minimum values ​​are used as the predicted electricity consumption values ​​for peak hours, normal hours and off-peak hours of the day, respectively.

[0032] Furthermore, the correction of the predicted values ​​of the prediction model based on the Bayesian algorithm includes:

[0033] Obtain the prior probability of the prediction model, and calculate the likelihood function of the prediction model based on the observation data corresponding to the prediction model;

[0034] Using Bayes' theorem, and combining the prior probability and likelihood function of the prediction model, the posterior distribution of the prediction model parameters is calculated, and the prediction values ​​of the prediction model are corrected using the posterior distribution.

[0035] Furthermore, based on the revised electricity consumption forecasts for each time period and the electricity price for each time period, the user's energy storage settlement fees are calculated, including:

[0036] Obtain electricity prices for peak, normal, and off-peak periods;

[0037] The revised peak-hour electricity consumption forecast is used as the discharge amount of the energy storage device during the peak hours, the revised normal-hour electricity consumption forecast is used as the charging amount of the energy storage device during the normal hours, and the revised off-peak-hour electricity consumption forecast is used as the charging amount of the energy storage device during the off-peak-hour.

[0038] The formula for calculating user energy storage settlement fees is as follows:

[0039]

[0040] Where M represents the user's energy storage settlement fee;

[0041] P peak M represents the discharge amount during peak hours. peak Electricity price during peak hours;

[0042] P avg η represents the charging amount during normal operation, η represents the charging and discharging efficiency of the energy storage device, and M represents the charging amount during normal operation. avg This refers to the electricity price during normal periods;

[0043] P valley For charging during off-peak hours, M valley This refers to the electricity price during off-peak hours.

[0044] According to another aspect of the present invention, a user energy storage auxiliary billing system is provided, the system comprising a data acquisition module, a feature data calculation module, a user electricity consumption prediction module, and an energy storage electricity fee settlement module:

[0045] The data acquisition module is used to acquire and preprocess user-side data, which includes user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data.

[0046] The feature data calculation module is used to calculate the degree of user load fluctuation, the degree of user electricity fluctuation, and the ratio of weekday to weekend usage time based on user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data.

[0047] The user electricity consumption prediction module is used to acquire electricity consumption time data for each time period, and based on the electricity consumption time data, the degree of user load fluctuation, the degree of user electricity consumption fluctuation, and the ratio of weekday to weekend usage time, combined with the linear regression model and Bayes' theorem, output and correct the predicted electricity consumption value for each time period.

[0048] The energy storage electricity billing module is used to calculate the user's energy storage billing fee based on the corrected electricity consumption forecast for each time period and the electricity price for each time period.

[0049] The beneficial effects of this invention are as follows:

[0050] (1) This invention analyzes users’ historical electricity consumption data and obtains the user’s load fluctuation degree, user’s electricity consumption fluctuation degree and user’s weekday-weekend usage time ratio as feature data through statistical analysis methods. Based on the extracted features, a model is constructed to predict the electricity consumption and obtain the predicted value. The predicted value is then corrected by Bayesian method, which can more accurately reflect the user’s electricity consumption habits and patterns, improve the accuracy of electricity consumption prediction, and use the predicted electricity consumption at different time periods as the charging or discharging amount of the energy storage device at different time periods, thereby realizing peak shaving and valley filling of the power grid.

[0051] (2) This invention calculates the discharge amount of the energy storage device during peak hours, the charging amount during normal hours and off-peak hours, and the electricity price of the corresponding time period, respectively, to obtain the cost of charging during off-peak hours and normal hours and the revenue of discharging during peak hours, and obtains the actual settlement cost of the energy storage device. This invention realizes the consideration of electricity prices in different time periods, improves the accuracy of the economic benefit calculation of the energy storage device in each time period, and can more accurately predict the final electricity bill of users. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. 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.

[0053] Figure 1 This is a flowchart of a user energy storage auxiliary settlement method according to an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of a user energy storage auxiliary settlement system according to an embodiment of the present invention.

[0055] In the picture:

[0056] 1. Data acquisition module; 2. Feature data calculation module; 3. User electricity consumption prediction module; 4. Energy storage electricity fee settlement module. Detailed Implementation

[0057] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0058] According to embodiments of the present invention, a user energy storage auxiliary settlement method and system are provided.

[0059] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, according to an embodiment of the present invention, a user energy storage auxiliary billing method is provided, the method comprising:

[0060] S1. Acquire and preprocess user-side data, including user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data.

[0061] S2. Based on user electricity consumption data, user electricity consumption habit index data and energy storage device usage time data, calculate the user load fluctuation degree, user electricity consumption fluctuation degree and weekday to weekend usage time ratio;

[0062] S3. Obtain electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation, user electricity consumption fluctuation and weekday to weekend usage time ratio, and combined with linear regression model and Bayes' theorem, output and correct the predicted electricity consumption value for each time period.

[0063] S4. Based on the corrected electricity consumption forecasts for each time period and the electricity price for each time period, calculate the user's energy storage settlement fee.

[0064] like Figure 2 As shown, according to another embodiment of the present invention, a user energy storage auxiliary billing system is provided, the system comprising: a data acquisition module 1, a feature data calculation module 2, a user electricity consumption prediction module 3, and an energy storage electricity fee settlement module 4.

[0065] Among them, the data acquisition module 1 is used to acquire and preprocess user-side data, and the user-side data includes user electricity consumption data, user electricity consumption habit index data and energy storage device usage time data.

[0066] Feature data calculation module 2 is used to calculate the degree of user load fluctuation, the degree of user electricity fluctuation, and the ratio of weekday to weekend usage time based on user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data.

[0067] User electricity consumption prediction module 3 is used to acquire electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation degree, user electricity consumption fluctuation degree and weekday to weekend usage time ratio, and combined with linear regression model and Bayes' theorem, output and correct the electricity consumption prediction value for each time period.

[0068] The energy storage electricity bill settlement module 4 is used to calculate the user's energy storage settlement fee based on the corrected electricity consumption forecast for each time period and the electricity price for each time period.

[0069] To facilitate understanding of the above technical solutions of the present invention, the following will further explain the above technical solutions of the present invention from the perspective of architecture and principle.

[0070] Smart meters with remote communication capabilities are installed at electricity consumption sites, and wireless communication modules are configured for each smart meter to transmit collected data to a cloud platform in real time. Data collection intervals are set on the smart meters, with data collected every hour. The smart meters automatically record electricity consumption for each time period, including total consumption and consumption categorized by peak, normal, and off-peak hours. User electricity consumption habits are also collected using smart meters. Smart meters installed on the grid side collect grid load data. The collected user-side and grid-side data are transmitted in real time to a cloud platform for processing and analysis via a communication network. Data processing services provided by the cloud computing platform are used for batch processing of the data. Filters are applied to remove high-frequency noise, and statistical analysis methods are used to calculate the mean and standard deviation of the data. Outliers are identified and removed using the Z-score method. A unified data format is determined, and data conversion tools on the cloud platform, including AWS Glue cloud data integration services or Azure Data, are used. Factory's cloud-based data integration system can convert data of different formats into a unified format. Features are extracted from user electricity consumption habit data to construct a linear regression model, including the ratio of weekday to weekend usage time, electricity consumption fluctuations, and load fluctuations. Using the extracted features as input and electricity consumption during peak, normal, and off-peak hours as output, the model is trained. The trained model then yields predicted peak, normal, and off-peak electricity consumption values. The predicted peak consumption value is used as the discharge amount of the energy storage device during peak hours, the predicted normal consumption value as the charging amount during normal hours, and the predicted off-peak consumption value as the charging amount during off-peak hours. The peak-hour electricity price is multiplied by the corresponding discharge amount to obtain the peak-hour discharge revenue. The normal-hour electricity price, charging / discharging efficiency, and corresponding charging amount are combined to obtain the normal-hour charging cost.

[0071] The off-peak charging cost is calculated by combining the electricity price, charging and discharging efficiency, and corresponding charging amount during off-peak hours. The total charging cost is obtained by adding the charging costs during normal hours and off-peak hours. The discharge revenue during peak hours is listed separately. The actual settlement cost of the energy storage device is obtained by combining the total charging cost and discharge revenue.

[0072] In one embodiment, S1 includes the following steps:

[0073] Smart meters are installed on both the user side and the grid side. The smart meters collect user electricity consumption data and grid power supply data every hour at preset time intervals, including electricity consumption data during peak hours, normal hours and off-peak hours, user electricity consumption habit index data and grid load data.

[0074] Specifically, smart meters with remote communication capabilities are installed at electricity consumption sites, and wireless communication modules are configured for these smart meters. This enables the meters to transmit collected data to a cloud platform in real time. Data collection intervals are set in the smart meters, with data collected every hour. The smart meters automatically record electricity consumption for each time period, including total electricity consumption and electricity consumption divided into peak, normal, and off-peak periods. The smart meters also collect data on users' electricity consumption habits. Smart meters installed on the power grid side collect power grid load data. The collected user-side and power grid-side data are transmitted in real time to the cloud platform for processing and analysis via a communication network.

[0075] In one embodiment, S2 includes the following steps:

[0076] The collected data is processed using cloud computing algorithms, high-frequency noise is removed using filters, outliers are identified and removed using statistical analysis methods, data of different formats are converted into a unified format, and data from different devices and systems are integrated.

[0077] Specifically, the data is processed in batches using data processing services provided by the cloud computing platform. Filters are applied to remove high-frequency noise, and statistical analysis methods are used to calculate the mean and standard deviation of the data. Outliers are identified and removed using the Z-score method. A unified data format is determined, and data conversion tools on the cloud platform, including either AWS Glue or Azure Data Factory, are used to convert data of different formats into a unified format.

[0078] In one embodiment, S3 includes the following steps:

[0079] The collected data, including daily electricity consumption and load, was cleaned by removing outliers and missing values ​​from the historical data of users over the past month. From the cleaned data, user electricity consumption indicators were extracted, including the degree of fluctuation in user load, the degree of fluctuation in user electricity consumption, and the ratio of user usage time between weekdays and weekends.

[0080] Record user electricity load data over a month, including peak, trough, and average values, to obtain the formula for calculating the degree of user load fluctuation:

[0081]

[0082] Among them, C degree For the degree of user load fluctuation, C highest C represents the user's peak load within a month. lowst C represents the lowest load value for users within a month. avg C represents the average load of users over a month. i Let 30 be the user's load value on day i, and 30 be the total number of days in a month.

[0083] The formula for calculating the fluctuation of user electricity consumption is derived by recording user electricity consumption data over a month:

[0084]

[0085] Among them, E degree E represents the degree of fluctuation in user electricity consumption. i E represents the user's electricity consumption on day i. avg This represents the user's average electricity consumption over a month, with 30 representing the total number of days in a month.

[0086] The system monitors the energy storage devices and uses smart meters to record daily usage time data for one month. After cleaning and removing outliers, features are extracted from the electricity consumption and usage time data, including average daily electricity consumption and standard deviation. A classification model is trained using a machine learning algorithm, taking the extracted features as input and outputting whether each day is a weekday or weekend. The collected data is categorized into weekdays and weekends, and the average usage time for each category is calculated. The average usage time for weekdays is compared with the average usage time for weekends, and the ratio is denoted as the weekday-weekend usage time ratio. It should be noted that in this invention, weekends refer to non-working days.

[0087] Formula for calculating the ratio of weekday to weekend usage time:

[0088]

[0089] Among them, T rate The ratio of weekday to weekend usage time, where n is the total number of weekdays in a month, and T is the total number of weekdays in a month. workday,i Let T be the usage duration of the energy storage device on the i-th working day, 30-n be the total number of weekend days in a month, and T be the total number of weekend days in a month. holiday,j The duration of energy storage device usage during the j-th weekend.

[0090] Actual electricity consumption data for each stage is collected, including peak, average, and off-peak periods. A linear regression model is constructed using features extracted from user electricity consumption habit indicators and actual electricity consumption data. These features include, through statistical analysis, any one of the following: weekday / weekend usage time ratio, electricity consumption fluctuation, load fluctuation, and peak, average, and off-peak electricity consumption times. The extracted features are used as input, and the electricity consumption during peak, average, and off-peak periods is used as output to train the model. After training, Bayesian methods are used to correct the predicted values. The prior distribution of model parameters is obtained based on historical data, and the model is then used to refine the model based on new observations. The likelihood function of the model is calculated using measured data. Bayes' theorem is used to combine prior probability and likelihood function to calculate the posterior distribution of model parameters. The posterior distribution is then used to correct the predicted values, resulting in corrected predicted values ​​for each time period. The corrected peak electricity consumption prediction is used as the discharge amount of the energy storage device during peak hours, the corrected normal electricity consumption prediction is used as the charging amount of the energy storage device during normal hours, and the corrected off-peak electricity consumption prediction is used as the charging amount of the energy storage device during off-peak hours. Using the trained model, based on feature data, peak electricity consumption, normal electricity consumption, and off-peak electricity consumption are predicted respectively, resulting in peak electricity consumption prediction values, normal electricity consumption prediction values, and off-peak electricity consumption prediction values.

[0091] Formula for calculating predicted electricity consumption:

[0092] y=α0+x_C degree ×α1+x_E degree ×α2+x_T rate ×α3+x_H i ×α4

[0093] Where y is the predicted electricity consumption, and x_C degree x_E degree x_T rate and x_H i C is the feature input to the model. degree E represents the degree of user load fluctuation. degree To account for fluctuations in user electricity consumption, T rate x_H represents the ratio of weekday to weekend usage time. i The user's electricity consumption time is represented by the following values: when i is 1, it represents the user's electricity consumption time during peak hours; when i is 2, it represents the user's electricity consumption time during normal hours; and when i is 3, it represents the user's electricity consumption time during off-peak hours. α0, α1, α2, α3, and α4 are all parameters of the model.

[0094] Extract the feature vector x = [x_C degree , x_E degree , x_T rate , x_H i ]T Substitute these values ​​into the model formula, and find the maximum, median, and minimum values ​​in the predicted electricity consumption for that day. These values ​​will be used as the predicted peak electricity consumption, normal electricity consumption, and off-peak electricity consumption for that day.

[0095] P peak =max(y)

[0096] P valley =min(y)

[0097] P avg =middle(y)

[0098] Where max(y) is the maximum value of the predicted electricity consumption for that day, min(y) is the minimum value of the predicted electricity consumption for that day, middle(y) is the average value of the predicted electricity consumption for that day, and P peak P represents the predicted peak electricity consumption for that day. avg P is the predicted average daily electricity consumption for that day. valley This is the predicted off-peak electricity consumption value for that day.

[0099] Specifically, electricity consumption data for users over a historical month is collected, including daily load, electricity consumption, and usage time. After processing for missing and outlier values, user electricity load data for the month is recorded, including peak, trough, and average values, to obtain a user load fluctuation value. Similarly, user electricity consumption data for the month is recorded, including the average value, to obtain a user electricity consumption fluctuation value. The data is categorized by weekdays and weekends (e.g., weekdays are Monday to Friday, weekends are Saturday and Sunday), and the average usage time for weekdays and weekends is calculated. The weekday-weekend usage time ratio is obtained by comparing the average weekday usage time with the average weekend usage time. Finally, user electricity consumption habit indicators and usage time data are analyzed. Features are extracted from the data to construct a linear regression model. The model takes any one of the following as input: the ratio of weekday to weekend usage time, the degree of electricity consumption fluctuation, the degree of load fluctuation, and the electricity consumption time during peak, normal, and off-peak hours. The extracted features are used as input, and the electricity consumption during peak, normal, and off-peak hours is used as output. The model is trained, and the predicted values ​​are corrected using a Bayesian method. The corrected peak electricity consumption prediction is used as the discharge amount of the energy storage device during peak hours, the corrected normal electricity consumption prediction is used as the charging amount of the energy storage device during normal hours, and the corrected off-peak electricity consumption prediction is used as the charging amount of the energy storage device during off-peak hours. The trained model is then used to obtain the predicted values ​​for peak, normal, and off-peak electricity consumption, respectively.

[0100] The new observation data includes input features and output variables. The input features are the feature data collected after the model is trained, including the ratio of usage time on weekdays to weekends, the degree of electricity consumption fluctuation, the degree of load fluctuation, and any one of the electricity consumption time during peak hours, normal hours, and off-peak hours. The output variables are the actual electricity consumption during peak hours, normal hours, and off-peak hours.

[0101] In one embodiment, S4 includes the following steps:

[0102] By obtaining peak, normal, and off-peak electricity consumption periods from the power system's electricity pricing policy, and acquiring the electricity prices for these periods from the power grid company's agency electricity purchase price list, the predicted peak electricity consumption is used as the discharge amount of the energy storage device during peak hours, the predicted normal electricity consumption is used as the charging amount during normal hours, and the predicted off-peak electricity consumption is used as the charging amount during off-peak hours. This allows us to obtain the costs of charging during off-peak and normal hours and the revenue from discharging during peak hours, thus obtaining the actual settlement cost of the energy storage device.

[0103] Actual settlement fee calculation formula:

[0104]

[0105] Where M represents the actual settlement cost of the energy storage equipment (user energy storage settlement cost), and P... peak M represents the discharge amount during peak hours. peak For peak-hour electricity prices, P avg η represents the charging amount during normal operation, η represents the charging and discharging efficiency of the energy storage device, and M represents the charging amount during normal operation. avg The electricity price during normal periods, P valley For charging during off-peak hours, M valley This refers to the electricity price during off-peak hours.

[0106] Specifically, understand the electricity pricing policy of the local power system, obtain the standards for dividing peak, normal, and off-peak electricity consumption periods, obtain the time range of peak, normal, and off-peak periods according to the pricing policy, and obtain the electricity purchase price list from the local power grid company to obtain electricity price information for peak, normal, and off-peak periods.

[0107] Specifically, for peak-hour discharge: the peak-hour electricity price is multiplied by the corresponding discharge amount to obtain the peak-hour discharge revenue.

[0108] For charging during normal periods, the charging cost during normal periods is calculated by combining the electricity price, charging and discharging efficiency, and the corresponding amount of charging.

[0109] For charging during off-peak hours, the off-peak charging cost is calculated by combining the electricity price, charging and discharging efficiency, and the corresponding amount of electricity charged during the off-peak hours.

[0110] The total charging cost is obtained by adding the charging costs during normal and off-peak periods, while the discharge revenue during peak periods is listed separately. The actual settlement cost of the energy storage device is obtained by comprehensively calculating the total charging cost and the discharge revenue.

[0111] In summary, this invention solves the technical problems of difficulty in extracting multiple feature data from users' electricity consumption habit index data to build a model, lack of data on charging and discharging amounts at different times based on the built model, difficulty in comprehensively calculating costs and benefits obtained at different times, and difficulty in obtaining the actual settlement cost of energy storage equipment.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A user energy storage auxiliary settlement method, characterized in that, The method includes: S1. Acquire and preprocess user-side data, including user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data. S2. Based on user electricity consumption data, user electricity consumption habit index data and energy storage device usage time data, calculate the user load fluctuation degree, user electricity consumption fluctuation degree and weekday to weekend usage time ratio; S3. Obtain electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation, user electricity consumption fluctuation and weekday to weekend usage time ratio, and combined with linear regression model and Bayes' theorem, output and correct the predicted electricity consumption value for each time period. The formula for calculating predicted electricity consumption is as follows: ； in, For the predicted electricity consumption, For the degree of user load fluctuation, For the degree of fluctuation in user electricity consumption, The ratio of weekday to weekend usage time. For users' electricity usage time, , , , and For different model parameters; S4. Based on the corrected electricity consumption forecast for each time period and the electricity price for each time period, calculate the user's energy storage settlement fee. The user's energy storage settlement fee, calculated based on the corrected electricity consumption forecasts for each time period and the electricity price for each time period, includes: Obtain electricity prices for peak, normal, and off-peak periods; The revised peak-hour electricity consumption forecast is used as the discharge amount of the energy storage device during the peak hours, the revised normal-hour electricity consumption forecast is used as the charging amount of the energy storage device during the normal hours, and the revised off-peak-hour electricity consumption forecast is used as the charging amount of the energy storage device during the off-peak-hour. The formula for calculating user energy storage settlement fees is as follows: ； in, Settle energy storage fees for users; This represents the discharge volume during peak hours. Electricity price during peak hours; This is the charging amount during normal periods. To improve the charging and discharging efficiency of energy storage devices. This refers to the electricity price during normal periods. This refers to the amount of electricity charged during off-peak hours. This refers to the electricity price during off-peak hours.

2. The user energy storage auxiliary settlement method according to claim 1, characterized in that, The calculations based on user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data include: User load fluctuation level, user electricity consumption fluctuation level, and weekday / weekend usage time ratio. Extract user power load data within a period from user electricity consumption habit index data, and the user power load data within a period includes the highest peak value, the lowest valley value and the average value; The degree of fluctuation in user load is calculated based on the highest peak, lowest valley, and average value of user power load data within the period. The degree of fluctuation in user electricity consumption is calculated based on user electricity consumption data; Based on the power consumption data of the energy storage equipment, and combined with the number of weekdays and weekends, the ratio of weekday to weekend usage time is calculated.

3. The user energy storage auxiliary settlement method according to claim 2, characterized in that, The calculation of user load fluctuation based on the highest peak, lowest trough, and average user power load data within the period includes: The degree of load fluctuation of a user is calculated by using the user's highest load peak value, lowest load trough value, average load value, and daily load value within a period.

4. The user energy storage auxiliary settlement method according to claim 2, characterized in that, The calculation of the fluctuation level of user electricity consumption based on user electricity consumption data includes: The degree of fluctuation in a user's electricity consumption is calculated by using the user's daily electricity consumption within a period and the user's average electricity consumption within the period.

5. A user energy storage auxiliary settlement method according to claim 2, characterized in that, The calculation of the weekday to weekend usage ratio based on the power consumption data of the energy storage device, combined with the number of weekdays and weekends, includes: Acquire daily electricity consumption data for users within a period, and extract features from user electricity consumption data and electricity consumption duration data to obtain average daily electricity consumption and standard deviation of electricity consumption; Based on the average daily electricity consumption and the standard deviation of electricity consumption, and combined with machine learning algorithms, a classification model is trained, and the attributes of each day are output using the classification model. Record the number of times each day is marked as a workday to obtain the total number of workdays within the period. Then, calculate the total number of weekend days within the period based on the difference between the total number of days within the period and the total number of workdays within the period. The ratio of weekday to weekend usage time is calculated by using the total number of working days in the cycle, the usage time of energy storage equipment on each working day, the total number of weekend days in the cycle, and the usage time of energy storage equipment on each weekend.

6. The user energy storage auxiliary settlement method according to claim 1, characterized in that, The acquisition of electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation, user electricity consumption fluctuation, and the ratio of weekday to weekend usage time, combined with a linear regression model and Bayes' theorem, outputting and correcting the predicted electricity consumption values ​​for each time period includes: Collect electricity consumption data for different time periods, including peak hours, normal hours, and off-peak hours; Using user load fluctuation, user electricity consumption fluctuation, weekday vs. weekend usage time ratio, and electricity usage time data as inputs and electricity consumption as output, the linear regression model is trained to obtain the prediction model. The predicted electricity consumption values ​​for each time period are output using the prediction model, and the predicted values ​​of the prediction model are corrected based on the Bayesian algorithm.

7. A user energy storage auxiliary settlement method according to claim 6, characterized in that, The predicted electricity consumption values ​​for each time period output by the prediction model include: In the predicted electricity consumption output by the prediction model, the maximum, median and minimum values ​​are used as the predicted electricity consumption values ​​for peak hours, normal hours and off-peak hours of the day, respectively.

8. A user energy storage auxiliary settlement method according to claim 6, characterized in that, The correction of the predicted values ​​of the prediction model based on the Bayesian algorithm includes: Obtain the prior probability of the prediction model, and calculate the likelihood function of the prediction model based on the observation data corresponding to the prediction model; Using Bayes' theorem, and combining the prior probability and likelihood function of the prediction model, the posterior distribution of the prediction model parameters is calculated, and the prediction values ​​of the prediction model are corrected using the posterior distribution.

9. A user energy storage auxiliary settlement system, used to implement the user energy storage auxiliary settlement method according to any one of claims 1-8, characterized in that, The system includes a data acquisition module, a feature data calculation module, a user electricity consumption prediction module, and an energy storage electricity fee settlement module. The data acquisition module is used to acquire and preprocess user-side data, which includes user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data. The feature data calculation module is used to calculate the degree of user load fluctuation, the degree of user electricity fluctuation, and the ratio of weekday to weekend usage time based on user electricity consumption data, user electricity consumption habit index data, and energy storage device usage time data. The user electricity consumption prediction module is used to acquire electricity consumption time data for each time period, and based on the electricity consumption time data, user load fluctuation, user electricity consumption fluctuation and weekday to weekend usage time ratio, and combined with linear regression model and Bayes' theorem, output and correct the electricity consumption prediction value for each time period. The energy storage electricity bill settlement module is used to calculate the user's energy storage settlement fee based on the corrected electricity consumption forecast for each time period and the electricity price for each time period.

Images