Electricity price scheme determination method and device, computer equipment and storage medium

[0026] In order to make the purpose, technical solutions, and advantages of this application clearer, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the application, and not used to limit the application.

[0027] The method for determining the electricity price plan provided in this application can be applied to figure 1 In the application environment shown. Wherein, the terminal 102 communicates with the server 104 through the network through the network.

[0028] Specifically, the server 104 may analyze the user's electricity bill under different electricity price plans according to the user's electricity consumption data, and recommend a suitable electricity price plan for the user. The server 104 can provide a client or page, and the terminal 102 can log in or access the provided client or page to obtain data such as information related to the recommended electricity price plan. The server can obtain the electricity forecast data corresponding to the target user. The electricity forecast data may include the historical electricity consumption data corresponding to the target user in the historical time period. The electricity forecast data is input into the trained target electricity consumption forecast model, and the target is predicted. The user’s target predicted electricity consumption in the target time period obtains a set of candidate electricity price calculation schemes. The candidate electricity price calculation scheme set may include multiple candidate electricity price calculation schemes. The electricity charge is calculated according to each candidate electricity price calculation scheme and the target predicted electricity consumption. Obtain the predicted electricity rate corresponding to each candidate electricity price calculation plan, and filter the candidate electricity price calculation plan set to obtain the target electricity price calculation plan corresponding to the target user in the target time period according to the predicted electricity rate corresponding to each candidate electricity price calculation plan. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server 104 may be implemented by an independent server or a server cluster composed of multiple servers.

[0029] In some embodiments, such as figure 2 As shown, a method for determining an electricity price plan is provided, which is applied to figure 1 Take the server in as an example for description, including the following steps:

[0030] S202: Obtain power prediction data corresponding to the target user, where the power prediction data includes historical power consumption data corresponding to the target user in a historical time period.

[0031] Specifically, the target user may be a resident user, and the electricity forecast data refers to data used for electricity forecasting, and may include historical electricity consumption data corresponding to the target user in a historical time period. The historical time period may be, for example, January 1, 2018. Number to January 31, 2018.

[0032] In some embodiments, the power forecast data further includes at least one of weather data, user power supply plan data, holiday data, or user power outage data. The user's power data, weather data, user power supply plan data, holiday data, and user power outage data can be obtained from various business systems; according to the power data, the user's time-sharing power consumption, daily power consumption and monthly power consumption can be calculated; Through the processing of weather data, the temperature difference over the same period can be calculated. Through the data processing of the power supply plan, invalid data is eliminated; through the data processing of holidays and power outages, valid data is filtered out. Thus, effective electricity forecast data is obtained.

[0033] S204: Input power prediction data into the trained target power consumption prediction model, and predict the target predicted power consumption of the target user in the target time period.

[0034] Specifically, the target time period is a time period in which electricity consumption needs to be predicted, and may be a time in the future, for example, the period from the beginning of the next month to the end of the next month. The target predicted power consumption may include time-sharing power consumption. Time-sharing refers to the whole hour, such as 24 whole points in a day. The target power consumption prediction model may be obtained by training the neural network model through historical power consumption data. The output of the target power consumption prediction model can be the predicted value of time-sharing power consumption, and the predicted value of daily power consumption and the predicted value of monthly power consumption can be calculated according to the output predicted value of time-sharing power consumption. It can also calculate peak power, flat power and valley power. Among them, the peak power refers to the power consumption corresponding to the peak period, the average power refers to the power consumption corresponding to the normal period, and the valley power refers to the power consumption corresponding to the valley period. The time periods included in the peak period are, for example, 9:00-11:30, 14:00-16:30, and 19:00-21:00; the time periods included in the normal period are, for example: 7:00-9:00, 11 :30-14:00, 16:30-19:00, and 21:00-23:00; the time period included in the valley period is, for example, 23:00 of the day to 7:00 of the next day.

[0035] S206: Obtain a set of candidate electricity price calculation schemes, where the candidate electricity price calculation scheme set includes multiple candidate electricity price calculation schemes.

[0036] Specifically, the set of candidate electricity price calculation schemes is a set of electricity price calculation schemes selectable by the target user, and is specifically determined according to the calculation scheme of the power supply bureau corresponding to the target user. For example, the set of candidate electricity price calculation schemes may include a tiered electricity price scheme and a time-of-use electricity price scheme.

[0037] S208: Calculate electricity charges according to each candidate electricity price calculation scheme and the target predicted power consumption, and obtain predicted electricity charges corresponding to each candidate electricity price calculation scheme.

[0038] Specifically, the predicted electricity rate corresponding to the tiered electricity price plan can be calculated according to the tiered electricity price scheme and the target predicted electricity consumption, and the predicted electricity rate corresponding to the time-of-use electricity price scheme can be calculated based on the time-of-use electricity price scheme and the target predicted electricity consumption.

[0039] For example, assuming that the peak power corresponding to the target predicted power consumption is X, the average power is Y, and the valley power is Z, calculate and compare the electricity charges corresponding to the first, second, and third gears in summer and non-summer:

[0040] (1) First gear in summer: X+Y+Z <=260

[0041] The electricity fee corresponding to the time-of-use tariff plan: 108.806875/100*X+66.286875/100*Y+33.576875/100*Z;

[0042] Electricity fee corresponding to the tiered price scheme: 66.286875/100*X+66.286875/100*Y+66.286875/100*Z;

[0043] Electricity fee difference: (108.806875-66.286875)/100*X-(66.286875-33.576875)/100*Z

[0044] =42.52/100*X-32.71/100*Z;

[0045] (2) Second gear in summer: 260 <=600

[0046] The electricity fee corresponding to the time-of-use tariff plan: 108.806875/100*X+66.286875/100*Y+33.576875/100*Z+(X+Y+Z-260)*5/100

[0047] Electricity fee corresponding to the tiered price scheme: 66.286875/100*260+(X+Y+Z-260)*71.286875/100

[0048] Electricity cost difference: time-sharing algorithm-ladder algorithm = 42.52/100*X-32.71/100*Z

[0049] (3) Third gear in summer: 600

[0050] Electricity fee corresponding to the time-of-use price plan: 108.806875/100*X+66.286875/100*Y+33.576875/100*Z+(600-260)*5/100+(X+Y+Z-600)*30/100

[0051] Electricity fee corresponding to the tiered price scheme: 66.286875/100*260+71.286875/100*340+(X+Y+Z-600)*96.286875/100

[0052] Electricity cost difference: time-sharing algorithm-ladder algorithm = 42.52/100*X-32.71/100*Z.

[0053] (4) Non-summer electricity consumption can be analyzed and calculated in the same way as the electricity fee corresponding to the time-of-use electricity price plan, the electricity fee corresponding to the tiered electricity price plan, and the electricity fee difference.

[0054] S210: According to the predicted electricity rates corresponding to the respective candidate electricity price calculation schemes, the target electricity price calculation scheme corresponding to the target user in the target time period is obtained by screening from the set of candidate electricity price calculation schemes.

[0055] Specifically, the solution with the lowest predicted electricity rate in the set of candidate electricity price calculation solutions may be used as the target electricity price calculation solution.

[0056] In some embodiments, it is possible to analyze the influencing factors of the electricity price, analyze the electricity price marketing factors, and recommend a target electricity price calculation solution for the user to save electricity costs.

[0057] In the above method for determining the electricity price plan, the electricity forecast data corresponding to the target user is obtained, and the electricity forecast data is input into the trained target electricity consumption prediction model to predict the target forecast electricity consumption of the target user in the target time period, and obtain candidates Electricity price calculation scheme set, calculate electricity cost according to each candidate electricity price calculation scheme and target predicted electricity consumption, and obtain the predicted electricity cost corresponding to each candidate electricity price calculation scheme. According to the predicted electricity cost corresponding to each candidate electricity price calculation scheme, calculate the electricity cost from the candidate electricity price calculation scheme. The target electricity price calculation scheme corresponding to the target user in the target time period is filtered from the collection. Therefore, a candidate electricity price calculation scheme suitable for the target user can be recommended to the target user, with high flexibility, and can reduce the electricity cost of the target user.

[0058] In some embodiments, step S210 is to screen and obtain the target electricity price calculation scheme corresponding to the target user in the target time period from the candidate electricity price calculation scheme set according to the predicted electricity price corresponding to each candidate electricity price calculation scheme. Respectively corresponding to the predicted electricity rate, the candidate electricity rate calculation plan with the smallest predicted electricity rate is selected from the set of candidate electricity rate calculation plans, as the target electricity price calculation plan corresponding to the target user in the target time period.

[0059] In the above embodiments of the present application, the candidate electricity price calculation solution with the smallest predicted electricity rate is used as the target electricity price calculation solution corresponding to the target user in the target time period, so that the predicted minimum electricity rate solution can be recommended to the user, so that the user can reduce the electricity cost.

[0060] In some embodiments, such as image 3 As shown, the method also includes:

[0061] S302: Compare the target electricity price calculation scheme with the current electricity price calculation scheme corresponding to the target user to obtain a comparison result.

[0062] S304: When the comparison result is inconsistent, push the push information corresponding to the target electricity price calculation scheme to the terminal corresponding to the target user.

[0063] Specifically, the push information may include information such as electricity charges corresponding to the current electricity price calculation scheme corresponding to the target time period, and electricity charges corresponding to the target electricity price calculation scheme. The comparison result may be that the target electricity price calculation scheme is the same (consistent) with the current electricity price calculation scheme corresponding to the target user, or the target electricity price calculation scheme is different from the current electricity price calculation scheme corresponding to the target user (inconsistent). When the target electricity price calculation scheme is different from the current electricity price calculation scheme corresponding to the target user, the push information corresponding to the target electricity price calculation scheme is pushed to the terminal corresponding to the target user.

[0064] In some embodiments, when the comparison result is inconsistent, the target can be used to determine the residential users who can choose to adjust the electricity price plan. Thus, it is possible to count the residential users who can choose to adjust the electricity price plan, and recommend suitable electricity price plans for these users.

[0065] In the above embodiment of the present application, when the comparison result is inconsistent, the push information corresponding to the target electricity price calculation scheme is pushed to the terminal corresponding to the target user, which enables the user to evaluate the impact of different electricity tariff packages on itself, and enables the user to accurately recognize Future electricity consumption, so as to choose a suitable electricity plan to reduce electricity costs.

[0066] In some embodiments, such as Figure 4 As shown, step S208 is to calculate electricity charges according to each candidate electricity price calculation scheme and the target predicted electricity consumption, and obtain the predicted electricity charges corresponding to each candidate electricity price calculation scheme, including:

[0067] S402: Obtain the electricity price calculation period corresponding to the candidate electricity price calculation scheme.

[0068] S404: Perform statistics according to the target predicted power consumption corresponding to the multiple target time periods corresponding to the electricity price calculation period to obtain the statistical predicted power consumption.

[0069] S406: Perform electricity fee calculation according to the candidate electricity price calculation scheme and the statistically predicted electricity consumption, and obtain the predicted electricity fee corresponding to each candidate electricity price calculation scheme.

[0070] Specifically, the electricity price calculation period of the candidate electricity price calculation scheme refers to the length of time corresponding to the electricity consumption on which the electricity price is determined. For example, the power supply bureau determines the user's electricity price calculation plan based on the user's daily power consumption, and one day is an electricity price calculation cycle. The duration of the electricity price calculation cycle is longer than the target time period. For example, an electricity price calculation cycle can be one day and the target time period is one hour. That is, the model outputs time-sharing power consumption, and the duration is relatively short, which can make the model output The forecasted electricity consumption is more accurate. Therefore, it is necessary to add up the electricity for each hour of the day to calculate the statistical electricity consumption, and calculate the electricity bill based on the candidate electricity price calculation plan and the statistical forecast electricity consumption to obtain the predicted electricity bill for the day when the candidate electricity price calculation plan is used. .

[0071] In some embodiments, such as Figure 5 As shown, the method further includes the step of obtaining the target power consumption prediction model, and the step of obtaining the target power consumption prediction model includes:

[0072] S502: Acquire first power consumption data corresponding to the first time period of the training user and second power consumption data corresponding to the second time period of the training user, where the first time period is before the second time period.

[0073] Specifically, the training user may refer to a resident user to which the data used to train the model belongs, that is, the model is trained by acquiring data related to the training user. The first time period and the second time period can be historical time periods of the same period. For example, the first time period is from January 1, 2017 to January 31, 2017, and the second time period is from January 1, 2018 to 2018. January 31 of the year. The first time period and the second time period can also be historical time periods of the same duration. For example, the first time period is from January 1, 2017 to January 31, 2017, and the second time period is March 1, 2017. Number to March 31, 2018. The first power consumption data may be the power consumption corresponding to the first time period, and the second power consumption data may be the power consumption corresponding to the second time period. Of course, the training users can also be other types of users besides resident users.

[0074] S504: Obtain a training feature of the trained user according to the first power consumption data, obtain a training label of the trained user according to the second power consumption data, and obtain a training sample according to the training feature and the training label.

[0075] Specifically, the first power consumption data can be used as the training feature. It is also possible to extract data features based on the first electricity consumption data as training features. The second power consumption data can be used as a training label. The training feature and the training label can be combined as a training sample.

[0076] In some embodiments, the training feature may include one of the power consumption of the same period of time, the power consumption of the same period of the last N days, the temperature of the same region in the same period, the temperature of the same region in the last N days, the same period holiday data, and the same period of power outage data, or Many kinds.

[0077] In some embodiments, the training characteristics of the trained user may be obtained according to the first power consumption data and other data, where the other data may include one or more of weather data, holiday data, and power outage data. The weather data may be, for example, daily weather data in the area where the resident user is located, obtained from a weather website. The holiday data may include, for example, Saturday, Sunday, and legal holidays.

[0078] In some embodiments, data related to forming training samples can be obtained from various platforms. For example, you can obtain 24 hourly electricity indications of residential users from the measurement automation system, including: power supply unit number, user number, data time, asset number, meter code, and comprehensive magnification; obtain weather data from the weather website, including: area name , Data time and average temperature; Synchronize the power supply plan data from the marketing system, including: user number, power consumption category, contract capacity, operating capacity, reporting date of shutdown, reporting date, power price plan, power price and user status; The azimuth system synchronizes the power outage data of residential users, including: power supply unit number, user number, power outage time and power restoration time. Among them, you can use the user number of the power supply plan ledger or the user number of the 24 time-point electrical energy representations as the query conditions to query and output the electricity consumption information.

[0079] In some embodiments, data related to forming training samples can be obtained from various platforms, and the obtained data can be processed to retain the available data. For example, it is possible to filter the 24 hour electricity indications, weather data, power supply plan data and resident user power outage data synchronized from the metering automation system, weather website, marketing system and customer all-round system to delete invalid Abnormal data, keep the available data, and calculate the data information that can be used in the feature model.

[0080] In some embodiments, data filtering processing can be performed on the power data of the user. Specifically, the missing zero-point data can be filled in to eliminate or automatically complete the resident users who have less than 24 zero-point data per day (for example, complete with zero power), and eliminate zero-power abnormal data, such as :The 0 o’clock data of the next day is smaller than the 0 o’clock data of today.

[0081] In some embodiments, the daily time-sharing power consumption, daily power consumption, and monthly power consumption of the residents can be calculated according to the processed power data. Specifically, the processed electricity quantity data can be summarized on an hourly basis. For example, the calculation method of the time-sharing power consumption may be: the time-sharing power consumption = the power data at 24 o'clock of the day-the power data at 23 o'clock of the day. The processed power data can be summarized by daily power. For example, the calculation method of daily electricity consumption may be: daily electricity consumption = electricity at 0 o'clock the next day-electricity at 0 o'clock today. Among them, multiple tables of residential users are consolidated and summarized on a daily basis, that is, the daily electricity consumption of residential users is a single piece of data. The processed power data can be summarized by daily power. For example, the calculation method of monthly power consumption may be: monthly power consumption = 0:00 power on the 1st of the following month-0:00 power on the 1st of this month.

[0082] In some embodiments, the temperature difference in the same period and the temperature change trend in the last N days can be calculated according to the collected weather temperature data.

[0083] S506: Perform model training according to the training samples to obtain a trained target power consumption prediction model.

[0084] Specifically, the model used in model training can be a neural network algorithm, a genetic algorithm, and other big data analysis algorithms. By inputting training samples into the model, let the model learn the relationship between training features and labels, thereby obtaining the target electricity consumption prediction model. The output of the target power consumption prediction model can include the power corresponding to each day in a month, or the power corresponding to each hour of the day.

[0085] In some embodiments, residential users can be classified, or the electricity consumption data and other data corresponding to users in the same category, analyze and calculate the overall characteristics of users of the same electricity consumption attribute group, and train to obtain the target electricity consumption corresponding to this type of user Forecast model. Of course, it is also possible to use the electricity consumption data and other data corresponding to the single user to analyze and calculate the characteristics of the electricity consumption of the single user, and to train the target electricity consumption prediction model corresponding to the single user.

[0086] In some embodiments, the power data output by the target power consumption prediction model can be obtained, the historical power consumption prediction curve is calculated according to the power data, and the historical power consumption prediction curve is compared with the historical actual power consumption curve to Verify the accuracy of the model. So as to analyze suitable big data analysis algorithms and feature models. Among them, the historical power consumption prediction curve is a curve obtained based on the predicted power of the historical power consumption, and the historical actual power consumption curve is a curve obtained based on the actual power of the historical power consumption. For example, the historical actual power consumption curve may be a curve obtained based on the actual daily power consumption of the previous month. The historical power consumption prediction curve may be a curve obtained according to the predicted power consumption of each day in the previous month output by the target power consumption prediction model. When the historical electricity consumption forecast curve is similar to the historical actual electricity consumption curve, the accuracy of the model can be considered to be high; when the historical electricity consumption forecast curve is less similar to the historical actual electricity consumption curve, the model can be considered accurate The rate is low.

[0087] In some embodiments, query conditions may be input through the interface, and the target power consumption prediction model may return the predicted power consumption results to the interface through the query conditions input in the interface, and the interface may also accept and display corresponding historical power consumption data. The query conditions may include, for example, the power supply unit, the unit of the default power supply personnel, the user number, and the user unit. Query can have permission restrictions. For example, the upper-level unit has the permission to query the data of the lower-level unit, and the lower-level unit has no permission to query the data of the upper-level unit; the upper-level unit has the permission to query the users of the lower-level unit, and the lower-level unit does not have the permission to query the users of the upper-level unit. The query result can include historical daily electricity consumption data and model forecast daily electricity consumption data at the same time. The interface can display historical daily electricity consumption data and model forecast daily electricity consumption data at the same time through a table, and can also display historical daily electricity consumption data and model forecast daily electricity consumption data at the same time through a curve. In order to achieve the comparison between historical daily electricity consumption data and forecast daily electricity consumption data of the model at the same time, to verify the accuracy of the model. For example, if the difference in power consumption between the curves is small, there are no particularly prominent abnormal points or scattered points, and the power consumption is evenly distributed, the model has a higher accuracy.

[0088] In some embodiments, the query conditions can be input through the interface. The date in the query conditions can be a date in the future. The target electricity consumption prediction model can obtain the query conditions input from the interface, and return the forecast of the electricity at a future date based on the query conditions. result. For example, the query conditions may include: the unit of the default person of the power supply unit, the user number, the user unit, and the predicted electricity consumption month, where the predicted electricity consumption month is the month for which forecast electricity consumption data is required, such as next month. The response result of the query condition may include user information and power prediction results. The user information may include, for example, user basic profile, user number, user name, power consumption type, contract capacity, and operating capacity. The power prediction result may include the predicted daily power consumption data for each day of the predicted month. The interface can display the response results of the query conditions.

[0089] In the foregoing embodiment of the present application, the training feature of the training user is obtained through the first power consumption data corresponding to the first time period of the training user, and the training label of the training user is obtained according to the second power consumption data corresponding to the second time period of the training user , Obtain training samples based on training features and training tags, so that the model can be trained based on historical power consumption data, and the accuracy of the model can be verified through historical power consumption data to obtain a target power consumption prediction model with high accuracy.

[0090] In some embodiments, there are multiple training samples corresponding to the same training user, such as Figure 6A As shown, step S506 is to perform model training according to the training samples to obtain the trained target power consumption prediction model including:

[0091] S602: Input the first power consumption data in each training sample into the power consumption prediction model to be trained to obtain training predicted power consumption data corresponding to the training user in the second time period.

[0092] Specifically, the power consumption prediction model to be trained may be a neural network model, and the power consumption prediction model to be trained can be trained through multiple training samples to obtain the trained target power consumption prediction model. The first power consumption data in each training sample can be input into the power consumption prediction model to be trained. The power consumption prediction model to be trained calculates each first power consumption data and outputs the corresponding first power consumption data. The training prediction power consumption data corresponding to the two time periods. A training sample can get a training prediction power consumption data. For example, if the first power consumption data is the time-sharing power consumption, the training predicted power consumption data is the predicted value of the corresponding time-sharing power consumption.

[0093] S604: Sort the training predicted power consumption data corresponding to the same training user in chronological order to obtain a training predicted power consumption data sequence.

[0094] Specifically, the training prediction power consumption data has a time sequence relationship. For example, each training prediction power consumption data is a prediction of the power consumption corresponding to different time sharing. The training prediction power consumption data can be sorted in chronological order to form a training prediction power consumption data sequence.

[0095] S606: Sort each second power consumption data corresponding to the same training user in chronological order to obtain a second power consumption data sequence.

[0096] Specifically, there is a time sequence relationship between the second power consumption data. For example, each second power consumption data is the actual power consumption corresponding to different time sharing. The second power consumption data can be sorted in chronological order to form a second power consumption data sequence. Therefore, the second power consumption data sequence is the actual value, and the training predicted power consumption data sequence is the predicted value of the second power consumption data sequence predicted by the model.

[0097] S608: Calculate a model loss value according to the difference between the training prediction power consumption data sequence and the second power consumption data sequence, and the model loss value and the difference have a negative correlation.

[0098] Specifically, the distance between the training prediction power consumption data sequence and the second power consumption data sequence, such as the Euclidean distance, can be calculated as the loss value of the model.

[0099] S610: Adjust the parameters of the power consumption prediction model to be trained according to the model loss value to obtain the target power consumption prediction model.

[0100] Specifically, the gradient descent method can be used to adjust the parameters of the power consumption prediction model to be trained in the direction of the model loss value to obtain the parameters of the power consumption prediction model to be trained that minimizes the model loss value, and this parameter corresponds to The power consumption prediction model to be trained is the target power consumption prediction model.

[0101] In the above embodiment of the present application, the second power consumption data sequence is the actual value, and the training predicted power consumption data sequence is the predicted value of the second power consumption data sequence predicted by the model. Therefore, it can be based on the second power consumption data sequence Train the predicted power consumption data sequence to obtain the model loss value, and obtain the target power consumption prediction model with high accuracy.

[0102] Of course, during model training, the historical electricity consumption corresponding to each of the 24 time-sharing can be input into the electricity consumption prediction model to be trained at the same time, and the model is trained to obtain the electricity consumption corresponding to the 24 time-sharing. Predictive value.

[0103] In some embodiments, the query conditions can be input through the interface, and the electricity consumption data and the electricity fee calculation process corresponding to each electricity price scheme corresponding to the query conditions can be displayed according to the query conditions. For example, the query conditions may include: power supply unit, user number, predicted peak power, predicted average power, predicted valley power, and predicted power consumption month. The power supply unit defaults to the login personnel unit. The system corresponding to the interface can respond to the request corresponding to the query condition, query the user's basic file according to the user number, and determine the user's power consumption attribute (normal power consumption, large amount of electricity or high demand power consumption) according to the user's power consumption attribute Identify the electricity price options available to the user. According to the input predicted peak power, predicted flat power and predicted valley power, calculate the electricity bill according to different electricity price schemes, and return the response result to the interface. The response result may include user profile information, electricity tariff calculation process of different electricity tariff schemes, and optimal electricity tariff analysis result of electricity tariff. The interface can display the response results.

[0104] In some embodiments, such as Figure 6B As shown, a method for determining an electricity price plan is provided, including:

[0105] Step 1. Prepare data.

[0106] Specifically, extract user power data, weather data, user power supply plan data, holiday data, and user power outage data from each associated business system;

[0107] Step 2. Process the data.

[0108] Specifically, calculate the user's time-sharing power consumption, daily power consumption, and monthly power consumption based on the power data, calculate the temperature difference in the same period based on the weather data; process the power supply plan data, and eliminate invalid data; process the holiday data; Process the power failure data; output the processed valid data set.

[0109] Step 3. Construct a feature model.

[0110] Specifically, the model constituent factors are defined, the electricity consumption prediction special diagnosis model is constructed, and the user's predicted electricity consumption is output.

[0111] Step 4. Verify the model.

[0112] Specifically, the historical power consumption curve is calculated according to the characteristic model and the actual power consumption curve is compared to verify the accuracy of the model.

[0113] Step 5. Predict the power.

[0114] Specifically, the historical electricity consumption prediction curve is calculated according to the characteristic model and the historical actual electricity consumption curve is compared to verify the accuracy of the model.

[0115] Step 6. Analyze the factors affecting electricity prices.

[0116] Specifically, analyze the electricity price marketing factors of ordinary electricity, and output users who can optimize the adjustment of electricity prices.

[0117] Step 7. Analyze the optimal electricity price for residential users.

[0118] Specifically, using the predicted electricity consumption of residents as input parameters, the electricity prices of residents are analyzed and calculated, and the results are compared to output the optimal plan.

[0119] Step 8, end.

[0120] Specifically, the determination of the residential user electricity price plan ends. Complete the closed-loop management determined by the residential user electricity price plan.

[0121] Such as Figure 6C As shown, a system for determining an electricity price plan is provided. The system includes a data preparation module, a characteristic model building module, a model verification module, an electricity forecasting module, an electricity price influencing factor analysis module, and a residential user electricity consumption analysis module. The back-end business system in the figure refers to a collection of business management systems that provide back-end information support for the electricity price plan determination system.

[0122] It should be understood that although the steps in the flowcharts of the foregoing embodiments are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless specifically stated in this article, the execution of these steps is not strictly restricted in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts of the foregoing embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The order of execution of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0123] In some embodiments, such as Figure 7 As shown, a device for determining an electricity price plan is provided, including: an electricity forecast data acquisition module 702, a target forecast electricity consumption acquisition module 704, a candidate electricity price calculation scheme set acquisition module 706, a forecast electricity tariff acquisition module 708, and a target electricity price calculation scheme. Module 710, where:

[0124] The power prediction data acquisition module 702 is configured to acquire power prediction data corresponding to the target user. The power prediction data includes historical power consumption data corresponding to the target user in a historical time period.

[0125] The target predicted power consumption obtaining module 704 is configured to input the power prediction data into the trained target power consumption prediction model to predict the target predicted power consumption of the target user in the target time period.

[0126] The candidate electricity price calculation solution set acquisition module 706 is configured to acquire a candidate electricity price calculation solution set, and the candidate electricity price calculation solution set includes multiple candidate electricity price calculation solutions.

[0127] The predicted electricity rate obtaining module 708 is configured to calculate the electricity rate according to each candidate electricity price calculation scheme and the target predicted power consumption, and obtain the predicted electricity rate corresponding to each candidate electricity price calculation scheme.

[0128] The target electricity price calculation scheme obtaining module 710 is configured to screen and obtain the target electricity price calculation scheme corresponding to the target user in the target time period from the set of candidate electricity price calculation schemes according to the predicted electricity rates corresponding to each candidate electricity price calculation scheme.

[0129] In some embodiments, the target electricity price calculation scheme obtaining module 710 is further configured to screen out the candidate electricity price calculation scheme with the smallest predicted electricity price from the set of candidate electricity price calculation schemes according to the predicted electricity costs corresponding to each candidate electricity price calculation scheme, as the target user The target electricity price calculation scheme corresponding to the target time period.

[0130] In some embodiments, the device further includes:

[0131] The comparison result obtaining module is used to compare the target electricity price calculation scheme with the current electricity price calculation scheme corresponding to the target user to obtain the comparison result.

[0132] The push information push module is used to push the push information corresponding to the target electricity price calculation scheme to the terminal corresponding to the target user when the comparison result is inconsistent.

[0133] In some embodiments, the predicted electricity bill obtaining module 708 includes:

[0134] The electricity price calculation period obtaining unit is configured to obtain the electricity price calculation period corresponding to the candidate electricity price calculation scheme.

[0135] The statistical forecast power consumption obtaining unit is used to perform statistics based on the target forecast power consumption corresponding to the multiple target time periods corresponding to the electricity price calculation period to obtain the statistical forecast power consumption.

[0136] The predicted electricity rate obtaining unit is used to calculate the electricity rate based on the candidate electricity price calculation plan and the statistically predicted electricity consumption, to obtain the predicted electricity rate corresponding to each candidate electricity price calculation plan.

[0137] In some embodiments, the device further includes a target power consumption prediction model obtaining module, and the target power consumption prediction model obtaining module includes:

[0138] The power consumption data acquisition unit is used to acquire the first power consumption data corresponding to the training user in the first time period and the second power consumption data corresponding to the training user in the second time period, where the first time period is in the second time period prior to.

[0139] The training sample obtaining unit is configured to obtain training features of the trained user according to the first power consumption data, obtain training labels of the trained user according to the second power consumption data, and obtain training samples according to the training features and training labels.

[0140] The target power consumption prediction model obtaining unit is used to perform model training according to the training samples to obtain the trained target power consumption prediction model.

[0141] In some embodiments, there are multiple training samples corresponding to the same training user, and the target power consumption prediction model obtaining unit is further used to input the first power consumption data in each training sample into the power consumption prediction to be trained. In the model, the training predicted power consumption data corresponding to the training user in the second time period is obtained; each training predicted power consumption data corresponding to the same training user is sorted in chronological order to obtain the training predicted power consumption data sequence; Each second power consumption data corresponding to the same training user is sorted in chronological order to obtain the second power consumption data sequence; the model loss is calculated according to the difference between the training predicted power consumption data sequence and the second power consumption data sequence The model loss value has a negative correlation with the difference; the parameters of the power consumption prediction model to be trained are adjusted according to the model loss value to obtain the target power consumption prediction model.

[0142] Regarding the specific definition of the electricity price scheme determination device, please refer to the above definition of the electricity price scheme determination method, which will not be repeated here. Each module in the device for determining a power price scheme can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the form of hardware or independent of the processor in the computer equipment, or may be stored in the memory of the computer equipment in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.

[0143] In some embodiments, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as Figure 8 Shown. The computer equipment includes a processor, a memory, and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by the processor to realize a method for determining an electricity price plan.

[0144] Those skilled in the art can understand, Figure 8 The structure shown in is only a block diagram of part of the structure related to the solution of the application, and does not constitute a limitation on the computer equipment to which the solution of the application is applied. The specific computer equipment may include more or Fewer parts, or combine some parts, or have a different arrangement of parts.

[0145] In some embodiments, a computer device is provided, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps of the method for determining the electricity price plan when the processor executes the computer program.

[0146] In some embodiments, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program is executed by a processor to implement the steps of the method for determining the electricity price plan.

[0147] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage. In the medium, when the computer program is executed, it may include the procedures of the above-mentioned method embodiments. Wherein, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and/or volatile memory. Non-volatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not a limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Channel (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0148] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should It is considered as the range described in this specification.

[0149] The above-mentioned embodiments only express a few implementation modes of the present application, and their description is relatively specific and detailed, but they should not be understood as a limitation on the scope of the invention patent. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can be made, and these all fall within the protection scope of this application. Therefore, the scope of protection of the patent of this application shall be subject to the appended claims.