Energy payment early warning method, electronic device and computer readable storage medium

Abstract

The invention provides a method for automatically reminding a merchant owner to pay before arrearage of an energy meter in a park. According to the method, the recharge balance of each energy meter is obtained regularly through a rear-end timed task, the energy consumption value of the next day is predicted in combination with an energy efficiency future consumption prediction algorithm, whether the arrearage risk exists in the day is judged, and if the arrearage risk exists, an alarm message is pushed to a merchant owner through multiple channels such as a WeChat official account, a short message and an email, so that the energy meter can be automatically and accurately recharged. A recharging link is provided in the alarm message, and the user can directly switch to a recharging interface of the merchant through the link, so that the payment efficiency and experience of the merchant are improved. In addition, the method and the device also support personalized configuration of alarm rules and alarm strategies for the park commercial tenants, and push of alarm messages is carried out by judging energy consumption alarm conditions independently set by the park commercial tenants. The invention further provides a function of allowing setting of the number of pushing times of the alarm message and avoiding repeated reminding and judgment of the message.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet of Things technology, and more specifically, to an energy payment early warning method, an electronic device, and a computer-readable storage medium. Background Art

[0002] Traditional energy meter management in industrial parks relies primarily on manual inspections and payment reminders, a method that is both inefficient and prone to errors. With the development of IoT technology, energy meters now have the ability to read data remotely. However, effectively utilizing this data to provide early warnings of impending overdue payments and improve industrial park service and management efficiency remains a major technical challenge. While existing technologies can remotely read energy meter data, they lack intelligent early warning, prediction, and automatic reminder mechanisms, resulting in inefficient and inconvenient payment management. Summary of the Invention

[0003] In order to solve the above technical problems existing in the prior art, the present invention proposes a method for automatic alarm reminder of energy payment in a park based on the Internet of Things platform, which can predict the future consumption of energy meters of park users, and push payment warning messages to their users for meters that are about to be in arrears and need to be paid, providing a reference for users' payment decisions on related energy meters, thereby improving the energy management efficiency and intelligence level of the park.

[0004] Specifically, the first aspect of the present invention provides a computer-implemented energy payment warning method, including steps S11 to S14: Step S11: Obtain the recharge balance of the user's energy meter; Step S12: Obtain the daily energy consumption value related to the energy meter; Step S13: Determine whether there is a risk of arrears in the recharge balance; if so, proceed to step S14: Push an alarm message to the user through at least one channel; if not, return to step S11 after a first predetermined time.

[0005] In some embodiments, obtaining the recharge balance of the energy meter includes: periodically obtaining the recharge balance of the energy meter.

[0006] In some embodiments, periodically obtaining the recharge balance of the energy meter includes: obtaining the recharge balance of the energy meter at a predetermined time every day.

[0007] In some embodiments, the at least one channel includes at least one of a WeChat public account, text messages, and emails.

[0008] In some embodiments, obtaining the daily energy consumption value associated with the energy meter includes: using an energy consumption prediction algorithm to predict the energy consumption value for the next day.

[0009] In some embodiments, the step of using the energy consumption prediction algorithm to predict the energy consumption value for the next day also includes: designing the energy consumption prediction algorithm, and designing the energy consumption prediction algorithm includes: collecting historical energy consumption data and storing it in a memory; preprocessing the historical energy consumption data stored in the memory to obtain preprocessed data that at least includes an energy consumption time series and a corresponding energy consumption amount; constructing an energy consumption prediction model based on the LSTM algorithm; dividing the preprocessed data into a training set and a validation set and inputting them into the energy consumption prediction model based on the LSTM algorithm to train the energy consumption prediction model based on the LSTM algorithm; using the trained model to predict the training set and the validation set to evaluate the model performance, and adjusting the parameters of the energy consumption prediction model based on the LSTM algorithm according to the evaluation results to obtain a model with adjusted parameters; and using the model with adjusted parameters to predict the energy consumption value for the next day.

[0010] In some embodiments, pushing the alarm message to the user through at least one channel further includes: providing the user with a recharge link while pushing the alarm message, so that the user can go to the recharge interface through the recharge link.

[0011] In some embodiments, determining whether the recharge balance has an arrears risk includes: determining whether the user has ever independently configured alarm rules and alarm strategies through a configuration page related to energy consumption alarms on the Internet; if not, determining whether the energy meter has an arrears risk based on the difference between the recharge amount of the user's energy meter and the determined daily energy consumption value; if yes, determining whether the energy meter has an arrears risk based on the personalized alarm configuration table independently configured by the user on the configuration page.

[0012] In some embodiments, pushing an alarm message to a user through at least one channel includes: determining whether the user has ever independently configured alarm rules and alarm policies through a configuration page related to energy consumption alarms on the Internet; if not, pushing the alarm message through a default channel; if yes, pushing the alarm message through a channel personalized configured by the user based on a personalized alarm configuration table independently configured by the user on the configuration page.

[0013] In some embodiments, after pushing the alarm message to the user through at least one channel, the energy payment warning method also includes: determining whether the user has ever independently configured alarm rules and alarm strategies through the configuration page related to energy consumption alarms on the Internet; if not, after a second predetermined time, returning to step S11 to determine the recharge balance of the user's energy meter to determine whether the user recharges after the alarm message is pushed to the user and pushing the alarm message to the user again when the judgment result shows that the user has not recharged or there is still a risk of arrears after recharging; if yes, deciding whether to determine whether the user responds to the alarm message after the alarm message is pushed to the user based on the personalized alarm configuration table independently configured by the user on the configuration webpage.

[0014] In some embodiments, the first predetermined time is greater than the second predetermined time.

[0015] A second aspect of the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-mentioned energy payment reservation method.

[0016] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned energy payment warning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The following drawings are examples for illustration purposes only, in accordance with various disclosed embodiments, and are not intended to limit the scope of the invention.

[0018] Figure 1 A flowchart of a computer-implemented energy payment early warning method according to an embodiment of the present disclosure is shown;

[0019] Figure 2 A flowchart of a computer-implemented energy payment early warning method according to an embodiment of the present disclosure is shown;

[0020] Figure 3 A flow chart of an energy payment alarm method based on an energy consumption prediction algorithm according to an embodiment of the present disclosure is shown;

[0021] Figure 4 A schematic diagram of an initial prediction model architecture according to an embodiment of the present disclosure is shown;

[0022] Figure 5 A flowchart of a method for training a prediction model according to an embodiment of the present disclosure is shown;

[0023] Figure 6A flowchart of a method for verifying a trained prediction model according to an embodiment of the present disclosure is shown;

[0024] Figure 7 A flowchart of a method for energy payment warning according to user personalized configuration according to an embodiment of the present disclosure is shown;

[0025] Figure 8 A schematic diagram showing a client personalized configuration interface according to an embodiment of the present disclosure; and

[0026] Figure 9 A block diagram of an energy bill payment early warning system according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0027] The present disclosure will now be described in more detail with reference to the following examples. It should be noted that the following description of some of the embodiments presented herein is for illustration and description purposes only. It is not intended to be exhaustive or limited to the precise forms disclosed.

[0028] The present disclosure relates to the field of Internet of Things technology, specifically to the field of smart park energy management, energy efficiency prediction, payment reminders and automation technology, and aims to provide a method and system that can automatically monitor park energy consumption, predict payment needs, and send payment reminders to users through multiple channels such as WeChat public accounts, text messages, and emails.

[0029] The inventors realized that in the energy management of a park, traditional energy meters (such as electricity meters, water meters, and gas meters) mainly rely on manual inspection and reminders to owners to pay their bills. This method is inefficient and prone to errors. With the development of Internet of Things technology, although energy meters have realized the function of remotely reading data, how to effectively use this data to provide early warning of overdue bills and improve management efficiency remains a current technical challenge. This disclosure proposes a method for automatically reminding owners to pay their bills before overdue bills are reported for energy meters (including electricity meters, water meters, and gas meters) in a park. For example, this method uses a back-end scheduled task written in Java language to regularly obtain the current balance of each energy meter. Combined with the "energy consumption prediction algorithm" proposed in this disclosure, daily energy consumption values ​​are predicted, such as the energy consumption value for the next day. The program then determines whether there is a risk of overdue bills on that day. If there is a risk of overdue bills, an alarm message is sent to the owner through channels such as WeChat official accounts, text messages, and emails. The alarm message includes a warning of today's overdue bills for the energy meter and provides a recharge link for the warning meter. Users can directly access the recharge interface of the current meter through the link to improve user payment efficiency and experience. Furthermore, the system supports personalized configuration of alarm rules and policies for campus users, enabling users to push alarm messages based on their own defined energy consumption alarm conditions. The system also allows users to set the number of push notifications and verifies the real-time balance before each push to avoid repeated reminders.

[0030] In some embodiments, the present disclosure provides a method and system for monitoring energy consumption of energy meters in a park, which may specifically involve monitoring the current balance of energy meters, recording and updating daily historical energy consumption data (in winter and summer), predicting consumption values ​​within a predetermined period in the future based on the daily historical energy consumption data of the meters, automatically reminding owners to pay at least one day in advance when the meters are about to be in arrears based on the predictions, pushing alarm messages according to the default method to avoid possible arrears risks, or pushing alarm messages according to the owner's personalized configuration settings of personalized alarm rules and alarm strategies to avoid possible arrears risks.

[0031] In some embodiments, the present disclosure provides a computer-implemented method for monitoring energy meter consumption, comprising: obtaining the current balance on the monitored energy meter; calling daily historical energy consumption data; comparing the current balance with the daily historical energy consumption data and determining whether the meter has a risk of arrears in the next week; if so, marking a red alert; if not, recording the comparison results and concluding that there is no risk of arrears in the short term.

[0032] As described above, the energy meter energy consumption monitoring method of the present invention involves, for example, an industrial park, a residential area, or a commercial and residential park; each park involves a large number of users, and each user includes multiple energy meters, mainly including electricity meters, water meters, and gas meters, etc. Therefore, the energy meter monitoring method of the present invention involves the monitoring of each energy meter for each user.

[0033] Figure 1 FIG. 1 shows a flow chart of a computer-implemented energy payment early warning method according to an embodiment of the present disclosure. Figure 1 As shown, the energy payment warning method includes steps S11 to S14: Step S11: Obtain the current remaining recharge balance of the user's energy meter; Step S12: Obtain the daily energy consumption value related to the energy meter; Step S13: Determine whether the recharge balance has an arrears risk; If so, proceed to step S14: Push an alarm message to the user through at least one channel; If not, return to step S11 after a first predetermined time.

[0034] Figure 1 The energy payment warning method shown is only performed for one energy meter of one user. For all users in a park or community, or even a larger area, where each user has multiple energy appliances (for example, at least an electricity meter, a water meter, and a gas meter), the warning can be executed in parallel or serially. Figure 1The energy payment warning method shown in the figure monitors the recharge balance and daily energy consumption of energy meters of all energy appliances of all users within the scope of concern, determines whether there is a risk of arrears based on the monitoring results, and pushes alarm messages to users when there is a possible risk of arrears. It can automatically monitor the energy consumption of all energy appliances of all users in a park, community or a larger area, predict payment needs, and then push alarm messages to users through specific channels. This automatic and regular monitoring method improves the management efficiency of energy consumption, reduces the number of staff, reduces detection costs, and greatly reduces the risk of energy supply interruption caused by arrears.

[0035] In some embodiments, obtaining the recharge balance of the energy meter may be, for example, obtaining the recharge balance of the energy meter on a regular basis. For example, the recharge amount of each energy meter may be obtained every week, every two days, or at a predetermined time every day. The channel for pushing the alarm message to the user in step S14 may be, for example, at least one of a WeChat official account, WeChat, text messages, and emails. For example, as described below, if the user has not personalized the alarm rules and alarm policies, the alarm message is pushed to the user in the system default initialization configuration method, for example, an alarm message is pushed to users at risk of arrears in the form of text messages; if the user has previously personalized the alarm rules and alarm policies, the alarm message is pushed in the user-configured method, for example, the alarm message is pushed using the user-set email as the channel.

[0036] In one embodiment, the daily energy consumption value may be the energy consumption value for the next day. In this case, the current remaining top-up amount of the user's energy meter is compared with the energy consumption value for the next day associated with the energy meter. If the current remaining top-up amount is found to be insufficient to cover the energy consumption value for the next day, an alert message is sent to the user. In this case, the alert message informs the user that there is a possibility of energy supply interruption in the next day and the user must top up the energy meter today.

[0037] In the above embodiment, if there is a risk of arrears, the user may not be able to recharge in time due to time constraints. In some embodiments, the daily energy consumption value of the present invention can be, for example, the energy consumption value for the next week, based on which it can be determined whether there is a risk of arrears one week later. This can give the user more time to recharge, thereby avoiding the harm caused by energy being stopped when recharging is inconvenient.

[0038] All of the above can be set on the user's personalized configuration page. Figure 1The daily consumption values, at least one channel, etc. involved in the steps of obtaining the daily energy consumption values ​​related to the energy meter and pushing the alarm message to the user through at least one channel in the energy payment warning method shown can be set on the user's personalized configuration page to adapt to the user's processing method and meet the user's needs, so as to better avoid the harm caused by the risk of arrears.

[0039] In one embodiment, in order to bring further convenience to the user, a recharge link may be provided to the user while pushing the alarm message, so that the user can go to the recharge interface through the recharge link.

[0040] Figure 2 FIG. 1 shows a flow chart of a computer-implemented energy payment early warning method according to an embodiment of the present disclosure. Figure 2 The energy consumption monitoring method of the energy meter shown includes steps S20 to S28.

[0041] In step S20, the energy consumption amount of the meter is obtained and a database including historical energy consumption data is created.

[0042] In some embodiments, in step S20, the energy consumption amount of each energy meter is obtained and stored in a historical energy consumption data database, in which, for example, the energy consumption amount of each day for at least the past period of time (for example, the past year) is stored (or the daily energy consumption amount is stored, and the daily energy consumption amount is calculated based on the unit price of the unit energy consumption amount; since the unit price of the unit energy consumption amount may change, for example, the electricity price per kilowatt-hour in a specific period may change slightly with the change of power supply capacity, so recording the daily energy consumption amount is more accurate for the judgment of the risk of arrears. The present invention is not limited to this, and the energy consumption amount of the meter or the energy consumption amount of the meter can be recorded in the database as needed). In one embodiment, the database may include a table with the vertical axis being time and the horizontal axis recording the daily energy consumption or the energy consumption amount. Based on this, the average energy consumption or the average energy consumption amount of each day over a period of time (for example, within a year, within a month, within a week) can be calculated as a basis for the subsequent judgment of the risk of arrears. For example, the average energy consumption or the average energy consumption amount of the four seasons of spring, summer, autumn and winter can be calculated based on the rules of the historical consumption data obtained. For example, for electricity meters, the daily energy consumption amount for each season of spring, summer, autumn and winter can be calculated, so that the risk of overdue payments can be judged more reliably and more reliable overdue payment alerts can be provided.

[0043] In step S21, a determination is made as to whether the user of the meter in question has independently configured alarm rules and policies through the energy consumption alarm configuration page of the network system (hereinafter referred to as the Web system). These independently configured alarm rules and policies are what the user desires, is more easily accepted, is less likely to be disturbed, and is more effective. In other words, the determination is made as to whether the user has previously customized the configuration within the Web system. For example, the Web system can be an application designed specifically for all users within a campus to monitor and manage the top-up amounts of all energy appliances used by all users, monitor daily energy consumption, and provide overdue payment reminders. This application can be referred to as an energy early warning app.

[0044] Every user in the park can use the web system (energy warning app) to check the remaining balance of each energy appliance, daily energy consumption, and other information. Users can also log in to the web system to configure their desired alarm rules and policies. For example, users are more sensitive to email content, paying more attention to it, and more likely to take action. Therefore, users can use the web system to set up overdue payment reminders via email. If an energy outage could result in significant losses for users (e.g., merchants), or if recharging the energy appliances involved is large and requires tedious financial work, users may want to receive reminders a week or more in advance of potential overdue payments so they can respond quickly and minimize the possibility of an energy outage. If users can plan their daily routines, they may prefer to receive only a single reminder, as they are confident they will recharge on time and that a missed recharge will not result in significant financial losses. Users may also prefer to receive multiple reminders throughout the day until the web system detects that they have recharged and are no longer at risk of overdue payments. Of course, if the user does not configure the alert rules and policies through the web system's energy consumption alarm configuration page, the web system will issue a default alert to the user indicating a potential arrears. For example, the default alert method includes notifying the user via WeChat when a potential arrears is present; two reminders are issued daily. Before the second reminder, the system will recheck the top-up balance. If the user has already responded to the previous alert, meaning they have already topped up and are no longer at risk, no further reminder will be issued.

[0045] For example, a commercial and residential campus includes both commercial users and ordinary residents. In this case, the risk of overdue payments is relatively high for commercial users, and the losses and harm caused by energy outages due to overdue payments are relatively large. Therefore, to minimize this risk, commercial users must ensure that the alarms issued by the web system truly alert users and that users can promptly or conveniently handle impending overdue payments after receiving the alarms. Commercial users can customize their configuration based on their individual circumstances, so that they can use customized alarm rules and policies to determine overdue payment risks and issue alarms, thereby reliably avoiding overdue payment risks. For users who do not customize their configuration, the web system will use the default method to push alarm messages. Of course, the web system can change the default alarm message push method based on actual circumstances or the effectiveness of the default alarm message push method.

[0046] When the judgment result in step S21 is yes, that is, the Web system shows that the user has performed personalized configuration, the method enters step S23, where the Web system will judge the risk of arrears through the personalized configured alarm rules and policies; and when the judgment structure in step S21 is no, that is, the Web system shows that the user has not performed personalized configuration, the method enters step S22, where the Web system will judge the risk of arrears through the system default method.

[0047] Regardless of whether personalized configuration has been performed before, in step S24, it is determined whether there is a risk of arrears. If the judgment result is no, the method proceeds to step S20, continuously or regularly obtains the energy consumption amount of the meter, and uses the newly obtained data to update the historical energy consumption data database, so that the historical energy consumption data is more complete and the future risk prediction is more accurate; if the judgment result in step S24 is yes, the method proceeds to step S26 or step S27, and in step S26, the alarm message is pushed in the system default manner, and in step S27, the alarm message is pushed according to the user's personalized alarm configuration. For example, the system default method of pushing alarm messages is to push alarm messages through WeChat, park announcements or WeChat public accounts. The present invention does not limit this, as long as there is a high probability that the alarm message can be reliably pushed to the user; the personalized alarm configuration method can, for example, push the alarm message by email or phone to ensure that the user can ensure the reliable push of the alarm message in the way he likes.

[0048] In step S28, the energy meter alarm message is recorded. For example, the energy meter alarm message can be stored in a database for subsequent use and reference. In some embodiments, the alarm messages can be recorded in a table, listing, for each user, the times when each energy meter (electricity meter, water meter, gas meter) was determined to be at risk of arrears and the alarm message was sent.

[0049] Figure 3 FIG. 1 shows a flow chart of an energy payment alarm method based on an energy consumption prediction algorithm according to an embodiment of the present disclosure. Figure 3 As shown, the energy payment warning method of this embodiment includes the following steps S30 to S38.

[0050] In step S30, on the one hand, based on the metadata of the Internet of Things (IoT) platform, the scheduled task of the Web system is used to obtain historical data, and in step S31, the historical data is stored in the database; on the other hand, based on the metadata of the Internet of Things (IoT) platform, the scheduled task of the Web system is used to obtain the current meter balance, and in step S33, the current meter balance is stored in the database.

[0051] In one example, for example, one day can be used as the time unit in the timed task to obtain the energy consumption amount of the meter as historical energy consumption data and store it in the database, so that in step S32, these historical energy consumption data will be used to train the energy consumption prediction model used for arrears risk judgment to obtain a more reliable prediction model. The daily historical energy consumption data can be stored in the database in the form of a list. The list can be stored in the database, for example, named energy_consumption_record to facilitate subsequent access. The main fields of the list include: device ID, device type (water meter, electricity meter, gas meter), device owner email address, device owner mobile phone number, time (day), measurement unit, stage (day) energy consumption amount (or energy consumption and unit price). In one example, the energy consumption data of the previous day is obtained in the early morning of each day and stored in the database as historical data, and the above-mentioned daily historical energy consumption data list is updated. The regular update of the historical data list is used to increase the accuracy of the prediction data. The data in this list is used to predict the daily energy consumption of a specific type of energy meter as daily historical energy consumption data. For example, a range of daily historical consumption data for each energy meter for each user is used. Based on this range of daily historical energy consumption data, it is determined whether the stored value balance in the energy meter is sufficient to cover the risk of overdue payments within the next week (or day). In other words, the energy consumption is predicted based on the stored value balance on the energy meter to determine whether an alarm should be issued.

[0052] In one example, in the energy consumption monitoring and early warning of each energy meter of each user in a park, the energy consumption forecast alarm function defaults to the alarm condition based on the meter alarm forecast amount obtained by the energy consumption forecast algorithm without self-configuration through the energy consumption alarm configuration page of the web system. The alarm method is to obtain the current balance of each energy meter through the back-end program on a regular basis (such as at a fixed time every day) and save the data to the database. For example, it can be stored in a table in the database, for example, it can be named energy_consumption_record, and the main fields include: device ID, device type (water, electricity, gas), device owner's email address, device owner's mobile phone number, time (day), measurement unit, device current balance, and device predicted consumption amount for the day.

[0053] Then, in step S34, the current meter balance obtained is compared with the daily historical energy consumption data obtained by the energy consumption prediction algorithm in step S32 to determine whether there is a risk of arrears; if not, the process ends at step S38; if there is a risk of arrears, a reminder is given at step S35 and a judgment is made whether to repeat the reminder (wherein, if the risk of arrears occurs on the same day, it is necessary to judge whether to repeat the reminder on the same day), if not, an alarm message is pushed at step S36; if yes, the process ends after the reminder, and returns to step S30 after a predetermined time, and re-judges whether the user has recharged after being reminded before to re-judge whether there is a risk of arrears; in step 37, the alarm message pushed in step S36 is stored in the energy warning message record table for subsequent use or reference.

[0054] In one example, after a warning message is pushed, it is stored in the database. The table involved, energy_consumption_message, has the following main fields: device ID, device type (water, electricity, gas), device owner, message type (WeChat official account, SMS, email), time (day), current device balance, predicted device consumption amount for the day, push time, and push status (success, failure).

[0055] In one embodiment, the data preparation step may include the following sub-steps: obtaining the energy consumption amount of a single meter, starting a scheduled task every morning to obtain the meter consumption amount of the previous day; because the energy meter can only obtain the real-time balance, the program needs to obtain and calculate the consumption amount of the previous day through two scheduled tasks; accessing the IoT platform through the back-end interface to obtain the current balance of the energy device, and obtaining the total amount of the payment bill of the current meter on the previous day through the IoT platform, and obtaining the energy consumption amount of the meter on the previous day by calculating under a single meter (current balance - total amount of the payment bill of the previous day - the last scheduled task call to obtain the meter balance); storing the obtained meter energy consumption amount and energy device related information in the database list device_energy_consumption_record as preliminary data preparation for the energy consumption prediction algorithm.

[0056] The device_energy_consumption_record list may include the following fields: device ID, event (event_date), meter energy consumption daily start amount (start_day_balance), meter energy consumption daily end amount (end_day_balance), current day energy consumption (day_consumption), update time (update_time), etc.

[0057] In one embodiment, the data preprocessing step may include: obtaining two columns of data, namely, energy consumption time series (event_date) and energy consumption amount (day_consumption), based on the stored energy meter historical data (device_energy_consumption_record table); filtering out invalid data in the above data, and then normalizing and serializing the data to finally obtain a time series data set as input for the LSTM model.

[0058] In one embodiment, the daily historical consumption data of each energy appliance of each user may be obtained by the following energy consumption prediction algorithm, which may represent the average value of the daily consumption data of each energy appliance of each user.

[0059] In one embodiment, the energy consumption prediction algorithm may include data preparation steps to collect historical energy consumption data; data preprocessing; building an energy consumption prediction model based on the LSTM algorithm; training the model; output and verification; and predicting tomorrow's consumption.

[0060] In one embodiment, Figure 4As shown in FIG, the energy consumption prediction model based on the LSTM algorithm is mainly designed to include an input layer, a first LSTM layer, a first Dropout layer, a second LSTM layer, a second Dropout layer, a fully connected Dense layer and an output layer.

[0061] The first LSTM layer has a hidden state dimension of 64 and contains 64 units (LSTM (64 units), which is used to learn high-dimensional features of the sequence. The second LSTM layer contains 32 neural units, reducing the dimensionality to extract higher-level abstract features. The first and second Dropout layers randomly discard 20% of the neuron outputs (effective only during training) to prevent overfitting and enhance generalization. The fully connected Dense layer contains 16 units to further reduce the dimensionality and map the LSTM output to a 16-dimensional space to further extract nonlinear features. The ReLU activation function is introduced in the fully connected Dense layer for nonlinearization to enhance the model's expressiveness. The output layer is used for regression task output and predicts the model's energy consumption.

[0062] In constructing Figure 4 After the prediction model architecture is shown, the model needs to be trained. Figure 5 A flowchart of a method for training a prediction model according to an embodiment of the present disclosure is shown. The method for training a prediction model according to an embodiment of the present disclosure may include steps S51 to S53.

[0063] In step S51: first, the acquired pre-processed data is divided into a training set and a validation set;

[0064] In step S52: Then, the input dimensions are adjusted according to the model design, which mainly include the number of samples, time steps and number of features; and

[0065] In step S53: the model training parameters are adjusted, the model is trained, and finally a model with the lowest prediction error rate is saved as the energy consumption prediction algorithm model.

[0066] In step S53, the parameters to be adjusted primarily include: time_steps (time window, number of days of historical data), number of LSTM layers and units, dropout ratio, number of epochs for model training, and batch_size. Experimentation has shown that actual prediction performance depends on the quality and quantity of data, with the best results achieved with more than two years of data points (700+).

[0067] In accordance with Figure 5 After training the model, you can use the following method steps: Figure 6 The steps shown are to output and verify the trained model. Specifically, Figure 6A flowchart of a method for verifying a trained prediction model according to an embodiment of the present disclosure is shown. Figure 6 The method for verifying the trained prediction model shown includes the following steps S61 to S64:

[0068] In step S61: using the trained model to predict the training set and the validation set to obtain the predicted value;

[0069] In step S62: the predicted value and the true value are restored from the normalized range to the original data range;

[0070] In step S63: calculate and output the mean absolute error (MAE) of the training set and the validation set to evaluate the model performance;

[0071] In step S64: adjust the model parameters according to the evaluation results to further improve the prediction accuracy, that is, re-perform the Figure 5 The method steps for training the prediction model are shown to obtain a further trained model for subsequent prediction, for example, using the trained model to predict the amount of energy consumption tomorrow.

[0072] Figure 7 FIG. 1 is a flow chart showing a method for energy payment warning according to a user's personalized configuration according to an embodiment of the present disclosure. Figure 7 As shown, the energy payment warning method of this embodiment includes the following steps S70 to S79.

[0073] In step S70, metadata of the Internet of Things (IoT) platform is obtained. As mentioned above, the metadata obtained by the scheduled task of the Web system includes the current meter balance and daily energy consumption;

[0074] In step S71, based on the metadata of the Internet of Things (IoT) platform, the current meter balance and daily energy consumption data are obtained by using the scheduled task of the Web system and stored in the database to form a list of relevant information of the user's energy consumption meter;

[0075] In step S72, the energy consumption alarm is configured, i.e., the alarm rules and alarm strategies for the energy consumption alarm are personalized. Here, the alarm conditions, the alarm channels, the number of alarms, and the alarm frequency can be configured. This disclosure does not limit this, and any configuration desired by the user can be used as the personalized configuration of the energy consumption alarm rules and alarm strategies.

[0076] In step S73, the energy consumption configuration data performed by the user is stored in the database to form an energy consumption alarm configuration table. In one example, each user in a park, such as a merchant, can independently configure alarm rules and alarm strategies through the energy consumption alarm configuration page of the Web system. Figure 8 As shown, the Web system energy consumption alarm configuration page in one embodiment of the present disclosure may include the following options: warning name (such as overdue payment reminder, tomorrow's overdue payment reminder, one-week overdue payment reminder, etc.), adding application devices (such as water meters, electricity meters, natural gas meters, etc.), device start query time, device balance query interval, whether to repeat reminders, message push channels, notes, etc. Figure 8 Just one example is shown. The energy consumption alarm configuration page of the Web system can also include multiple channel push, single channel push, push interval, push times, etc. After the merchant has configured the alarm configuration independently, it can alarm according to the personalized alarm rules and strategies instead of relying on the energy consumption prediction algorithm. The database involved includes the merchant's personalized alarm configuration table (such as energy_consumption_configuration), which mainly includes the fields device ID, device type (water, electricity, gas), device owner's email address, device owner's mobile phone number, interval time (hours), alarm strategy (less than, less than or equal to), alarm amount, alarm repetition times (on the same day), alarm channel, and alarm name. However, the present disclosure is not limited to this. In the alarm rules configured by the user independently, the user can also choose to rely on the energy consumption prediction algorithm to alarm.

[0077] In step S74, the risk of arrears is judged based on the energy consumption alarm configuration table. After the user has made personalized configuration requirements, the energy payment warning method of the present invention will judge the risk of arrears based on the personalized configuration requirements (which may include personalized configuration of energy consumption alarm alarm rules and alarm strategies). In some embodiments, for example, the current meter balance obtained and the user's personalized configuration requirements in the energy consumption alarm configuration table can be compared to determine whether there is a risk of arrears; if there is no risk of arrears, the process ends at step S79; if there is a risk of arrears, a reminder is made at step S75 and a judgment is made on whether to repeat the reminder (that is, whether it meets the repetitive requirements set by the user). If the judgment result shows that repeated reminders are not required, the method proceeds to step S79 to end the processing; if the judgment result shows that repeated reminders are required, a push channel is selected in step S76 according to the push channel configured in the energy consumption alarm configuration table stored in the memory set by the user; in step 77, the alarm message is pushed using the push channel selected in step S76, and in step S78, the alarm message is stored in the energy warning message record table for subsequent use or reference.

[0078] exist Figure 7 In the illustrated embodiment, the energy payment alarm method disclosed herein supports user-customized configuration of alarm rules and alarm strategies, and pushes alarm messages by determining energy consumption alarm conditions set independently by users.

[0079] In this embodiment, in step S74, the risk of arrears can be determined based on the user expectations configured in the energy consumption alarm configuration table. In fact, this determination step can also be adopted Figure 3 The embodiment described herein uses an energy consumption prediction algorithm to determine the risk of arrears. Of course, the present invention is not limited to this. Risk determination in step S74 can be tailored to the user's needs. For example, the user can configure a default risk to occur when their top-up balance is less than a predetermined value, rather than determining the risk based on daily energy consumption. This ensures that users can reliably avoid losses caused by energy outages based on their individual circumstances.

[0080] As described above, in some embodiments, for example, the energy consumption alarm implementation method of each energy appliance of each user in the park can include two implementation methods as described above: one is to use the meter alarm prediction amount obtained by the energy consumption prediction algorithm as a condition for alarming. For example, if it is found after monitoring that there will be insufficient balance in the next week, the alarm will be issued according to the system default method; the second is that if the user of the park, such as the merchant, has independently configured the alarm rules and alarm strategies through the web system energy consumption alarm configuration page, the alarm will be pushed according to the rules configured by the user.

[0081] The above two alarm implementation methods will traverse the energy consumption meter balances of all users in the park, and make a risk assessment of arrears, and then push alarms based on whether the users involved have made personalized customizations.

[0082] Furthermore, the risk of overdue payments is judged: by combining the data table (device_energy_consumption) obtained from the energy consumption forecast or the merchant's personalized alarm configuration table (energy_consumption_config table), it is determined whether the energy meter will have a risk of overdue payments tomorrow; if there is a risk of overdue payments, the alarm mechanism is triggered and the merchant meter pre-arrears message is pushed to the user through different message channels.

[0083] That is, if the arrears risk judgment conditions are met, an alarm message will be pushed. For the alarm method that has been personalized configured by the merchant, the alarm push channel can be selected independently, including the WeChat public account bound to the merchant or the email address or phone number text message uploaded by the user; for the alarm method that has not been personalized configured by the merchant, the default is to use the meter alarm prediction amount obtained by the energy consumption prediction algorithm model as the condition for alarm, and the alarm push channel defaults to pushing messages through the WeChat public account bound to the user; the message content includes the arrears warning information of the energy meter and provides a recharge link for the warning meter. Users can directly go to the recharge interface of the current meter through the link to improve user payment efficiency and experience.

[0084] During this process, repeated reminders should be avoided: users are allowed to set the total number of times alarm messages are pushed on the same day; before each overdue payment warning message is pushed, the real-time balance under the meter is obtained and it is verified whether the push conditions are met to ensure that reminder messages are no longer pushed after recharging.

[0085] Based on the above, the energy consumption monitoring method of the energy meter disclosed in the present invention may also include, before determining whether there is a risk of arrears and pushing an alarm message: determining whether the current user has personalized energy consumption data; if so, further determining whether the current user's current energy consumption user has personalized energy consumption data; if so, pushing an alarm message according to the alarm rules and alarm strategies in the personalized energy consumption data; if not, pushing the alarm message in the system default manner.

[0086] According to the energy consumption monitoring method of energy appliances according to the embodiment of the present disclosure, personalized energy consumption alarm rules and alarm strategies can be performed for each user and for each energy consumption appliance of each user. For example, the reminder and alarm methods are different between different users, and the reminder and alarm methods are different between different energy appliances of a user. This can bring convenience to users and provide more effective reminders.

[0087] Figure 9 A block diagram of an electronic device provided in an embodiment of the present disclosure.

[0088] Reference Figure 9 An embodiment of the present disclosure provides an electronic device, which includes: at least one processor 901; at least one memory 902, and one or more I / O interfaces 903, connected between the processor 901 and the memory 902; wherein the memory 902 stores one or more computer programs that can be executed by the at least one processor 901, and the one or more computer programs are executed by the at least one processor 901 to enable the at least one processor 901 to execute the above-mentioned energy payment warning method.

[0089] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor / processing core, implements the aforementioned energy payment warning method. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.

[0090] It will be understood by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable storage medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium).

[0091] As is well known to those skilled in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information (such as computer-readable program instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technology, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those skilled in the art, communication media typically contains computer-readable program instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0092] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0093] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0094] The computer program product described herein may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0095] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0096] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0097] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0098] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the prescribed logical function. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the prescribed function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0099] The embodiments of the present disclosure achieve the following effects: intelligent early warning: using the energy efficiency future consumption prediction algorithm to predict energy consumption values ​​in advance and realize intelligent early warning of arrears risks; convenient payment: providing a recharge link through the pushed alarm message, so that merchants can pay directly and improve payment efficiency and experience; avoiding repeated reminders: verifying the real-time balance before each alarm message is pushed to ensure that no more reminder messages are pushed after recharge, so as to avoid unnecessary disturbance to merchants; energy consumption prediction algorithm based on LSTM neural network, the features include data preprocessing, model training and parameter adjustment; personalized alarm settings: including different merchants independently setting alarm conditions and alarm push strategies (whether to push repeatedly, push interval time, message push channel).

[0100] The exemplary embodiments of the present invention have been described in detail above for the purposes of illustration and description. This is not intended to be exhaustive or to limit the invention to the precise forms of the embodiments disclosed. Therefore, the above description should be considered illustrative rather than restrictive. It should be understood that changes and modifications may be made to the described embodiments by those skilled in the art without departing from the scope of the invention as defined by the appended claims, and such changes and modifications should fall within the scope of protection of the present invention.

Claims

1. A computer-implemented energy payment early warning method, comprising steps S11 to S14: Step S11: obtaining the current remaining recharge balance of the user's energy meter; Step S12: obtaining daily energy consumption values ​​related to the energy meter; Step S13: Determine whether the recharge balance has a risk of arrears; if so, proceed to step S14: Push an alarm message to the user through at least one channel; if not, return to step S11 after a first predetermined time.

2. The energy payment early warning method according to claim 1, wherein: Obtaining the recharge balance of the energy meter includes: periodically obtaining the recharge balance of the energy meter.

3. The energy payment early warning method according to claim 2, wherein: Regularly obtaining the recharge balance of the energy meter includes: obtaining the recharge balance of the energy meter at a predetermined time every day.

4. The energy payment early warning method according to claim 1, wherein: The at least one channel includes at least one of a WeChat public account, text messages, and emails.

5. The energy payment early warning method according to claim 1, wherein: Acquiring the daily energy consumption value related to the energy meter includes: using an energy consumption prediction algorithm to predict the energy consumption value for the next day.

6. The energy payment early warning method according to claim 5, wherein: The step of using the energy consumption prediction algorithm to predict the energy consumption value for the next day also includes: designing the energy consumption prediction algorithm, and Designing the energy consumption prediction algorithm includes: Collect historical energy consumption data and store it in memory; Preprocessing the historical energy consumption data stored in the memory to obtain preprocessed data including at least an energy consumption time series and a corresponding energy consumption amount; Construct an initial energy consumption prediction model based on the LSTM algorithm; Dividing the preprocessed data into a training set and a validation set and inputting the resultant data into the initial energy consumption prediction model based on the LSTM algorithm to train the initial energy consumption prediction model based on the LSTM algorithm; Using the trained model to predict the training set and the validation set to evaluate model performance, and adjusting the parameters of the energy consumption prediction model based on the LSTM algorithm according to the evaluation results to obtain a model with adjusted parameters; and The model with adjusted parameters is used to predict the energy consumption value for the next day.

7. The energy payment early warning method according to claim 1, wherein: Pushing the alarm message to the user through at least one channel also includes: providing the user with a recharge link while pushing the alarm message, so that the user can go to a recharge interface through the recharge link.

8. The energy payment early warning method according to claim 1, wherein: Determining whether the recharge balance has a risk of arrears includes: Determining whether the user has ever independently configured alarm rules and alarm policies through a configuration page related to energy consumption alarms on the Internet; If not, determining whether the energy meter has a risk of arrears based on the difference between the recharge amount of the user's energy meter and the determined daily energy consumption value; If yes, it is determined whether the energy meter has a risk of arrears based on the personalized alarm configuration table configured by the user on the configuration page.

9. The energy payment early warning method according to claim 1, wherein: Pushing an alarm message to a user through at least one channel includes: Determining whether the user has ever independently configured alarm rules and alarm policies through a configuration page related to energy consumption alarms on the Internet; If not, push the alarm message through the default channel; If yes, then according to the personalized alarm configuration table configured by the user on the configuration page, an alarm message is pushed through the channel configured by the user.

10. The energy payment early warning method according to claim 1, wherein: After pushing the alarm message to the user through at least one channel, the energy payment warning method further includes: Determining whether the user has ever independently configured alarm rules and alarm policies through a configuration page related to energy consumption alarms on the Internet; If not, after a second predetermined time, the process returns to step S11 to determine the recharge balance of the user's energy meter, so as to determine whether the user has recharged after the alarm message is pushed to the user, and if the judgment result shows that the user has not recharged or there is still a risk of arrears after recharging, the alarm message is pushed to the user again; If yes, then it is determined whether to judge whether the user responds to the alarm message after the alarm message is pushed to the user according to the personalized alarm configuration table configured by the user on the configuration webpage.

11. The energy payment early warning method according to claim 10, wherein: The first predetermined time is greater than the second predetermined time.

12. An electronic device comprising a processor, a memory, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the energy payment early warning method according to any one of claims 1 to 11.

13. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

Images