A household energy consumption management method and device combined with a multifunctional electric energy meter

Abstract

The present invention relates to the field of household energy consumption management technology, and specifically to a household energy consumption management method and device combined with a multifunctional electric energy meter, the method comprising: using a multifunctional electric energy meter to collect energy consumption data of household electrical appliances in real time, the collected data has a timestamp, and constitutes an energy consumption data set arranged in time series; performing data preprocessing on the collected energy consumption data set, including removing outliers, data standardization, and data cleaning; extracting eigenvalues ​​from the energy consumption data set after preprocessing; training a household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted eigenvalues; predicting the energy consumption of household electrical appliances through the trained model and managing the energy consumption according to the prediction results. The present invention combines a multifunctional electric energy meter with machine learning technology to collect energy consumption data of household electrical appliances in real time and perform intelligent analysis and prediction, timely discover anomalies and manage them, save energy and reduce household energy consumption costs.

Description

Technical Field

[0001] The present invention relates to the technical field of household energy consumption management, and in particular to a household energy consumption management method and device combined with a multifunctional electric energy meter. Background Art

[0002] With the development of society and the improvement of people's living standards, the number of electrical appliances in households is increasing, and the problem of household energy consumption is becoming increasingly prominent. Traditional household energy consumption management methods are mostly based on simple meter readings and manual estimates. They lack accurate analysis and effective management of the energy consumption of household appliances, and cannot make early judgments on energy consumption anomalies. On the one hand, it is difficult for users to accurately know the energy consumption of household appliances and cannot take targeted energy-saving measures; on the other hand, it is impossible to effectively predict household energy consumption patterns, and cannot timely discover and adjust energy consumption anomalies before they occur. In addition, some existing energy consumption management devices have single functions and cannot meet the energy consumption management needs in complex household environments. Therefore, there is an urgent need for an intelligent household energy consumption management method and device. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the present invention provides a household energy consumption management method and device combined with a multifunctional electric energy meter.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for managing household energy consumption in combination with a multifunctional electric energy meter is provided, the method comprising the following steps:

[0006] Step S10: using a multifunctional electric energy meter to collect energy consumption data of household electrical appliances in the home in real time, including voltage, current, power and power consumption duration, the collected data is time-stamped, and constitutes an energy consumption data set arranged in time series;

[0007] Step S20: performing data preprocessing on the collected energy consumption data set, including removing outliers, data standardization and data cleaning;

[0008] Step S30: extracting characteristic values ​​from the energy consumption data set after preprocessing, including the daily average energy consumption, energy consumption peak, energy consumption peak occurrence time and energy consumption fluctuation frequency of the household electrical appliances. The extracted characteristic values ​​are used to reflect the energy consumption mode and usage regularity of the household electrical appliances.

[0009] Step S40: constructing a household energy consumption prediction model based on LSTM, and training the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted feature values;

[0010] Step S50: predicting the energy consumption of household electrical appliances through the trained model, managing household energy consumption according to the prediction results, and judging whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, an energy consumption abnormality warning is issued to the user, and the potential cause of the abnormality is pointed out according to the abnormal content. For example, excessive energy consumption indicates that one or some household electrical appliances are faulty or aging. When there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household electrical appliances according to the household energy consumption information. At the same time, the household energy consumption prediction model re-outputs the prediction results according to the user's adjustment, and the cycle repeats.

[0011] The multifunctional electric energy meter used in step S10 has the following functions:

[0012] Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement;

[0013] Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement;

[0014] Data storage: Store the collected data according to the measurement date and time;

[0015] In the step S40 of constructing a household energy consumption prediction model based on LSTM, an improved LSTM long short-term memory algorithm is adopted. By changing the number of hidden layers and fully connected layers, a Dropout layer is added between the LSTM layers to optimize the training of the model, making it more suitable for learning and predicting the energy consumption data of household electrical appliances in the home.

[0016] Preferably, the data collected in step S10 carries a timestamp, constituting an energy consumption data set arranged in a time series, and is implemented by setting a real-time clock unit in the multi-function electric energy meter, and regularly synchronizing the time through the network to provide accurate time information; after the energy consumption data is collected, the current time provided by the real-time clock unit is used as a timestamp, and is stored together with the collected energy consumption data to form an energy consumption data set arranged in a time series.

[0017] Preferably, the step S20 performs data preprocessing on the collected energy consumption data set, including:

[0018] Removing outliers: The threshold method is used to set the threshold range to identify and remove outliers. The normal power range of household electrical appliances is P min To P max , then the threshold is set as shown in formula (1):

[0019] P 阈值 =k×(P max -P min )+P min (1)

[0020] Where k is a coefficient, which is set according to the specific household electrical appliance and has a value range of 1.05≤k≤1.95. When the collected power value exceeds the set threshold, it is determined as an abnormal value and removed;

[0021] Data standardization: Z-score standardization is used to convert the collected data into a standard normal distribution with a mean of 0 and a standard deviation of 1. The calculation formula is shown in formula (2):

[0022]

[0023] Where x is a data point in the data set, μ is the mean of the data set, σ is the standard deviation of the data set, and x is the standardized x;

[0024] Data cleaning: missing values ​​were filled using the mean method.

[0025] Preferably, extracting characteristic values ​​from the energy consumption data set after preprocessing in step S30 specifically includes:

[0026] Daily average energy consumption: Divide the total daily energy consumption by the number of statistical days to obtain the daily average energy consumption of the household appliance. The calculation formula is shown in formula (3):

[0027]

[0028] Where E is the average daily energy consumption, n is the number of days, E j is the energy consumption on the jth day;

[0029] Energy consumption peak: The peak value of energy consumption is determined by traversing the energy consumption data set. The energy consumption data set sequence is P = [P 1 ,P 2 ,...,P m ], set P max =P 1 , then for i = 2 to i = m, when P i >P max When P max =P 1 , the final P obtained after traversing the energy consumption data set max It is the peak energy consumption;

[0030] Energy consumption peak occurrence time: The data collected in step S10 is time-stamped, forming an energy consumption data set arranged in time series. When the energy consumption peak data is obtained, the peak data is time-stamped, and the energy consumption peak occurrence time is obtained;

[0031] Energy consumption fluctuation frequency: The energy consumption fluctuation frequency is calculated by analyzing the change of energy consumption data over time. The calculation formula is shown in formula (4):

[0032]

[0033] Where f is the energy consumption fluctuation frequency, a is the number of data points, Δt is the time interval, P b is the energy consumption value at the bth time point, P b+1 is the energy consumption value at the b+1th time point.

[0034] Preferably, in step S40, a household energy consumption prediction model based on LSTM is constructed, and the household energy consumption prediction model based on LSTM is trained according to the energy consumption data set and the extracted eigenvalues. The model construction and training process includes:

[0035] Dataset division: The energy consumption dataset is divided into a training set and a validation set, with the training set accounting for 80% and the validation set accounting for 20%;

[0036] Model construction: Adopt LSTM-based long-term and short-term neural network architecture, including input layer, hidden layer, and fully connected layer; the input layer contains 5 nodes, corresponding to the energy consumption data set and the four extracted feature values ​​of daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency; use LSTM layer as hidden layer, and add Dropout layer between each LSTM layer. Each LSTM layer contains 64 LSTM units, and each LSTM unit has input gate, forget gate and output gate for processing sequence data; the fully connected layer is used as the output layer, which contains 5 nodes, corresponding to the 5 nodes of the input layer, and is used to map the output of the hidden layer to the final output. The output value is the predicted energy consumption of household appliances and daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency;

[0037] Model training: The mean square error loss function is used, and the formula is shown in formula (5):

[0038]

[0039] Where a is the number of data points, y i is the i-th true value, is the i-th predicted value, the true value and pre-test include energy consumption data, average daily energy consumption, energy consumption peak, time of occurrence of energy consumption peak and frequency of energy consumption fluctuation; Adam optimizer is used as the optimization algorithm to update the weights in model training to minimize the loss function; for each data in the training set, it is input into the LSTM-based household energy consumption prediction model in batches, and the output is calculated by forward propagation; the weight is calculated through the loss function according to the output and the true value; the gradient of the loss function for each weight is calculated through back propagation according to the weight; the weight is updated through the Adam optimizer according to the gradient; the above steps are repeated to traverse all batches of data to complete a training round; after each training round, the performance of the model is evaluated using the validation set, and the model training is completed when the loss no longer decreases.

[0040] Preferably, the step S50 manages the household energy consumption according to the prediction results and determines whether the household energy consumption is abnormal according to the set energy consumption threshold. The threshold setting method is consistent with the threshold setting method in the above-mentioned step of removing abnormal values. When the result predicted by the household energy consumption prediction model exceeds the threshold, an energy consumption abnormality warning is issued to the user by means of an alarm, text message and flashing light.

[0041] Preferably, in step S50, the potential cause of the abnormality is pointed out according to the abnormality type by obtaining data on the energy consumption abnormality type and the corresponding cause of the abnormality from the Internet, and at the same time serving as the input of the household energy consumption prediction model. When an abnormality occurs, an energy consumption abnormality warning is issued to the user, and a matching abnormality type and potential cause of the abnormality are given.

[0042] In addition, to achieve the above-mentioned purpose, the present invention also proposes a household energy consumption management device combined with a multifunctional electric energy meter, comprising:

[0043] Multifunctional energy meter data collection module: used to use the multifunctional energy meter to collect energy consumption data of household electrical appliances in real time, including voltage, current, power and power consumption duration. The collected data is time-stamped and forms an energy consumption data set arranged in time series;

[0044] Energy consumption data preprocessing module: used to preprocess the collected energy consumption data set, including removing outliers, data standardization and data cleaning;

[0045] Energy consumption data feature extraction module: used to extract feature values ​​from the pre-processed energy consumption data set, including the average daily energy consumption, peak energy consumption, peak energy consumption time and energy consumption fluctuation frequency of household electrical appliances. The extracted feature values ​​are used to reflect the energy consumption mode and usage rules of household electrical appliances.

[0046] Household energy consumption prediction model construction and training module: used to build a household energy consumption prediction model based on LSTM, and train the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted feature values;

[0047] Household energy consumption management module: used to predict the energy consumption of household appliances through the trained model, manage household energy consumption according to the prediction results, and judge whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, it will issue an energy consumption abnormality warning to the user and point out the potential cause of the abnormality according to the abnormal content. For example, high energy consumption indicates that one or some household appliances are faulty or aging. When there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household appliances according to the household energy consumption information. At the same time, the household energy consumption prediction model will re-output the prediction results according to the user's adjustment, and the cycle will be repeated.

[0048] The functions of the multifunctional electric energy meter used in the multifunctional electric energy meter data acquisition module include:

[0049] Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement;

[0050] Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement;

[0051] Data storage: Store the collected data according to the measurement date and time;

[0052] In the household energy consumption prediction model construction and training module, an improved LSTM long short-term memory algorithm is adopted. By changing the number of hidden layers and fully connected layers, a Dropout layer is added between the LSTM layers to optimize the training of the model, making it more suitable for learning and predicting the energy consumption data of household electrical appliances in the home.

[0053] In addition, to achieve the above-mentioned objectives, the present invention also proposes a household energy consumption management device combined with a multi-function electric energy meter, the device comprising: a memory, a processor, and programs such as LSTM-based household energy consumption prediction stored on the memory and run on the processor, the programs such as LSTM-based household energy consumption prediction are steps to implement a household energy consumption management method combined with a multi-function electric energy meter as described above.

[0054] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer program product, which includes programs such as LSTM-based household energy consumption prediction. When the LSTM-based household energy consumption prediction programs are executed by the processor, a household energy consumption management method combined with a multi-function electricity meter as described above is implemented.

[0055] The advantages and effects of the present invention are:

[0056] The present invention proposes a household energy consumption management method and device combined with a multifunctional electric energy meter. The method uses a multifunctional electric energy meter combined with a neural network algorithm to accurately collect and predict the energy consumption data of household electrical appliances in the home, providing technical support for accurate energy consumption analysis, and can timely discover abnormal energy consumption, effectively helping users save energy and reduce household energy consumption costs; in addition, the various modules of the proposed household energy consumption management device combined with a multifunctional electric energy meter have clear division of labor and work in coordination, which can provide users with comprehensive and convenient energy consumption management services and improve users' ability to control household energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 The present invention is a flow chart of a household energy consumption management method combined with a multifunctional electric energy meter.

[0059] Figure 2 The present invention is a schematic structural diagram of a household energy consumption management device combined with a multifunctional electric energy meter.

[0060] Figure 3 The present invention is a schematic block diagram of the structure of a household energy consumption management electronic device combined with a multifunctional electric energy meter. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] The present invention provides a household energy consumption management method combined with a multifunctional electric energy meter, such as Figure 1 As shown, the following steps are included:

[0063] Step S10: The energy consumption data of each electrical appliance in the home is collected in real time through a multifunctional electricity meter, including parameters such as voltage, current, power and power consumption duration. The multifunctional electricity meter has a high-precision measurement function and can accurately obtain the power information on the branch circuits of different electrical appliances. The collected data is timestamped to form an energy consumption data set arranged in time series.

[0064] Among them, the functions of the multifunctional energy meter used include:

[0065] Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement; active electric energy measurement is used to measure the actual electric energy consumed by household electrical appliances, and the unit is kilowatt-hour (kW / h); reactive electric energy measurement is used to measure the electric energy of reactive power in household electrical appliances, and the unit is kilovar-hour (kVar / h); total electric energy measurement is used to measure the total electric energy in the circuit, and the unit is kilovolt-ampere-hour (kVA / h);

[0066] Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement;

[0067] Data storage: Store the collected data according to the measurement date and time.

[0068] Specifically, the data collected in step S10 has a timestamp and constitutes an energy consumption data set arranged in time series, which is implemented as follows:

[0069] A real-time clock unit is set in the multifunctional electric energy meter to synchronize the time regularly through the network to provide accurate time information;

[0070] After the energy consumption data is collected, the current time provided by the real-time clock unit is used as a timestamp and stored together with the collected energy consumption data to form an energy consumption data set arranged in time series.

[0071] Step S20: preprocessing the collected energy consumption data set, including removing outliers, such as abnormally high energy consumption data points caused by short-term circuit failures; data standardization to make data of different magnitudes comparable; data cleaning to fill in missing values, etc.

[0072] Specifically, data preprocessing of the collected energy consumption data set includes:

[0073] Removing outliers: The threshold method is used to set the threshold range to identify and remove outliers. The normal power range of household electrical appliances is P min To P max , then the threshold is set as shown in formula (1):

[0074] P 阈值 =k×(P max -P min )+P min (1)

[0075] Where k is a coefficient, which is set according to the specific household electrical appliance and has a value range of 1.05≤k≤1.95. When the collected power value exceeds the set threshold, it is determined as an abnormal value and removed;

[0076] Data standardization: Z-score standardization is used to convert the collected data into a standard normal distribution with a mean of 0 and a standard deviation of 1. The calculation formula is shown in formula (2):

[0077]

[0078] Where x is a data point in the data set, μ is the mean of the data set, σ is the standard deviation of the data set, and x is the standardized x;

[0079] Data cleaning: missing values ​​were filled using the mean method.

[0080] Step S30: extracting characteristic values ​​from the energy consumption data set after preprocessing, including the average daily energy consumption, energy consumption peak, energy consumption peak occurrence time and energy consumption fluctuation frequency of household electrical appliances. The extracted characteristic values ​​are used to reflect the energy consumption mode and usage rules of household electrical appliances.

[0081] Among them, extracting characteristic values ​​from the energy consumption data set after preprocessing includes:

[0082] Daily average energy consumption: Divide the total daily energy consumption by the number of statistical days to obtain the daily average energy consumption of the household appliance. The calculation formula is shown in formula (3):

[0083]

[0084] Where E is the average daily energy consumption, n is the number of days, E j is the energy consumption on the jth day;

[0085] Energy consumption peak: The peak value of energy consumption is determined by traversing the energy consumption data set. The energy consumption data set sequence is P = [P 1 ,P 2 ,...,P m ], set P max =P 1 , then for i = 2 to i = m, when P i >P max When P max =P 1 , the final P obtained after traversing the energy consumption data set max It is the peak energy consumption;

[0086] Energy consumption peak occurrence time: The data collected in step S10 is time-stamped, forming an energy consumption data set arranged in time series. When the energy consumption peak data is obtained, the peak data is time-stamped, and the energy consumption peak occurrence time is obtained;

[0087] Energy consumption fluctuation frequency: The energy consumption fluctuation frequency is calculated by analyzing the change of energy consumption data over time. The calculation formula is shown in formula (4):

[0088]

[0089] Where f is the energy consumption fluctuation frequency, a is the number of data points, Δt is the time interval, P b is the energy consumption value at the bth time point, P b+1 is the energy consumption value at the b+1th time point.

[0090] Step S40: construct a household energy consumption prediction model based on LSTM, and train the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted eigenvalues; in the household energy consumption prediction model based on LSTM, an improved LSTM long short-term memory algorithm is adopted, and the training of the model is optimized by changing the number of hidden layers and fully connected layers and adding a Dropout layer between the LSTM layers, so that it is more suitable for the learning and prediction of energy consumption data of household electrical appliances in the home.

[0091] Specifically, the construction and training process of the LSTM-based household energy consumption prediction model includes:

[0092] Dataset division: The energy consumption dataset is divided into a training set and a validation set, with the training set accounting for 80% and the validation set accounting for 20%;

[0093] Model construction: Adopt LSTM-based long-term and short-term neural network architecture, including input layer, hidden layer, and fully connected layer; the input layer contains 5 nodes, corresponding to the energy consumption data set and the four extracted feature values ​​of daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency; use LSTM layer as hidden layer, and add Dropout layer between each LSTM layer. Each LSTM layer contains 64 LSTM units, and each LSTM unit has input gate, forget gate and output gate for processing sequence data; the fully connected layer is used as the output layer, which contains 5 nodes, corresponding to the 5 nodes of the input layer, and is used to map the output of the hidden layer to the final output. The output value is the predicted energy consumption of household appliances and daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency;

[0094] Model training: The mean square error loss function is used, and the formula is shown in formula (5):

[0095]

[0096] Where a is the number of data points, y i is the i-th true value, is the i-th predicted value, the true value and pre-test include energy consumption data, average daily energy consumption, energy consumption peak, time of occurrence of energy consumption peak and frequency of energy consumption fluctuation; Adam optimizer is used as the optimization algorithm to update the weights in model training to minimize the loss function; for each data in the training set, it is input into the LSTM-based household energy consumption prediction model in batches, and the output is calculated by forward propagation; the weight is calculated through the loss function according to the output and the true value; the gradient of the loss function for each weight is calculated through back propagation according to the weight; the weight is updated through the Adam optimizer according to the gradient; the above steps are repeated to traverse all batches of data to complete a training round; after each training round, the performance of the model is evaluated using the validation set, and the model training is completed when the loss no longer decreases.

[0097] Step S50: Predict the energy consumption of household electrical appliances through the trained model, manage the household energy consumption according to the prediction results, and determine whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, an energy consumption abnormality warning is issued to the user and the potential cause of the abnormality is pointed out according to the abnormal content. For example, excessive energy consumption indicates that one or some household electrical appliances are faulty or aging; when there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household electrical appliances according to the household energy consumption information. At the same time, the household energy consumption prediction model re-outputs the prediction results according to the user's adjustment, and the cycle repeats. The threshold setting method is consistent with the threshold setting method in the above-mentioned step of removing abnormal values. When the result predicted by the household energy consumption prediction model exceeds the threshold, an energy consumption abnormality warning is issued to the user through an alarm, text message and flashing.

[0098] Specifically, in step S50, the potential cause of the abnormality is pointed out according to the abnormality type by obtaining data on the energy consumption abnormality type and the corresponding cause of the abnormality from the Internet, and at the same time serving as the input of the household energy consumption prediction model. When an abnormality occurs, an energy consumption abnormality warning is issued to the user, and a matching abnormality type and potential cause of the abnormality are given.

[0099] In addition, the present invention also proposes a household energy consumption management device combined with a multifunctional electric energy meter, please refer to Figure 2 ,include:

[0100] Multifunctional energy meter data collection module: used to use the multifunctional energy meter to collect energy consumption data of household electrical appliances in real time, including voltage, current, power and power consumption duration. The collected data is time-stamped and forms an energy consumption data set arranged in time series;

[0101] Energy consumption data preprocessing module: used to preprocess the collected energy consumption data set, including removing outliers, data standardization and data cleaning;

[0102] Energy consumption data feature extraction module: used to extract feature values ​​from the pre-processed energy consumption data set, including the average daily energy consumption, peak energy consumption, peak energy consumption time and energy consumption fluctuation frequency of household electrical appliances. The extracted feature values ​​are used to reflect the energy consumption mode and usage rules of household electrical appliances.

[0103] Household energy consumption prediction model construction and training module: used to build a household energy consumption prediction model based on LSTM, and train the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted feature values;

[0104] Household energy consumption management module: used to predict the energy consumption of household appliances through the trained model, manage household energy consumption according to the prediction results, and judge whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, it will issue an energy consumption abnormality warning to the user and point out the potential cause of the abnormality according to the abnormal content. For example, high energy consumption indicates that one or some household appliances are faulty or aging. When there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household appliances according to the household energy consumption information. At the same time, the household energy consumption prediction model will re-output the prediction results according to the user's adjustment, and the cycle will be repeated.

[0105] The functions of the multifunctional electric energy meter used in the multifunctional electric energy meter data acquisition module include:

[0106] Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement;

[0107] Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement;

[0108] Data storage: Store the collected data according to the measurement date and time;

[0109] In the household energy consumption prediction model construction and training module, an improved LSTM long short-term memory algorithm is adopted. By changing the number of hidden layers and fully connected layers, a Dropout layer is added between the LSTM layers to optimize the training of the model, making it more suitable for learning and predicting the energy consumption data of household electrical appliances in the home.

[0110] The present application provides a household energy consumption management device combined with a multifunctional electric energy meter, which adopts a household energy consumption management method combined with a multifunctional electric energy meter in the above embodiment, and can solve the technical problem that the traditional household energy consumption management method cannot predict energy consumption anomalies and make timely adjustments. Compared with the prior art, the beneficial effects of the household energy consumption management device combined with a multifunctional electric energy meter provided by the present application are the same as the beneficial effects of the household energy consumption management method combined with a multifunctional electric energy meter provided by the above embodiment, and the other technical features of the household energy consumption management device combined with a multifunctional electric energy meter are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0111] The present application provides a household energy consumption management device combined with a multi-function electricity meter, and the household energy consumption management device combined with a multi-function electricity meter includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a household energy consumption management method combined with a multi-function electricity meter in the above-mentioned embodiment one.

[0112] Reference below Figure 3 , which shows a schematic diagram of the structure of a household energy consumption management device combined with a multifunctional electric energy meter suitable for implementing an embodiment of the present application. A household energy consumption management device combined with a multifunctional electric energy meter in an embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The household energy consumption management device combined with a multi-function electric energy meter shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0113] Figure 3The household energy consumption management device combined with a multifunctional electric energy meter shown may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of a household energy consumption management device combined with a multifunctional electric energy meter are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow a household energy consumption management device combined with a multifunctional electric energy meter to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a household energy consumption management device combined with a multifunctional electric energy meter with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0114] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0115] The present application provides a household energy consumption management device combined with a multifunctional electric energy meter, which adopts a household energy consumption management method combined with a multifunctional electric energy meter in the above embodiment, and can solve the technical problem that the traditional household energy consumption management method lacks accurate analysis and effective management of the energy consumption of various household electrical appliances. Compared with the prior art, the beneficial effects of the household energy consumption management device combined with a multifunctional electric energy meter provided by the present application are the same as the beneficial effects of the household energy consumption management method combined with a multifunctional electric energy meter provided by the above embodiment, and the other technical features of the household energy consumption management device combined with a multifunctional electric energy meter are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0116] The various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0117] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the household energy consumption management method combined with a multi-function electric energy meter as described above.

[0118] The computer program product provided by the present application can solve the technical problem that the traditional household energy consumption management method cannot predict energy consumption anomalies and make timely adjustments. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the household energy consumption management method combined with a multi-function electric energy meter provided in the above embodiment, which will not be described in detail here.

[0119] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A household energy consumption management method combined with a multifunctional electric energy meter, characterized in that: The method comprises the following steps: Step S10: using a multifunctional electric energy meter to collect energy consumption data of household electrical appliances in the home in real time, including voltage, current, power and power consumption duration, the collected data is time-stamped, and constitutes an energy consumption data set arranged in time series; Step S20: performing data preprocessing on the collected energy consumption data set, including removing outliers, data standardization and data cleaning; Step S30: extracting characteristic values ​​from the energy consumption data set after preprocessing, including the daily average energy consumption, energy consumption peak, energy consumption peak occurrence time and energy consumption fluctuation frequency of the household electrical appliances. The extracted characteristic values ​​are used to reflect the energy consumption mode and usage regularity of the household electrical appliances. Step S40: constructing a household energy consumption prediction model based on LSTM, and training the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted feature values; Step S50: predicting the energy consumption of household electrical appliances through the trained model, managing household energy consumption according to the prediction results, and judging whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, an energy consumption abnormality warning is issued to the user and the potential cause of the abnormality is pointed out according to the abnormality type; when there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household electrical appliances according to the household energy consumption information, and the household energy consumption prediction model re-outputs the prediction results according to the user's adjustment, and the cycle repeats; The multifunctional electric energy meter used in step S10 has the following functions: Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement; Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement; Data storage: Store the collected data according to the measurement date and time; In the step S40 of constructing a household energy consumption prediction model based on LSTM, an improved LSTM long short-term memory algorithm is adopted. By changing the number of hidden layers and fully connected layers, a Dropout layer is added between the LSTM layers to optimize the training of the model, making it more suitable for learning and predicting the energy consumption data of household electrical appliances in the home.

2. A household energy consumption management method combined with a multifunctional electric energy meter according to claim 1, characterized in that: The data collected in step S10 is time-stamped and constitutes an energy consumption data set arranged in time series. The implementation method is to set a real-time clock unit in the multi-function electric energy meter and synchronize the time regularly through the network to provide accurate time information; After the energy consumption data is collected, the current time provided by the real-time clock unit is used as a timestamp and stored together with the collected energy consumption data to form an energy consumption data set arranged in time series.

3. A household energy consumption management method combined with a multifunctional electric energy meter according to claim 1, characterized in that: The step S20 performs data preprocessing on the collected energy consumption data set, including: Removing outliers: The threshold method is used to set the threshold range to identify and remove outliers. The normal power range of household electrical appliances is P min To P max , then the threshold is set as shown in formula (1): P 阈值 =k×(P max -P min )+P min (1) Where k is a coefficient, which is set according to the specific household electrical appliance and has a value range of 1.05≤k≤1.

95. When the collected power value exceeds the set threshold, it is determined as an abnormal value and removed; Data standardization: Z-score standardization is used to convert the collected data into a standard normal distribution with a mean of 0 and a standard deviation of 1. The calculation formula is shown in formula (2): Where x is a data point in the data set, μ is the mean of the data set, σ is the standard deviation of the data set, and x is the standardized x; Data cleaning: missing values ​​were filled using the mean method.

4. A household energy consumption management method combined with a multifunctional electric energy meter according to claim 1, characterized in that: The step S30 extracts characteristic values ​​from the energy consumption data set after preprocessing, specifically including: Daily average energy consumption: Divide the total daily energy consumption by the number of statistical days to obtain the daily average energy consumption of the household appliance. The calculation formula is shown in formula (3): in is the daily average energy consumption, n is the number of days, E j is the energy consumption on the jth day; Energy consumption peak: The peak value of energy consumption is determined by traversing the energy consumption data set. The energy consumption data set sequence is P = [P1, P2, ..., P m ], set P max = P1, then for i = 2 to i = m, when P i >P max When P max = P1, the final P obtained after traversing the energy consumption data set max It is the peak energy consumption; Energy consumption peak occurrence time: The data collected in step S10 is time-stamped, forming an energy consumption data set arranged in time series. When the energy consumption peak data is obtained, the peak data is time-stamped, and the energy consumption peak occurrence time is obtained; Energy consumption fluctuation frequency: The energy consumption fluctuation frequency is calculated by analyzing the change of energy consumption data over time. The calculation formula is shown in formula (4): Where f is the energy consumption fluctuation frequency, a is the number of data points, Δt is the time interval, P b is the energy consumption value at the bth time point, P b+1 is the energy consumption value at the b+1th time point.

5. A household energy consumption management method combined with a multifunctional electric energy meter according to claim 1, characterized in that: In step S40, a household energy consumption prediction model based on LSTM is constructed, and the household energy consumption prediction model based on LSTM is trained according to the energy consumption data set and the extracted feature values. The model construction and training process includes: Dataset division: The energy consumption dataset is divided into a training set and a validation set, with the training set accounting for 80% and the validation set accounting for 20%; Model construction: Adopt LSTM-based long-term and short-term neural network architecture, including input layer, hidden layer, and fully connected layer; the input layer contains 5 nodes, corresponding to the energy consumption data set and the four extracted feature values ​​of daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency; use LSTM layer as hidden layer, and add Dropout layer between each LSTM layer. Each LSTM layer contains 64 LSTM units, and each LSTM unit has input gate, forget gate and output gate for processing sequence data; the fully connected layer is used as the output layer, which contains 5 nodes, corresponding to the 5 nodes of the input layer, and is used to map the output of the hidden layer to the final output. The output value is the predicted energy consumption of household appliances and daily average energy consumption, energy consumption peak, energy consumption peak occurrence time, and energy consumption fluctuation frequency; Model training: The mean square error loss function is used, and the formula is shown in formula (5): Where a is the number of data points, y i is the i-th true value, is the i-th predicted value, the true value and pre-test include energy consumption data, average daily energy consumption, energy consumption peak, time of occurrence of energy consumption peak and frequency of energy consumption fluctuation; Adam optimizer is used as the optimization algorithm to update the weights in model training to minimize the loss function; for each data in the training set, it is input into the LSTM-based household energy consumption prediction model in batches, and the output is calculated by forward propagation; the weight is calculated through the loss function according to the output and the true value; the gradient of the loss function for each weight is calculated through back propagation according to the weight; the weight is updated through the Adam optimizer according to the gradient; the above steps are repeated to traverse all batches of data to complete a training round; after each training round, the performance of the model is evaluated using the validation set, and the model training is completed when the loss no longer decreases.

6. A household energy consumption management method combined with a multifunctional electric energy meter according to claim 1, characterized in that: The step S50 manages the household energy consumption according to the prediction result and determines whether the household energy consumption is abnormal according to the set energy consumption threshold. The threshold setting method is consistent with the threshold setting method in the above-mentioned step of removing abnormal values. When the result predicted by the household energy consumption prediction model exceeds the threshold, an energy consumption abnormality warning is issued to the user through an alarm, text message and flashing light; in the step S50, the potential cause of the abnormality is pointed out according to the abnormality type. The data of the energy consumption abnormality type and the corresponding cause of the abnormality are obtained from the Internet, and are used as the input of the household energy consumption prediction model. When an abnormality occurs, an energy consumption abnormality warning is issued to the user, and a matching abnormality type and a potential cause of the abnormality are given.

7. A household energy consumption management device combined with a multifunctional electric energy meter, characterized in that: The household energy consumption management device combined with a multifunctional electric energy meter comprises: Multifunctional energy meter data collection module: used to use the multifunctional energy meter to collect energy consumption data of household electrical appliances in real time, including voltage, current, power and power consumption duration. The collected data is time-stamped and forms an energy consumption data set arranged in time series; Energy consumption data preprocessing module: used to preprocess the collected energy consumption data set, including removing outliers, data standardization and data cleaning; Energy consumption data feature extraction module: used to extract feature values ​​from the pre-processed energy consumption data set, including the average daily energy consumption, peak energy consumption, peak energy consumption time and energy consumption fluctuation frequency of household electrical appliances. The extracted feature values ​​are used to reflect the energy consumption mode and usage rules of household electrical appliances. Household energy consumption prediction model construction and training module: used to build a household energy consumption prediction model based on LSTM, and train the household energy consumption prediction model based on LSTM according to the energy consumption data set and the extracted feature values; Household energy consumption management module: used to predict the energy consumption of household appliances through the trained model, manage household energy consumption according to the prediction results, and judge whether the household energy consumption is abnormal according to the set energy consumption threshold. When an abnormality occurs, it will issue an energy consumption abnormality warning to the user and point out the potential cause of the abnormality according to the abnormal content. For example, high energy consumption indicates that one or some household appliances are faulty or aging. When there is no abnormality, the current household energy consumption information is displayed on the display screen, so that the user can adjust the use of household appliances according to the household energy consumption information. At the same time, the household energy consumption prediction model will re-output the prediction results according to the user's adjustment, and the cycle will be repeated. The functions of the multifunctional electric energy meter used in the multifunctional electric energy meter data acquisition module include: Basic electric energy measurement: including active electric energy measurement, reactive electric energy measurement and total electric energy measurement; Electrical parameter measurement: including voltage measurement, current measurement, power measurement, power usage time measurement and frequency measurement; Data storage: Store the collected data according to the measurement date and time; In the household energy consumption prediction model construction and training module, an improved LSTM long short-term memory algorithm is adopted. By changing the number of hidden layers and fully connected layers, a Dropout layer is added between the LSTM layers to optimize the training of the model, making it more suitable for learning and predicting the energy consumption data of household electrical appliances in the home.

8. A household energy consumption management device combined with a multifunctional electric energy meter, characterized in that: The household energy consumption management device combined with a multifunctional electric energy meter comprises: A memory, a processor, and an LSTM-based household energy consumption prediction program stored in the memory and executable on the processor. When the LSTM-based household energy consumption prediction program is executed by the processor, a household energy consumption management method combined with a multi-function electric energy meter is implemented as described in any one of claims 1 to 6.

9. A computer program product, characterized in that The computer program product includes a LSTM-based household energy consumption prediction program, which, when executed by a processor, implements a household energy consumption management method combined with a multi-function electric energy meter as described in any one of claims 1 to 6.

Images