Household energy consumption prediction and optimization control method based on intelligent router

Abstract

The invention discloses a household energy consumption prediction and optimization control method based on an intelligent router, relates to the technical field of energy consumption prediction analysis, and solves the technical problem that the prediction result is inaccurate due to the fact that historical energy consumption data of a workday may be utilized to predict energy consumption data of a rest day in the prior art. The intelligent router collects energy consumption data, real-time temperature and user behavior data of electrical equipment through a wireless network; predicting the energy consumption in a set time period according to the energy consumption data of the electrical equipment to obtain predicted energy consumption data; correcting the predicted energy consumption data according to the real-time temperature and the user behavior data to obtain corrected energy consumption data; acquiring real-time electricity price information, and adjusting parameters of the electrical equipment according to the real-time electricity price information and the corrected energy consumption data; sending the adjusted parameters of the electrical equipment to a user for confirmation, and adjusting the parameters of the electrical equipment according to a confirmation result of the user; and the accuracy of the prediction result is improved.

Description

Technical Field

[0001] The present invention belongs to the field of energy consumption prediction and analysis, and relates to the technology of home energy consumption prediction and optimization control. Specifically, it is a method for home energy consumption prediction and optimization control based on an intelligent router. Background Art

[0002] An intelligent router is a router with intelligent management. It can serve as the central node of a smart home system to connect and control smart home devices, realizing the intelligent management of the devices; the intelligent router can accurately predict the future energy consumption trend by real-time detecting the energy consumption data of home devices and using big data and machine learning technologies, which helps home users understand the energy consumption situation in advance and reasonably arrange the electricity usage plan; the intelligent router can intelligently allocate energy consumption according to the prediction results, giving priority to ensuring the power supply of key devices while reducing the energy consumption of non-essential devices. By reasonably and intelligently allocating energy consumption, the resource utilization rate can be significantly improved and the waste of energy consumption can be reduced; the intelligent router supports the remote monitoring function, allowing users to view the home energy consumption situation at any time and perform remote control.

[0003] When predicting the home energy consumption currently, the energy consumption data of the home is collected by intelligent sensors. After storing and integrating the collected energy consumption data, historical energy consumption data is obtained; according to the historical energy consumption data and weather data, a model is used to predict the energy consumption data of the home to obtain the predicted energy consumption data; however, the energy consumption data of each family is different at different time periods. When using the historical energy consumption data to predict the energy consumption data of a set time period, it may occur that the historical energy consumption data of weekdays is used to predict the energy consumption data of weekends, resulting in inaccurate prediction results and affecting the user experience.

[0004] The present invention provides a method for home energy consumption prediction and optimization control based on an intelligent router to solve the above technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present invention proposes a method for home energy consumption prediction and optimization control based on an intelligent router, which is used to solve the technical problem that when using historical energy consumption data to predict the energy consumption data of a set time period in the prior art, it may occur that the historical energy consumption data of weekdays is used to predict the energy consumption data of weekends, resulting in inaccurate prediction results and affecting the user experience.

[0006] To achieve the above object, the first aspect of the present invention provides a method for home energy consumption prediction and optimization control based on an intelligent router, including:

[0007] Step S1: The intelligent router collects the energy consumption data, real-time temperature, and user behavior data of electrical appliances through the wireless network;

[0008] Step S2: Predict the energy consumption within a set time period based on the energy consumption data of the electrical appliances to obtain predicted energy consumption data; correct the predicted energy consumption data according to the real-time temperature and user behavior data to obtain corrected energy consumption data;

[0009] Step S3: Obtain the real-time electricity price information, and adjust the parameters of the electrical appliances according to the real-time electricity price information and the corrected energy consumption data;

[0010] Step S4: Send the adjusted parameters of the electrical appliances to the user for confirmation, and adjust the parameters of the electrical appliances according to the user's confirmation result.

[0011] Preferably, the predicting the energy consumption within a set time period based on the energy consumption data of the electrical appliances includes:

[0012] Retrieve the energy consumption data of the electrical appliances; wherein, the energy consumption data includes historical power consumption data, historical gas consumption data, and historical water consumption data;

[0013] Divide the energy consumption data into several energy consumption arrays according to the set time period; sort the several energy consumption data in chronological order to obtain an energy consumption prediction array;

[0014] Analyze the energy consumption prediction array using an energy consumption prediction model to obtain predicted energy consumption data; wherein, the energy consumption prediction model is constructed based on an artificial intelligence model.

[0015] The present invention divides the energy consumption data into several energy consumption arrays according to the set time period, sorts the energy consumption arrays in chronological order to obtain an energy consumption prediction array, and analyzes the energy consumption prediction array using a trained energy consumption prediction model to obtain predicted energy consumption data, which can predict the energy consumption data within a set time period based on the existing historical energy consumption data, making the prediction process more convenient and conducive to improving the prediction rate.

[0016] Preferably, the energy consumption prediction model is constructed based on an artificial intelligence model, including:

[0017] Select a suitable model type and deep learning framework from several artificial intelligence models; define the structure of the model and set the training parameters according to the selected deep learning framework to obtain a training model;

[0018] Obtain a standard data set; wherein, the standard data set includes standard input data consistent with the content attributes of the energy consumption prediction array and standard output data consistent with the content attributes of the predicted energy consumption data;

[0019] The standard data set is used to train, validate, optimize, and test the training model according to the set ratio. When the test index of the training model is greater than the set standard, the training model is marked as the energy consumption prediction model; otherwise, the artificial intelligence model is reconstructed and trained again.

[0020] Preferably, predicting the energy consumption within a set time period according to the energy consumption data of the electrical equipment includes:

[0021] Retrieving the energy consumption data of the electrical equipment; obtaining the time label of the set time period to be predicted; wherein, the time label includes: weekdays, rest days, and holidays;

[0022] Dividing the energy consumption data of the electrical equipment according to the data type, and extracting the characteristic energy consumption data corresponding to the time from the divided energy consumption data according to the time label of the set time period to be predicted; integrating the characteristic energy consumption data into a discrete characteristic curve, and analyzing the fluctuation coefficient of the discrete characteristic curve; calculating the predicted energy consumption data according to the mean value and the fluctuation coefficient of the characteristic energy consumption data.

[0023] It should be noted that the way to obtain the fluctuation coefficient is: calculating the average value of the characteristic energy consumption data; calculating the standard deviation of the characteristic energy consumption data according to the standard deviation calculation formula; calculating the percentage of the quotient of the standard deviation and the average value to obtain the fluctuation coefficient; the fluctuation coefficient is a coefficient used to reflect the fluctuation degree of the characteristic energy consumption data.

[0024] The present invention extracts the corresponding characteristic energy consumption data from the energy consumption data of the electrical equipment according to the time label of the set time period to be predicted, integrates the characteristic energy consumption data into a discrete characteristic curve, and calculates the predicted energy consumption data according to the fluctuation coefficient and the mean value of the characteristic energy consumption data; it can predict according to the characteristics of the actual predicted time period and the historical energy consumption data, making the prediction result more in line with the actual situation, which is beneficial to improving the accuracy of the prediction result.

[0025] Preferably, correcting the predicted energy consumption data according to the real-time temperature and the user behavior data includes:

[0026] Retrieving the real-time temperature and the user behavior data; comparing the real-time temperature with the temperature threshold; wherein, the temperature threshold is set according to the user behavior data; the temperature threshold includes: a low temperature threshold and a high temperature threshold;

[0027] When the real-time temperature is greater than or equal to the high temperature threshold or the real-time temperature is less than or equal to the low temperature threshold, the predicted energy consumption data is corrected according to the real-time temperature;

[0028] Obtain the instruction data of the user; among them, the instruction data includes: the operating state of the electrical appliance, the device parameters of the electrical appliance, and the operating time; when the operating state of the electrical appliance is on, calculate the operating energy consumption data according to the device parameters and operating time of the electrical appliance, and calculate the sum of the operating energy consumption data and the predicted energy consumption data to obtain the corrected energy consumption data.

[0029] It should be noted that different users have different sensitivities to temperature. Some users like a cool environment, and some users like a warm environment. Setting the temperature threshold according to the user's behavior data can ensure that the temperature controlled by the smart home makes the user more comfortable and satisfied.

[0030] The present invention sets the temperature threshold according to the user's behavior data, and compares the real-time temperature with the temperature threshold. When the real-time temperature is greater than or equal to the high temperature threshold, or the real-time temperature is less than or equal to the low temperature threshold, it will control some smart homes to start running, resulting in a deviation in the predicted energy consumption data. Therefore, correcting the energy consumption data according to the real-time temperature can ensure the accuracy of the prediction result; and calculating the operating energy consumption data according to the user's instructions, and calculating the corrected energy consumption data according to the operating energy consumption data and the predicted energy consumption data can dynamically adjust the prediction result according to the user's instructions, which is beneficial to making the prediction result more accurate.

[0031] Preferably, the correction of the predicted energy consumption data according to the real-time temperature includes:

[0032] Retrieve the real-time temperature, and calculate the difference between the real-time temperature and the temperature threshold; analyze the sensitivity coefficient between the energy consumption data and the real-time temperature according to the energy consumption data;

[0033] Calculate the corrected energy consumption data through the difference, the sensitivity coefficient, and the mapping relationship between the predicted energy consumption data and the corrected energy consumption data.

[0034] It should be noted that the calculation formula for the corrected energy consumption data is: XN = (1 + α × MG × |CZ|) × YN; where XN represents the corrected energy consumption data; MG represents the sensitivity coefficient; CZ represents the difference; YN represents the predicted energy consumption data; α represents the error adjustment coefficient, which is used to reduce the error caused by inaccurate acquisition, inaccurate predicted energy consumption data and other factors during the correction process.

[0035] The present invention corrects the predicted energy consumption data according to the difference between the real-time temperature and the temperature threshold and the sensitivity coefficient, considering the sensitivity coefficient between the energy consumption data and the real-time temperature, making the corrected energy consumption data closer to the actual energy consumption data, which is beneficial to providing accurate prediction results for users and improving the user experience and satisfaction.

[0036] Preferably, the analysis of the sensitivity coefficient between the energy consumption data and the real-time temperature according to the energy consumption data includes:

[0037] The energy consumption analysis model is constructed as follows: Among them, NH represents energy consumption data; MG represents the sensitivity coefficient; SW represents the real-time temperature; GT represents the high-temperature threshold; DT represents the low-temperature threshold; β1 and β2 are both the reference values of energy consumption data;

[0038] Retrieve the energy consumption data and the corresponding historical temperature data; use the energy consumption data and the historical temperature data to train the energy consumption analysis model, and calculate the corresponding sensitivity coefficient.

[0039] It should be noted that the reference values β1 and β2 of the energy consumption data are set according to the actual situation. β1 is the energy consumption reference value in spring, and β2 is the energy consumption reference value in autumn. When the difference between the two energy consumption reference values is small, β1 and β2 can be set to the same value.

[0040] Preferably, adjusting the parameters of the electrical equipment according to the real-time electricity price information and the corrected energy consumption data includes:

[0041] Retrieve the real-time electricity price information and the corrected energy consumption data; divide the set time period into a normal price area and a special price area according to the real-time electricity price information; when the corrected energy consumption data is greater than the energy consumption threshold, adjust the parameters of the electrical equipment; otherwise, keep the parameters of the electrical equipment running;

[0042] Obtain the unit energy consumption and the running time of the parameters of the electrical equipment; compare the running time with the divided areas of the set time period to obtain the first running time and the second running time; calculate the total energy consumption corresponding to several parameters in each running time according to the unit energy consumption and the running time of the electrical equipment; calculate the electricity price of several parameters in each running time according to the electricity price information; screen out the adjustment parameters from the corresponding parameters according to the electricity price.

[0043] It should be noted that the energy consumption threshold is the average energy consumption of the previous year. The purpose of comparing the corrected energy consumption data with the energy consumption threshold is to adjust the parameters of the electrical equipment when the corrected energy consumption data exceeds the average energy consumption, so as to reduce the electricity price cost.

[0044] The present invention divides the set time period into a normal price area and a special price area according to the real-time electricity price; when the corrected energy consumption data is greater than the energy consumption threshold, obtain the unit energy consumption and the running time of several parameters of the electrical equipment; calculate the total energy consumption of each running time according to the unit energy consumption and the running time of the electrical equipment, and calculate the electricity price according to the electricity price information, providing a data basis for subsequent analysis and screening of adjustment parameters.

[0045] Preferably, screening out the adjustment parameters from the corresponding parameters according to the electricity price includes:

[0046] Retrieve the electricity prices of several parameters for each period of running time; divide the electricity prices according to the running time and number them as YDij; where, i represents the i-th period of running time, i = 1, 2; j represents the j-th parameter; j = 1, 2, …, n, and n is a positive integer representing the number of parameters;

[0047] Obtain the parameter ranges of each period of running time set by the user, and eliminate the electricity prices of each period of running time that are not within the parameter ranges; by calculating the sum of the electricity prices of two periods of running time, obtain the corresponding total electricity cost; sort the corresponding parameters in ascending order of the total electricity cost to obtain a parameter screening table; select the parameter with the lowest electricity cost as the adjustment parameter.

[0048] According to the parameter ranges of each period of running time, the present invention eliminates the electricity prices that are not within the parameter ranges, calculates the total sum of the electricity prices of two periods of running time, obtains the corresponding total electricity cost and constructs a parameter screening table, and selects the parameter with the lowest electricity cost from the parameter screening table as the adjustment parameter; it can provide users with affordable adjustment parameters for electrical equipment, which is beneficial to reducing the electricity cost while meeting the needs of users.

[0049] Preferably, adjusting the parameters of the electrical equipment according to the confirmation result of the user includes:

[0050] Retrieve the confirmation result of the user; where, the confirmation result of the user includes: confirm execution and not execute; when the confirmation result of the user is confirm execution, then adjust the parameters of the electrical equipment according to the selected adjustment parameters;

[0051] When the confirmation result of the user is not execute, then obtain the adjustment information of the user; if the user sends an adjustment instruction, then adjust the parameters of the electrical equipment according to the adjustment instruction of the user; otherwise, keep the parameters of the electrical equipment running.

[0052] The present invention adjusts the parameters of the electrical equipment according to the confirmation information of the user, enabling users to fully understand the electricity consumption situation of the family and the optimal parameters of the electrical equipment. Only after the user confirms, the parameters of the electrical equipment are adjusted, which is beneficial to improving the user experience and enhancing user satisfaction.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] 1. The present invention divides energy consumption data into several energy consumption arrays according to a set time period, and sorts the energy consumption arrays in chronological order to obtain an energy consumption prediction array; analyzes the energy consumption prediction array using a trained energy consumption prediction model to obtain predicted energy consumption data; can predict the energy consumption data for a set time period based on existing historical energy consumption data, making the prediction process more convenient and conducive to improving the prediction rate; extracts corresponding characteristic energy consumption data from the energy consumption data of electrical equipment according to the time label of the predicted set time period, and integrates the characteristic energy consumption data into a discrete characteristic curve, and calculates the predicted energy consumption data according to the fluctuation coefficient of the discrete characteristic curve and the mean value of the characteristic energy consumption data; can predict according to the characteristics of the actual predicted time period and historical energy consumption data, making the prediction result more in line with the actual situation and conducive to improving the accuracy of the prediction result; sets a temperature threshold according to the user's behavior data, and compares the real-time temperature with the temperature threshold. When the real-time temperature is greater than or equal to the high temperature threshold, or the real-time temperature is less than or equal to the low temperature threshold, some smart homes will be controlled to start running, resulting in a deviation in the predicted energy consumption data. Therefore, correcting the energy consumption data according to the real-time temperature can ensure the accuracy rate of the prediction result; and calculates the operating energy consumption data according to the user's instruction, and calculates the corrected energy consumption data according to the operating energy consumption data and the predicted energy consumption data, can dynamically adjust the prediction result according to the user's instruction, which is conducive to making the prediction result more accurate; corrects the predicted energy consumption data according to the difference between the real-time temperature and the temperature threshold and the sensitivity coefficient, considering the sensitivity coefficient of the energy consumption data and the real-time temperature, making the corrected energy consumption data closer to the actual energy consumption data, which is conducive to providing accurate prediction results for users and improving the user experience and satisfaction.

[0055] 2. The present invention divides the set time period into a normal price area and a special price area according to the real-time electricity price; when the corrected energy consumption data is greater than the energy consumption threshold, the unit energy consumption and running time of several parameters of the electrical equipment are obtained; calculates the total energy consumption of each running time according to the unit energy consumption and running time of the electrical equipment, and calculates the electricity price according to the electricity price information, providing a data basis for subsequent analysis and screening of adjustment parameters; according to the parameter range of each running time, eliminates the electricity prices outside the parameter range, and calculates the sum of the electricity prices of two running times to obtain the corresponding total electricity cost and constructs a parameter screening table, and selects the parameter with the lowest electricity cost from the parameter screening table as the adjustment parameter; can provide users with economical and affordable adjustment parameters for electrical equipment, which is conducive to reducing the electricity cost while meeting the user's needs; adjusts the parameters of the electrical equipment according to the user's confirmation information, enabling users to fully understand the electricity consumption situation of the family and the optimal parameters of the electrical equipment. Only when the user confirms, the parameters of the electrical equipment are adjusted, which is conducive to improving the user experience and enhancing the user's satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0057] Figure 1 It is a schematic diagram of the overall solution steps of the present invention;

[0058] Figure 2 It is a schematic diagram of the energy consumption data prediction and correction steps of the present invention;

[0059] Figure 3 It is a schematic diagram of the parameter adjustment steps of the electrical equipment of the present invention. Specific embodiments

[0060] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0061] Please refer to Figure 1 , an embodiment of the first aspect of the present invention provides a home energy consumption prediction and optimization control method based on an intelligent router, including:

[0062] Step S1: The intelligent router collects the energy consumption data, real-time temperature, and user behavior data of the electrical equipment through the wireless network;

[0063] Step S2: Predict the energy consumption within a set time period based on the energy consumption data of the electrical equipment to obtain predicted energy consumption data; correct the predicted energy consumption data according to the real-time temperature and user behavior data to obtain corrected energy consumption data;

[0064] Step S3: Obtain the real-time electricity price information, and adjust the parameters of the electrical equipment according to the real-time electricity price information and the corrected energy consumption data;

[0065] Step S4: Send the adjusted parameters of the electrical equipment to the user for confirmation, and adjust the parameters of the electrical equipment according to the user's confirmation result.

[0066] Please refer to Figure 2, the smart router collects the energy consumption data, real-time temperature, and user behavior data of electrical appliances through the wireless network; retrieves the energy consumption data of electrical appliances; obtains the time tags for the predicted set time period; where the time tags include: weekdays, rest days, and holidays; the energy consumption data includes: historical electricity consumption data, historical gas consumption data, and historical water consumption data;

[0067] Divide the energy consumption data of electrical appliances according to the data type, and extract the characteristic energy consumption data corresponding to the time from the divided energy consumption data according to the time tags of the predicted set time period; integrate the characteristic energy consumption data into a discrete characteristic curve, and analyze the fluctuation coefficient of the discrete characteristic curve; calculate the predicted energy consumption data based on the mean value and the fluctuation coefficient of the characteristic energy consumption data.

[0068] It should be noted that the way to obtain the fluctuation coefficient is: calculate the average value of the characteristic energy consumption data; calculate the standard deviation of the characteristic energy consumption data according to the standard deviation calculation formula; calculate the percentage of the quotient of the standard deviation and the average value to obtain the fluctuation coefficient; the fluctuation coefficient is a coefficient used to reflect the degree of fluctuation of the characteristic energy consumption data.

[0069] It should be explained that the calculation formula for the predicted energy consumption data is: predicted energy consumption data = mean value of the characteristic data × (1 + fluctuation coefficient); the mean value reflects the benchmark value of the energy consumption data, and the fluctuation coefficient can dynamically adjust the benchmark value.

[0070] For example: Suppose we are now predicting the electricity consumption data in the predicted energy consumption data for Saturday, and the obtained time tag for Saturday is a rest day; screen out the historical electricity consumption data with the time tag of rest day from the energy consumption data; integrate the historical electricity consumption data into a discrete characteristic curve, and through analysis and calculation, the fluctuation coefficient of the discrete characteristic curve is obtained as 0.16; calculate the mean value of the historical electricity energy consumption data as 2.5 kW; then calculate the predicted energy consumption data as 2.9 kW according to the calculation formula for the predicted energy consumption data.

[0071] Retrieve the real-time temperature and user behavior data; compare the real-time temperature with the temperature threshold; where the temperature threshold is set according to the user's behavior data; the temperature threshold includes: low temperature threshold and high temperature threshold;

[0072] When the real-time temperature is greater than or equal to the high temperature threshold or the real-time temperature is less than or equal to the low temperature threshold, then correct the predicted energy consumption data according to the real-time temperature;

[0073] Obtain the user's instruction data; where the instruction data includes: the operating status of the electrical appliance, the device parameters of the electrical appliance, and the operating time; when the operating status of the electrical appliance is on, then calculate the operating energy consumption data according to the device parameters and the operating time of the electrical appliance, and calculate the sum of the operating energy consumption data and the predicted energy consumption data to obtain the corrected energy consumption data;

[0074] The energy consumption analysis model is constructed as follows: Among them, NH represents energy consumption data; MG represents the sensitivity coefficient; SW represents the real-time temperature; GT represents the high-temperature threshold; DT represents the low-temperature threshold; β1 and β2 are both reference values of energy consumption data; retrieve the energy consumption data and the corresponding historical temperature data; use the energy consumption data and the historical temperature data to train the energy consumption analysis model, and calculate the corresponding sensitivity coefficient.

[0075] Retrieve the real-time temperature and calculate the difference between the real-time temperature and the temperature threshold; through the difference, the sensitivity coefficient, and the mapping relationship between the predicted energy consumption data and the corrected energy consumption data, calculate the corrected energy consumption data.

[0076] It should be noted that the calculation formula for the corrected energy consumption data is: XN = (1 + α × MG × |CZ|) × YN; where XN represents the corrected energy consumption data; MG represents the sensitivity coefficient; CZ represents the difference; YN represents the predicted energy consumption data; α represents the error adjustment coefficient, which is used to reduce the errors caused by inaccurate acquisition, inaccurate predicted energy consumption data, and other factors during the correction process.

[0077] It should be noted that the more sensitive the corresponding energy consumption data is to temperature, the greater the corresponding sensitivity coefficient. The greater the difference between the real-time temperature and the temperature threshold, the greater the difference between the corrected energy consumption data and the predicted energy consumption data.

[0078] For example: Assume that the predicted energy consumption data is corrected, compare the real-time temperature with the temperature threshold, and the real-time temperature is greater than the high-temperature threshold; correct the predicted energy consumption data according to the sensitivity coefficient, and set the error adjustment coefficient α to 1; as shown in the following table:

[0079] Energy consumption type Temperature difference Sensitivity coefficient Predicted energy consumption data Corrected energy consumption data Electricity 5 0.15 2.9 kW 5.075 kW Gas 5 0.05 <![CDATA[1.5m 3 > <![CDATA[1.875m 3 > Water volume 5 0.03 320 liters 368 liters

[0080] Table 1: Schematic diagram of the correction result

[0081] In another preferred embodiment, predicting the energy consumption within a set time period according to the energy consumption data of the electrical equipment includes:

[0082] Retrieve the energy consumption data of the electrical equipment; among them, the energy consumption data includes: historical power consumption data, historical gas consumption data, and historical water consumption data;

[0083] Divide the energy consumption data into several energy consumption arrays according to the set time period; sort the several energy consumption data in chronological order to obtain an energy consumption prediction array;

[0084] Use the energy consumption prediction model to analyze the energy consumption prediction array to obtain the predicted energy consumption data; among them, the energy consumption prediction model is constructed based on an artificial intelligence model.

[0085] It should be noted that the energy consumption prediction model is constructed based on an artificial intelligence model, including:

[0086] Select a suitable model type and deep learning framework from several artificial intelligence models; define the structure of the model and set the training parameters according to the selected deep learning framework to obtain a training model;

[0087] Obtain a standard data set; where the standard data set includes standard input data consistent with the content attributes of the energy consumption prediction array and standard output data consistent with the content attributes of the predicted energy consumption data;

[0088] Train, validate, optimize, and test the training model with the standard data set according to a set ratio; when the test index of the training model is greater than the set standard, mark the training model as the energy consumption prediction model; otherwise, reconstruct and train the artificial intelligence model again.

[0089] It should be noted that the ratio for dividing the standard data set is set by experimental simulation, and the test indexes include: accuracy, recall rate, stability, and F1 score; the standard is set by expert evaluation; when reconstructing the model, the ratio for dividing the standard data set can be adjusted, or the internal parameters of the constructed model can be adjusted.

[0090] Please refer to Figure 3 to retrieve the real-time electricity price information and corrected energy consumption data; divide the set time period into a normal price area and a special offer price area according to the real-time electricity price information; when the corrected energy consumption data is greater than the energy consumption threshold, adjust the parameters of the electrical equipment; otherwise, keep the parameters of the electrical equipment running;

[0091] Obtain the unit energy consumption and running time of the parameters of the electrical equipment; compare the running time with the divided areas of the set time period to obtain the first running time and the second running time; calculate the total energy consumption corresponding to several parameters during each running time according to the unit energy consumption and running time of the electrical equipment; calculate the electricity price of several parameters during each running time according to the electricity price information.

[0092] For example: Assume that Saturday in the set time period is obtained, and the corrected energy consumption data for Saturday calculated according to the above steps is 5.075 kW; which is greater than the energy consumption threshold of 4.32 kW; then analyze the electricity price; according to the real-time electricity price information, divide 00:00 - 6:00 and 22:00 - 23:59 on Saturday into the preferential price area, and divide 6:00 - 22:00 into the normal price area; now adjust the parameters of the air conditioner; the operating time of the air conditioner is 13:00 - 23:00; then divide the operating time of the air conditioner into two segments as 13:00 - 22:00 and 22:00 - 23:00 and mark them as operating time period 1 and operating time period 2; calculate the electricity price of the air conditioner in each operating time period as shown in the following table:

[0093] Parameter Operating period 1 Operating period 2 30 degrees Celsius 6.13 0.46 28 degrees Celsius 5.11 0.38 26 degrees Celsius 7.16 0.53 …… …… …… 20 degrees Celsius 11.76 0.88

[0094] Table 2: Schematic table of electricity prices

[0095] Retrieve the electricity prices of several parameters for each operating time period; divide the electricity prices according to the operating time and number them as YDij; where, i represents the i-th operating time period, i = 1, 2; j represents the j-th parameter; j = 1, 2,..., n, and n is a positive integer representing the number of parameters;

[0096] Obtain the parameter range set by the user for each operating time period, and eliminate the electricity prices of each operating time period that are not within the parameter range; by calculating the sum of the electricity prices of the two operating time periods, obtain the corresponding total electricity cost; sort the corresponding parameters in ascending order of the total electricity cost to obtain a parameter screening table; select the parameter with the lowest electricity cost as the adjustment parameter.

[0097] For example: Assume that the parameter range in operating time period 1 is 20 - 26 degrees Celsius, and the parameter range in operating time period 2 is 26 - 30 degrees Celsius; then eliminate 28 and 30 degrees Celsius in operating time period 1, and eliminate 20 degrees Celsius in operating time period 2, and calculate the total electricity cost; obtain that the air conditioner parameter setting in operating time period 1 is 26 degrees Celsius, and the air conditioner parameter setting in operating time period 2 is 28 degrees Celsius.

[0098] Send the parameters of the adjusted electrical equipment to the user for confirmation; where, the user's confirmation result includes: confirm execution and not execute; when the user's confirmation result is confirm execution, then adjust the parameters of the electrical equipment according to the selected adjustment parameters; when the user's confirmation result is not execute, then obtain the user's adjustment information; if the user sends an adjustment instruction, then adjust the parameters of the electrical equipment according to the user's adjustment instruction; otherwise, keep the parameters of the electrical equipment running.

[0099] Some of the data in the above formula are taken as numerical values after removing the dimension. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the actual situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0100] The working principle of the present invention: The intelligent router of the present invention collects the energy consumption data, real-time temperature and user behavior data of electrical appliances through the wireless network; predicts the energy consumption within a set time period according to the energy consumption data of the electrical appliances to obtain predicted energy consumption data; corrects the predicted energy consumption data according to the real-time temperature and user behavior data to obtain corrected energy consumption data; obtains the real-time electricity price information, and adjusts the parameters of the electrical appliances according to the real-time electricity price information and the corrected energy consumption data; sends the adjusted parameters of the electrical appliances to the user for confirmation, and adjusts the parameters of the electrical appliances according to the confirmation result of the user.

[0101] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for predicting and optimizing the control of home energy consumption based on an intelligent router, characterized in that, Including: Step S1: The intelligent router collects the energy consumption data, real-time temperature, and user behavior data of electrical appliances through the wireless network. Step S2: Predict the energy consumption within a set time period based on the energy consumption data of electrical appliances to obtain predicted energy consumption data; correct the predicted energy consumption data according to the real-time temperature and user behavior data to obtain corrected energy consumption data. Step S3: Obtain the real-time electricity price information, and adjust the parameters of electrical appliances according to the real-time electricity price information and the corrected energy consumption data. Step S4: Send the adjusted parameters of the electrical appliances to the user for confirmation, and adjust the parameters of the electrical appliances according to the user's confirmation result.

2. The home energy consumption prediction and optimization control method based on an intelligent router according to claim 1, wherein The prediction of the energy consumption within a set time period based on the energy consumption data of electrical appliances includes: Retrieve the energy consumption data of electrical appliances; among them, the energy consumption data includes: historical power consumption data, historical gas consumption data, and historical water consumption data. Divide the energy consumption data into several energy consumption arrays according to the set time period; sort the several energy consumption data in chronological order to obtain an energy consumption prediction array. Analyze the energy consumption prediction array using an energy consumption prediction model to obtain predicted energy consumption data; among them, the energy consumption prediction model is constructed based on an artificial intelligence model.

3. The method for predicting and optimizing the control of household energy consumption based on an intelligent router according to claim 2, wherein, The energy consumption prediction model is constructed based on an artificial intelligence model, including: Select a suitable model type and deep learning framework from several artificial intelligence models; define the structure of the model and set the training parameters according to the selected deep learning framework to obtain a training model. Obtain a standard data set; among them, the standard data set includes standard input data consistent with the content attributes of the energy consumption prediction array, and standard output data consistent with the content attributes of the predicted energy consumption data. Train, verify, optimize, and test the training model with the standard data set according to the set ratio; when the test index of the training model is greater than the set standard, mark the training model as an energy consumption prediction model; otherwise, reconstruct and train the artificial intelligence model again.

4. A method for predicting and optimizing control of household energy consumption based on an intelligent router according to claim 1, characterized in that, The prediction of the energy consumption within a set time period based on the energy consumption data of electrical appliances includes: Retrieve the energy consumption data of electrical appliances; obtain the time label of the predicted set time period; among them, the time label includes: weekdays, rest days, and holidays. Divide the energy consumption data of electrical appliances according to the data type, and extract the characteristic energy consumption data corresponding to the time from the divided energy consumption data according to the time label of the predicted set time period; integrate the characteristic energy consumption data into a discrete characteristic curve, and analyze the fluctuation coefficient of the discrete characteristic curve; calculate the predicted energy consumption data according to the mean value and fluctuation coefficient of the characteristic energy consumption data.

5. The method for predicting and optimizing control of household energy consumption based on an intelligent router according to claim 1, wherein The correction of the predicted energy consumption data according to the real-time temperature and user behavior data includes: Retrieve the real-time temperature and user behavior data; compare the real-time temperature with the temperature threshold; among them, the temperature threshold is set according to the user's behavior data; the temperature threshold includes: low temperature threshold and high temperature threshold. When the real-time temperature is greater than or equal to the high temperature threshold or the real-time temperature is less than or equal to the low temperature threshold, correct the predicted energy consumption data according to the real-time temperature. Obtain the instruction data of the user; among them, the instruction data includes: the operating state of the electrical appliance, the device parameters of the electrical appliance, and the operating time; when the operating state of the electrical appliance is on, calculate the operating energy consumption data according to the device parameters and the operating time of the electrical appliance, and calculate the sum of the operating energy consumption data and the predicted energy consumption data to obtain the corrected energy consumption data.

6. The method for predicting and optimizing control of home energy consumption based on an intelligent router according to claim 5, characterized in that, The correction of the predicted energy consumption data according to the real-time temperature includes: Retrieve the real-time temperature, and calculate the difference between the real-time temperature and the temperature threshold; analyze the sensitivity coefficient between the energy consumption data and the real-time temperature according to the energy consumption data. Calculate the corrected energy consumption data through the difference, the sensitivity coefficient, and the mapping relationship between the predicted energy consumption data and the corrected energy consumption data.

7. The method for predicting and optimizing the control of household energy consumption based on an intelligent router according to claim 6, wherein The analysis of the sensitivity coefficient between the energy consumption data and the real-time temperature according to the energy consumption data includes: The energy consumption analysis model is constructed as follows: Among them, NH represents energy consumption data; MG represents the sensitivity coefficient; SW represents the real-time temperature; GT represents the high-temperature threshold; DT represents the low-temperature threshold; β1 and β2 are both the reference values of energy consumption data; Retrieve the energy consumption data and the corresponding historical temperature data; use the energy consumption data and the historical temperature data to train the energy consumption analysis model, and calculate the corresponding sensitivity coefficient.

8. The home energy consumption prediction and optimization control method based on an intelligent router according to claim 1, characterized in that The adjustment of the parameters of the electrical appliance according to the real-time electricity price information and the corrected energy consumption data includes: Retrieve the real-time electricity price information and the corrected energy consumption data; divide the set time period into a normal price area and a special price area according to the real-time electricity price information; when the corrected energy consumption data is greater than the energy consumption threshold, adjust the parameters of the electrical appliance; otherwise, keep the parameters of the electrical appliance running. Obtain the unit energy consumption and the operating time of the parameters of the electrical appliance; compare the operating time with the divided areas of the set time period to obtain the first-stage operating time and the second-stage operating time; calculate the total energy consumption corresponding to several parameters in each stage of operating time according to the unit energy consumption and the operating time of the electrical appliance; calculate the electricity price of several parameters in each stage of operating time according to the electricity price information; screen out the adjustment parameters from the corresponding parameters according to the electricity price.

9. The method for predicting and optimizing the control of household energy consumption based on an intelligent router according to claim 8, wherein The screening of the adjustment parameters from the corresponding parameters according to the electricity price includes: Retrieve the electricity prices of several parameters in each stage of operating time; divide the electricity prices according to the operating time and number them YDij; where i represents the i-th stage of operating time, i = 1, 2; j represents the j-th parameter; j = 1, 2,..., n, n is a positive integer representing the number of parameters. Obtain the parameter range set by the user for each stage of operating time, and eliminate the electricity prices of each stage of operating time that are not within the parameter range; obtain the corresponding total electricity cost by calculating the sum of the electricity prices of the two stages of operating time; sort the corresponding parameters in ascending order of the total electricity cost to obtain a parameter screening table; select the parameter with the lowest electricity cost as the adjustment parameter.

10. A method for predicting and optimizing the control of household energy consumption based on an intelligent router according to claim 1, characterized in that, The adjustment of the parameters of the electrical appliance according to the confirmation result of the user includes: Retrieve the confirmation result of the user; among them, the confirmation result of the user includes: confirm execution and not execute; when the confirmation result of the user is confirm execution, adjust the parameters of the electrical appliance according to the selected adjustment parameters. When the confirmation result of the user is not execute, obtain the adjustment information of the user; if the user sends an adjustment instruction, adjust the parameters of the electrical appliance according to the adjustment instruction of the user; otherwise, keep the parameters of the electrical appliance running.

Images