Household energy storage and smart home collaborative scheduling system

Abstract

The invention provides a household energy storage and smart home collaborative scheduling system, and belongs to the field of collaborative scheduling. The problem of low household power resource scheduling efficiency is solved; the energy storage planning module is used for acquiring historical electricity consumption data, predicting expected electricity consumption and formulating a charging scheme of energy storage equipment; the smart home scheduling module is used for predicting the use time and the power consumption of the adjustable load; judging whether the electric quantity stored by the energy storage equipment is sufficient or not; if yes, the working time of the adjustable load is adjusted according to the time-of-use electricity price of the user region; if not, reformulating the charging scheme of the energy storage equipment; the scheduling updating module is used for monitoring electricity price change and actual electricity consumption of the user in real time and updating a charging scheme or working time; by analyzing the power consumption of the user in each time period and combining the time-of-use electricity price, the working time and the working state are selected for the energy storage equipment and the smart home, and the power consumption cost of the user is reduced.

Description

Technical Field

[0001] This invention relates to a residential energy storage and smart home collaborative scheduling system, belonging to the field of collaborative scheduling. Background Technology

[0002] Existing systems for residential energy storage and smart home management have the following shortcomings: Lack of energy storage charging planning: Some dispatching systems only rely on simple experience or fixed schedules to arrange the charging of energy storage devices without fully considering the changes in users' actual electricity demand and the dynamic differences in electricity prices. They may charge energy storage devices during periods of high electricity prices or overcharge them during periods of low user electricity demand, resulting in the energy stored in the energy storage devices not being effectively utilized at the appropriate time.

[0003] The scheduling of smart home devices lacks rationality: some scheduling systems simply operate according to the mode manually set by the user, without dynamically adjusting based on the power status of energy storage devices and electricity prices; when multiple high-power smart home devices are running simultaneously, it may cause excessive instantaneous load on the power grid, which not only affects the normal operation of the devices, but may also damage the home circuit.

[0004] Lack of optimized energy allocation: The lack of coordination between some energy storage devices and smart home devices may lead to energy storage devices failing to prioritize power supply to the home when there is sufficient power, and continuing to draw power from the grid; or when the energy storage capacity is insufficient, there is no reasonable arrangement for charging and device use, resulting in ineffective energy consumption. Summary of the Invention

[0005] In view of the shortcomings of existing technologies, the purpose of this invention is to provide a household energy storage and smart home collaborative scheduling system, which aims to solve the problem of low efficiency in household power resource scheduling.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a residential energy storage and smart home collaborative scheduling system comprising: Energy storage planning module: used to acquire users' historical electricity consumption data, predict users' expected electricity consumption in different time periods, and then, in combination with the time-of-use electricity price in the user's region, the working mode of the energy storage equipment, and weather conditions, formulate a charging plan for the energy storage equipment. Smart home scheduling module: used to obtain the user's historical operation behavior of adjustable loads, predict the user's usage time and power consumption of adjustable loads in future time periods; based on the actual power consumption of fixed loads and the user's cumulative power consumption, determine whether the energy storage device has sufficient power; if sufficient, adjust the working time of adjustable loads according to the time-of-use electricity price in the user's area; if insufficient, revise the charging plan for energy storage device. The scheduling update module is used to monitor changes in electricity prices at the user's location and the user's actual hourly electricity consumption in real time, and to update existing energy storage device charging schemes or adjustable load working times.

[0007] Furthermore, the process for developing a charging solution is as follows: The SARIMA model is used to perform time series analysis on historical electricity consumption data to predict users' expected electricity consumption at different times of the day and at different times of the next day. Calculate the total expected electricity consumption e for each time period of the day. （tod） The total expected electricity consumption for each time period of the next day, e （tom） ; Get the user's cumulative electricity consumption for the day （con） And combined with the existing electricity stored in the energy storage device e （dev） Determine whether the energy storage device has sufficient stored power and formulate different charging schemes for the energy storage device; Scenario 1: If the energy storage device has sufficient stored electricity, compared to and Based on the size of the energy storage device, a charging solution is developed. like If so, then do not process it; like Then determine whether the energy storage device is connected to an external photovoltaic panel and formulate the working mode of the energy storage device; Scenario 2: If the energy storage device does not have enough stored electricity, the energy storage device needs to be fully charged from ti to 24:00.

[0008] Furthermore, the workflow for scenario one is as follows: Scenario a1: If the energy storage device is not connected to an external photovoltaic panel, the charging amount of the energy storage device at each time will be adjusted according to the time-of-use electricity price and power supply efficiency at the user's location. Obtain the expected electricity consumption from time ti to 24:00, and calculate the rechargeable capacity re of the energy storage device from time ti to 24:00. （ti） ~re （24） ; Determine re （ti） ~re （24） The sum is greater than or equal to Does it meet the requirements? If satisfied, obtain the electricity price at the user's location from time ti to 24:00, and calculate the additional electricity cost consumed by the energy storage device from time ti to 24: ec （ti） ~ec （24） ; re （ti） ~re （24） Press EC （ti） ~ec （24） Arranged in ascending order, we get: rc（1） ~rc （td） Where td represents the time difference between time ti and 24:00; Calculate rc （1） ~rc （td） The sum of the first z terms, such that rc （1） ~rc （td） The sum of the first z terms ≥ ; in turn in rc （1） ~rc （z） The corresponding time period is for replenishing the energy storage device with RC. （1） ~rc （z） The amount of electricity, non-RC （1） ~rc （z） The corresponding time period will not be processed; If the conditions are not met, then obtain the electricity price for the user's location from 1:00 AM to 12:00 AM and recalculate rc. （1） ~rc （z） The steps to replenish the energy storage device within one day （tom） The amount of electricity; Scenario a2: If the energy storage device is connected to an external photovoltaic panel, the working status of the photovoltaic panel will be adjusted according to the weather conditions.

[0009] Furthermore, the workflow for scenario a2 is as follows: Obtain the solar declination angle δ （sun） and solar hour angle ω （sun） Calculate the solar altitude angle s （sun） And solar azimuth angle a （sun） ; Adjust the angle of the photovoltaic panels according to the actual weather conditions: If the weather is sunny, adjust the tilt angle of the photovoltaic panels to: s （sun） The complementary angle is used to adjust the azimuth angle of the photovoltaic panel to: a （sun） ; If the weather is normal, adjust the tilt angle of the photovoltaic panels to: latitude φ of the user's location. （geo） According to φ （geo） ω （sun） and δ （sun） Calculate the azimuth angle γ （azi） And adjust the azimuth angle of the photovoltaic panels to γ （azi） ; Based on the user's local historical weather records, the solar intensity for the remaining sunshine hours is estimated. Then, based on the actual power generation of the photovoltaic panels, the expected power generation of the photovoltaic panels during the remaining sunshine hours is estimated, and it is determined whether the expected power generation is greater than or equal to e. （tom） ; If so, then do not process it; If not, then calculate e. （tom）The difference between the expected power generation and the expected power generation is taken as the surplus power generation. The processing steps for use case a1 are to supplement the surplus power generation of the energy storage device. If the weather is severe, the photovoltaic panels will be shut down, and the procedure for use case a1 will be to replenish the power to the energy storage device.

[0010] Furthermore, the workflow for scenario two is as follows: Obtain the electricity price from time ti to 24:00, and organize the matrix p. （r） ; Let the charging power of the energy storage device from time ti to 24 hours be: po （ti） ~po （24） And organize it into a power matrix P （o） ; Obtain the hourly charging time t of the energy storage device （re） Define constraint 1 and construct the objective function J; Obtain the maximum rated power Po of the energy storage device during charging. （max） Construct constraint 2; Introduce a barrier parameter u greater than zero （ob） Secondary construction of the objective function J （u） ; Introducing Lagrange multipliers and upper and lower bound constraints, based on the objective function J... （u） Construct the Lagrangian function L; Define the iterative process: [The process involves] setting po... （ti） ~po （24） The initial value is set to Set the convergence threshold ε; Initialize u （ob） The value of v is calculated when the partial derivative of function L is zero, representing the upper and lower bound constraint values. （t） and w （t） and matrix P （o） ; According to v （t） and w （t） The value of u decreases in the opposite direction. （ob） The value; During the iteration process, obtain the power matrix P for any two consecutive iterations. （o1） and P （o2） ; Calculate matrix P （o1） Sum of matrix P （o2） The function value J corresponding to the objective function J （o1） and J （o2） ; If J （o1） With J （o2） The difference is less than ε, or matrix P （o2） For the first time, more than po appeared in China （ti） ~po （24）If the upper and lower bounds are not found, the iteration stops, and matrix P is modified. （o1） The value in the value is used as the charging power of the energy storage device.

[0011] Furthermore, the workflow for predicting the usage time and power consumption of adjustable loads is as follows: The names and types of various adjustable loads are numbered, the usage time of adjustable loads is used as time characteristics, the name, type and power of adjustable loads are used as load characteristics, the outdoor environmental information when using adjustable loads is used as environmental characteristics, and the historical operation behavior of users using various adjustable loads is transformed into three-dimensional time series samples of users using various adjustable loads. Analyze the probability distribution of user usage patterns for different adjustable loads at different time periods; Select a prediction model, divide the 3D time series samples into training and test sets proportionally, and train the prediction model: In the first stage of model training, time features and environmental features are used as inputs, and the probability of a user using a certain adjustable load during a certain time period is output based on the outdoor environment. In the second stage of model training, time features, load features and environmental features are used as inputs to output the usage duration and power of a user using an adjustable load at a certain time in the future. During model training, the prediction results generated by the model based on the training set and the real results in the test set are obtained. The prediction error between the prediction results and the real results is calculated, and the weights of the adaptive features of the model for time features, load features and environmental features are updated until the model reaches the maximum number of iterations. Based on the user's outdoor weather on that day, a predictive model is used to predict the usage time and electricity consumption of each adjustable load from time ti to 24:00.

[0012] Furthermore, the procedure for adjusting the adjustable load working time is as follows: Obtain the power consumption ue of the first to the second adjustable loads. （1） ~ue （ad） ; Calculate the user's electricity consumption (lo) of the fixed load from time ti to 24 hours based on the actual power consumption of the fixed load. （ti） ~lo （24） ; Get the user's expected electricity consumption ex from time ti to 24 hours. （ti） ~ex （24） ; Use ex （ti） ~ex （24） Subtract lo in sequence （ti） ~lo （24） The allocable electricity ca from time ti to 24 hours is obtained. （ti） ~ca（24） ; Compare the stored capacity of energy storage devices with The size of the energy storage device determines whether it has sufficient electrical energy. If sufficient, obtain the time-of-use electricity price (PR) for the user's location from ti to 24:00. （ti） ~pr （24） ; ca （ti） ~ca （24） Press pr （ti） ~pr （24） Arranged in ascending order, we get aa （1） ~aa （td） Where td represents the time difference from time ti to 24:00, and is aa （1） Adjustable loads are allocated to the corresponding time periods; If the amount is insufficient, calculate ca. （ti） ~ca （24） The sum of the charges and the additional charging capacity of the energy storage device are used to revise the charging scheme for the energy storage device.

[0013] Furthermore, for aa （1） The process for allocating adjustable load to the corresponding time period is as follows: Set a binary variable s （r） s （r） Indicates whether the r-th adjustable load is assigned to aa （1） The corresponding time period, 1 indicates allocation, 0 indicates no allocation; Define the objective function M, and let ue （1） ~ue （ad） Arranged in descending order, the sequence se is obtained. （1） ~se （ad） ; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The front b （u） Item and ma （1） , which serves as the upper bound of the objective function; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The after c （u） Item and ma （2） , which serves as a lower bound for the objective function; If ma （1） with ma （2） If they are equal, then ma （2） Change to: sequence se （1） ~se （ad） After (c) （u） -1) terms and; According to ma （1） with ma （2） Define constraints; Based on the constraints, the adjustable load that maximizes the objective function M is used as the alternative load.

[0014] Furthermore, for aa （1） The subsequent process for allocating adjustable load to the corresponding time period is as follows: Obtain the usage time corresponding to each alternative load; In the historical operation behavior of users using various adjustable loads, the usage time with a basic usage probability greater than 0.5 corresponding to each alternative load is extracted and used as the reference time for each alternative load; If the usage time of a candidate load falls within the reference time of that load, then the load is retained; if the usage time of a candidate load does not fall within the reference time of that load, then the load is removed. The selected alternative loads are compiled and used as the final loads; Calculate and determine the total power consumption taa of the load, and then use aa （1） The value becomes Continue selecting and choosing adjustable loads from the other adjustable loads; Adjust the usage time corresponding to the determined load to aa （1） The corresponding time period.

[0015] Compared with the prior art, the beneficial effects of the present invention are: Precise planning of energy storage and charging: This invention comprehensively integrates key factors such as time-of-use electricity price information of the user's region, the working characteristics of the energy storage device itself, and real-time weather conditions. During periods of low electricity prices, the energy storage device stores sufficient electrical energy, while during peak electricity prices, the charging activity of the energy storage device is reduced. In cases where cloudy or rainy weather may affect photovoltaic power generation (if there is supporting infrastructure) and thus affect energy storage charging, the charging strategy is adjusted in advance, thereby reducing the overall electricity cost of the household and saving expenses.

[0016] Optimized Smart Home Scheduling: This invention analyzes the historical operational behavior of users using adjustable loads, covering multi-dimensional data such as usage time, duration, power, and outdoor environmental information. Based on this data, it predicts the specific usage of adjustable loads by users in different time periods in the future. Then, based on the actual power usage of fixed loads and the user's cumulative electricity consumption, it flexibly adjusts the working time of adjustable loads according to time-of-use pricing, avoiding high electricity bills due to excessive electricity consumption.

[0017] Ensuring electrical safety: This invention continuously monitors the power and operating power of the energy storage device, and replenishes it in a timely manner when the energy storage is about to run out, or adjusts the charging scheme of the energy storage device, so as to avoid safety hazards such as equipment failure, circuit damage or even fire caused by unreasonable use of electricity. Attached Figure Description

[0018] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the system of the present invention; Figure 2 This is a schematic diagram illustrating the scenario of the present invention; Figure 3 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] Please see Figure 1 and Figure 3 The residential energy storage and smart home collaborative scheduling system includes: Energy storage planning module: used to acquire users' historical electricity consumption data, predict users' expected electricity consumption in different time periods, and then, in combination with the time-of-use electricity price in the user's region, the working mode of the energy storage equipment, and weather conditions, formulate a charging plan for the energy storage equipment. The workflow of the energy storage planning module is as follows: The system obtains the user's daily and hourly electricity consumption data for the past year as the user's historical electricity consumption data; it uses the SARIMA model to perform time series analysis on the historical electricity consumption data to predict the user's expected electricity consumption in each time period of the day and the expected electricity consumption in each time period of the next day (in this invention, each hour is a time period). Calculate the total expected electricity consumption e of the user in each time period of the day. （tod） ; Calculate the total expected electricity consumption e of the user in each time period of the next day. （tom） ; Get the current time ti and the user's cumulative electricity consumption e for the day. （con） And combined with the existing electricity stored in the energy storage device e （dev） Determine whether the energy storage device has sufficient stored power and formulate different charging schemes for the energy storage device; Scenario 1: If This indicates that the energy storage device has sufficient stored electricity, which is relatively... and Based on the size of the energy storage device, a charging solution is developed. like If so, then do not process it; like Then, determine whether the energy storage device is connected to an external photovoltaic panel and determine the operating mode of the energy storage device: Scenario a1: If the energy storage device is not connected to an external photovoltaic panel, the charging amount of the energy storage device at each time will be adjusted according to the time-of-use electricity price and the power supply efficiency of the (power supply system) at the user's location. Get the user's expected electricity consumption from ti to 24:00 on the same day: qe （ti） ~qe （24） ; Obtain the power supply from time ti to 24:00 at the user's location (power supply system), and calculate the amount of electricity the user can receive during that time period: el （ti） ~el （24） ; Use el （ti） ~el （24） Subtract qe in sequence （ti） ~qe （24） The rechargeable capacity of the energy storage device from time ti to 24 hours is obtained. （ti） ~re （24） ; Determine re （ti） ~re （24） The sum is greater than or equal to Does it meet the requirements? If satisfied, obtain the electricity price for the user's location from time ti to 24:00, and then combine it with re （ti） ~re （24） Calculate the additional electricity cost consumed by the energy storage device from time ti to 24:00 (i.e., the electricity price at the user's location from time ti to 24:00, multiplied by "re"). （ti） ~re （24） "Get: Additional electricity cost consumed by the energy storage device from ti to 24:00): ec" （ti） ~ec （24） ; re （ti） ~re （24） Press EC （ti） ~ec （24） Arranged in ascending order, we get: rc （1） ~rc （td） Where td represents the time difference between time ti and 24:00; Calculate rc （1） ~rc （td） The sum of the first z terms, such that rc （1） ~rc （td） The sum of the first z terms ≥ ; in turn in rc （1） ~rc （z） The corresponding time period is for replenishing the energy storage device with RC. （1）~rc （z） The amount of electricity, non-RC （1） ~rc （z） The corresponding time period is not processed; where the value of z ranges from 1 to td. If the conditions are not met, then obtain the electricity price for the user's location from 1:00 AM to 12:00 AM and recalculate rc. （1） ~rc （z） The steps to replenish the energy storage device within one day （tom） The amount of electricity; Scenario a2: If the energy storage device is connected to an external photovoltaic panel, the operating status of the photovoltaic panel should be adjusted according to the weather conditions. Obtain the solar declination angle δ （sun） and solar hour angle ω （sun） Calculate the solar altitude angle s （sun） And solar azimuth angle a （sun） ; Adjust the angle of the photovoltaic panels according to the actual weather conditions: If the weather is sunny, adjust the tilt angle of the photovoltaic panels to: s （sun） The complementary angle (i.e., the tilt angle of the photovoltaic panel = 90°) o -s （sun） ), so that the surface of the photovoltaic panel is perpendicular to the sunlight; adjust the azimuth angle of the photovoltaic panel to: a （sun） This ensures that the photovoltaic panels are oriented in the same direction as the sun; If the weather is average (i.e., the local weather is cloudy or overcast with no rain or snow), adjust the tilt angle of the photovoltaic panels to: the latitude φ of the user's location. （geo） Adjust the azimuth angle of the photovoltaic panel to γ （azi） : ; After adjusting the angle of the photovoltaic panels, the solar intensity for the remaining sunshine hours in the user's local area is estimated based on historical weather records. Then, based on the actual power generation of the photovoltaic panels, the expected power generation of the panels during the remaining sunshine hours is estimated, and it is determined whether the expected power generation is greater than or equal to e. （tom） ; If yes, then do nothing; if not, then calculate e. （tom） The difference between the expected power generation and the expected power generation is taken as the surplus power generation. The processing steps for use case a1 are to supplement the surplus power generation of the energy storage device. If the weather is severe (i.e., the local weather for the user is rain or snow), the photovoltaic panels will be shut down, and the procedure for use case a1 will be to replenish the power to the energy storage device. Scenario 2: If If the energy storage device is not charged enough, it means that the energy storage device needs to be fully charged from ti to 24:00. Get the electricity price for the user's location from time ti to 24: pr （ti） ~pr （24） And rearrange it into a (td×1) matrix p （r） ; Let the charging power of the energy storage device from time ti to 24 hours be: po （ti） ~po （24） And rearranged into a (td×1) power matrix P （o） ; Obtain the hourly charging time t of the energy storage device （re） Define constraint 1: Among them, po （v） This represents the charging power of the energy storage device at time v; the value of v ranges from ti to 24. Construct the objective function J: ;in, Represents the identity matrix of (td×td) (i.e. In the matrix, only the elements on the main diagonal are 1, and all other elements are zero. Obtain the maximum rated power Po of the energy storage device during charging. （max） Construct inequalities As constraint 2 (constraint 2 is also po) （ti） ~po （24） (upper and lower bounds); Introduce a barrier parameter u greater than zero （ob） Secondary construction of the objective function J （u） : ; Introducing Lagrange multipliers λ and upper and lower bound constraints v for inequalities （t） and w （t） Construct the Lagrange function L: ; Define po （ti） ~po （24） Iterative process: po （ti） ~po （24） The initial values ​​are all set to And set the convergence threshold ε; Initialize u （ob） Find the value of the partial derivative of the Lagrange function L and set the partial derivative to zero. Then calculate the upper and lower bound constraint values ​​v when the partial derivative of the Lagrange function L is zero. （t） and w （t） and matrix P （o） ; According to v（t） and w （t） The value of u decreases in the opposite direction. （ob） The value of makes the median of the power matrix gradually approach the value of . （ti） ~po （24） The upper and lower bounds; During the iteration process, obtain the power matrix P for any two consecutive iterations. （o1） and P （o2） ; Calculate matrix P （o1） Sum of matrix P （o2） The function value J corresponding to the objective function J （o1） and J （o2） ; If J （o1） With J （o2） The difference is less than ε, or matrix P （o2） For the first time, more than po appeared in China （ti） ~po （24） If the upper and lower bounds are not found, the iteration stops, and matrix P is modified. （o1） The value in the table represents the hourly charging power of the energy storage device from time ti to 24:00.

[0021] The smart home scheduling module is used to acquire historical user behavior data for adjustable loads (including usage time, duration, power consumption, and outdoor environmental information for each adjustable load). It then predicts the user's usage time and electricity consumption for adjustable loads in future time periods. Based on the actual power consumption of fixed loads and the user's cumulative electricity consumption, it determines whether the energy storage device has sufficient power. If sufficient, it adjusts the operating time of the adjustable loads according to the user's local time-of-use electricity price. If insufficient, it re-determines the charging plan for the energy storage device. This invention categorizes smart home systems into: adjustable load and fixed load; Adjustable load refers to smart home devices that do not need to work continuously for 24 hours and whose working hours can be freely adjusted, such as smart washing machines, robot vacuums, and smart air conditioners. Fixed load refers to smart home devices that need to operate 24 hours a day without interruption, such as smart refrigerators, smart door locks, and smart surveillance systems. Please see Figure 2 The specific process for predicting the usage time and operating power of adjustable loads in different time periods in the future is as follows: The system obtains the user's usage time, duration, and power consumption for various adjustable loads over the past month, thus revealing the user's historical operational behavior for each adjustable load. The names and types of various adjustable loads are numbered (the numbers are in numerical form). The usage time of the adjustable loads is used as the time feature, the name, type and power of the adjustable loads are used as the load feature, and the outdoor environmental information when using the adjustable loads is used as the environmental feature. The historical operation behavior of users using various adjustable loads is transformed into three-dimensional time series samples of users using various adjustable loads (i.e., time features, load features and environmental features). Using statistical learning and unsupervised analysis, we analyze the probability distribution of user usage patterns for different adjustable loads at different time periods: Statistically analyze the historical usage frequency of each adjustable load in each time period, and divide by the total number of days to obtain the basic usage probability of each adjustable load in each time period (e.g., the usage probability of a smart washing machine at 3 PM = the total number of times the smart washing machine was used at 3 PM in the past year / the total number of days in the year); design sliding windows with a time length of weeks or months to identify the periodic patterns of user usage of various adjustable loads; use the Apriori algorithm or FP-Growth algorithm to analyze the usage duration and power consumption of users when using adjustable loads in different time periods, and identify the usage duration and power consumption patterns of users when using various adjustable loads; use the K-Means or DBSCAN algorithm to analyze the environmental information of users when using various adjustable loads, and identify the outdoor environmental patterns of users when using various adjustable loads. Choose a suitable machine learning model (such as Gradient Boosting Tree (GBDT) or Neural Network (NN)) to define the outdoor environmental patterns of users using various adjustable loads as usage rules. Check one by one whether the usage conditions corresponding to various adjustable loads meet the established rules and correct the basic usage probability of various adjustable loads in different outdoor environments. For example: Basic usage probability: The basic usage probability of a smart washing machine at 14:00 is 50%; Usage rules: If it is a sunny day outdoors and the user is at home, the probability of the smart washing machine being used at 2 PM increases by 20%. Outdoor conditions: Sunny day, and the user is at home; The corrected probability is: 50% + 20% = 70%; Select a prediction model (such as a time series model LSTM or a machine learning model gradient boosting tree), and divide the three-dimensional time series samples of users using various adjustable loads into training and test sets according to the proportions to train the prediction model. In the first stage of model training, time features and environmental features are used as inputs to output the probability of a user using a certain adjustable load in a certain time period based on the outdoor environment. In the second stage of model training, time features, load features, and environmental features are used as inputs to output the usage duration and power of a user using a certain adjustable load in a future time period. During the model training process, obtain the prediction results generated by the model based on the training set and the real results in the test set, calculate the prediction error between the prediction results and the real results, and update the weights of the adaptive features of time features, load features, and environmental features until the model reaches the maximum number of iterations (preset by the user). Based on the user's outdoor weather on that day, a predictive model is used to predict the usage time, duration, and power consumption of each adjustable load from time ti to 24:00. Calculate the power consumption of each adjustable load based on its usage duration and power consumption. Define the operating state of each adjustable load using binary tuples [tt] （1） ue （1） ]~[tt （ad） ue （ad） ]; where ad is the number of adjustable loads; tt （1） ~tt （ad） Indicates the usage time of the first to the second adjustable load, ue （1） ~ue （ad） This represents the electricity consumption of the first to the ath adjustable loads; Calculate the user's electricity consumption (lo) of the fixed load from time ti to 24 hours based on the actual power consumption of the fixed load. （ti） ~lo （24） ; Get the user's expected electricity consumption ex from time ti to 24 hours. （ti） ~ex （24） ; Use ex （ti） ~ex （24） Subtract lo in sequence （ti） ~lo （24） The allocable electricity ca from time ti to 24 hours is obtained. （ti） ~ca （24） ; Compare the stored capacity of energy storage devices with The size of the energy storage device determines whether it has sufficient electrical energy. If the amount of electricity stored in the energy storage device is greater than or equal to This indicates that the energy storage device has sufficient electrical energy stored. Get the time-of-use electricity price (PR) for the user's location from time ti to 24:00. （ti） ~pr （24） ; ca （ti） ~ca （24） Press pr （ti） ~pr （24） Arranged in ascending order, we get aa （1） ~aa （td） Where td represents the time difference from ti to 24:00; for aa （1） Adjustable load allocation for the corresponding time period: Set a binary variable s （r） s （r） Indicates whether the r-th adjustable load is assigned to aa （1） The corresponding time period; s （r） ∈{0,1}, where 1 represents allocation and 0 represents unallocated; where the range of r is 1 to ad; Define the objective function M: Among them, ue （r） This represents the electricity consumption of the r-th adjustable load; ue （1） ~ue （ad） Arranged in descending order, the sequence se is obtained. （1） ~se （ad） ; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The front b （u） Item and ma （1） (i.e. ma （1） It is a sequence se （1） ~se （ad） Add from left to right), add ma （1） This serves as an upper bound for the selectable adjustable load in the objective function; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The after c （u） Item and ma （2） (i.e. ma （2） It is a sequence se （1） ~se （ad） Add from right to left), add ma （2） As a lower bound for the objective function, the adjustable load can be selected; If ma （1） with ma （2） If they are equal, then ma （2） Change to: sequence se （1） ~se（ad） After (c) （u） -1) terms and; According to ma （1） with ma （2） Define constraints: ; Based on the constraints, use an optimization solver (such as Gurobi or CPLEX) to find the adjustable load that maximizes the objective function M, and use it as the alternative load. In tt （1） ~tt （ad） In the process, obtain the usage time corresponding to each alternative load; In the historical operation behavior of users using various adjustable loads, the usage time with a basic usage probability greater than 0.5 corresponding to each alternative load is extracted and used as the reference time for each alternative load; If the usage time of a candidate load falls within the reference time of that load, then the load is retained; if the usage time of a candidate load does not fall within the reference time of that load, then the load is removed. The selected alternative loads are compiled and used as the final loads; Calculate and determine the total power consumption taa of the load, and then use aa （1） The value becomes Continue selecting and choosing adjustable loads from the other adjustable loads (i.e., removing the "alternative loads that have been removed" and the "retained alternative loads"); Adjust the usage time corresponding to the determined load to aa （1） The corresponding time period; Repeat as aa （1） The same process applies to allocating adjustable loads for time periods corresponding to other allocable power supplies: If the amount of electricity stored in the energy storage device is less than This indicates that the energy storage device does not store enough electricity, and the calculation of ca... （ti） ~ca （24） The sum of the charges is used as additional charging capacity for the energy storage device. Using the processing steps of scenario one or scenario two in the above-mentioned energy storage planning module, the charging scheme for the energy storage device is re-formulated.

[0022] The scheduling update module is used to monitor changes in electricity prices at the user's location and the user's actual hourly electricity consumption in real time, and to update existing energy storage device charging schemes or adjustable load working times.

[0023] The above formulas are all dimensionless calculations. The formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. For example, there are weighting coefficients and proportional coefficients. The values ​​set are to quantify each parameter to obtain a specific value, which is convenient for subsequent comparison. The values ​​of the weighting coefficients and proportional coefficients are only required to not affect the proportional relationship between the parameters and the quantified values.

[0024] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of protection of the claims.

Claims

1. A residential energy storage and smart home collaborative scheduling system, characterized in that, The system includes: Energy storage planning module: used to acquire users' historical electricity consumption data, predict users' expected electricity consumption in different time periods, and then, in combination with the time-of-use electricity price in the user's region, the working mode of the energy storage equipment, and weather conditions, formulate a charging plan for the energy storage equipment. Smart home scheduling module: used to obtain the user's historical operation behavior of adjustable loads, predict the user's usage time and power consumption of adjustable loads in future time periods; based on the actual power consumption of fixed loads and the user's cumulative power consumption, determine whether the energy storage device has sufficient power; if sufficient, adjust the working time of adjustable loads according to the time-of-use electricity price in the user's area; if insufficient, revise the charging plan for energy storage device. The scheduling update module is used to monitor changes in electricity prices and actual electricity consumption in real time, and to update existing energy storage device charging schemes or adjustable load working times.

2. The household energy storage and smart home collaborative scheduling system according to claim 1, characterized in that, The process for developing a charging solution is as follows: Time series analysis of historical electricity consumption data is used to predict users' expected electricity consumption at different times of the day and at different times of the next day. Calculate the total expected electricity consumption e for each time period of the day. （tod） The total expected electricity consumption for each time period of the next day, e （tom） ; Get the user's cumulative electricity consumption for the day （con） And combined with the existing electricity stored in the energy storage device e （dev） Determine whether the energy storage device has sufficient stored power and formulate different charging schemes for the energy storage device; Scenario 1: If the energy storage device has sufficient stored electricity, compared to and Based on the size of the energy storage device, a charging solution is developed. like If so, then no action will be taken; like Then determine whether the energy storage device is connected to an external photovoltaic panel and formulate the working mode of the energy storage device; Scenario 2: If the energy storage device does not have enough stored electricity, the energy storage device needs to be fully charged from ti to 24:

00.

3. The household energy storage and smart home collaborative scheduling system according to claim 2, characterized in that, The workflow for scenario one is as follows: Scenario a1: If the energy storage device is not connected to an external photovoltaic panel, the charging amount of the energy storage device at each time will be adjusted according to the time-of-use electricity price and power supply efficiency at the user's location. Obtain the expected electricity consumption from time ti to 24:00, and calculate the rechargeable capacity re of the energy storage device from time ti to 24:

00. （ti） ~re （24） ; Determine re （ti） ~re （24） The sum is greater than or equal to Does it meet the requirements? If satisfied, obtain the electricity price at the user's location from time ti to 24:00, and calculate the additional electricity cost consumed by the energy storage device: ec （ti） ~ec （24） ; re （ti） ~re （24） Press EC （ti） ~ec （24） Arranged in ascending order, we get: rc （1） ~rc （td） Where td represents the time difference between time ti and 24:00; Calculate rc （1） ~rc （td） The sum of the first z terms, such that rc （1） ~rc （td） The sum of the first z terms ≥ ; in turn in rc （1） ~rc （z） The corresponding time period is for replenishing the energy storage device with RC. （1） ~rc （z） The amount of electricity, non-RC （1） ~rc （z） The corresponding time period will not be processed; If the conditions are not met, then obtain the electricity price for the user's location from 1:00 AM to 12:00 AM and recalculate rc. （1） ~rc （z） The steps to replenish the energy storage device within one day （tom） The amount of electricity; Scenario a2: If the energy storage device is connected to an external photovoltaic panel, the working status of the photovoltaic panel will be adjusted according to the weather conditions.

4. The household energy storage and smart home collaborative scheduling system according to claim 3, characterized in that, The workflow for scenario a2 is as follows: Obtain the solar declination angle δ （sun） and solar hour angle ω （sun） Calculate the solar altitude angle s （sun） And solar azimuth angle a （sun） ; Adjust the angle of the photovoltaic panels according to the actual weather conditions: If the weather is sunny, adjust the tilt angle of the photovoltaic panels to: s （sun） The complementary angle is used to adjust the azimuth angle of the photovoltaic panel to: a （sun） ; If the weather is normal, adjust the tilt angle of the photovoltaic panels to: latitude φ of the user's location. （geo） According to φ （geo） ω （sun） and δ （sun） Calculate the azimuth angle γ （azi） And adjust the azimuth angle of the photovoltaic panels to γ （azi） ; Based on the user's local historical weather records, the solar intensity for the remaining sunshine hours is estimated. Then, based on the actual power generation of the photovoltaic panels, the expected power generation of the photovoltaic panels during the remaining sunshine hours is estimated, and it is determined whether the expected power generation is greater than or equal to e. （tom） ; If so, then do not process it; If not, then calculate e. （tom） The difference between the expected power generation and the expected power generation is taken as the surplus power generation. The processing steps for use case a1 are to supplement the surplus power generation of the energy storage device. If the weather is severe, the photovoltaic panels will be shut down, and the procedure for use case a1 will be to replenish the power to the energy storage device.

5. The household energy storage and smart home collaborative scheduling system according to claim 2, characterized in that, The workflow for scenario two is as follows: Obtain the electricity price from time ti to 24:00, and organize the matrix p. （r） ; Let the charging power of the energy storage device from time ti to 24 hours be: po （ti） ~po （24） And organize it into a power matrix P （o） ; Obtain the hourly charging time t of the energy storage device （re） Define constraint 1 and construct the objective function J; Obtain the maximum rated power Po of the energy storage device during charging. （max） Construct constraint 2; Introduce a barrier parameter u greater than zero （ob） Secondary construction of the objective function J （u） ; Introducing Lagrange multipliers and upper and lower bound constraints, based on the objective function J... （u） Construct the Lagrangian function L; Define the iterative process: po （ti） ~po （24） The initial value is set to Set the convergence threshold ε; Initialize u （ob） The value of v is calculated when the partial derivative of function L is zero, representing the upper and lower bound constraint values. （t） and w （t） and matrix P （o） ; According to v （t） and w （t） The value of u decreases in the opposite direction. （ob） The value; During the iteration process, obtain the power matrix P for any two consecutive iterations. （o1） and P （o2） ; Calculate matrix P （o1） Sum of matrix P （o2） The function value J corresponding to the objective function J （o1） and J （o2） ; If J （o1） With J （o2） The difference is less than ε, or matrix P （o2） For the first time, more than po appeared in China （ti） ~po （24） If the upper and lower bounds are not found, the iteration stops, and matrix P is modified. （o1） The value in the value is used as the charging power of the energy storage device.

6. The household energy storage and smart home collaborative scheduling system according to claim 1, characterized in that, The workflow for predicting the usage time and power consumption of adjustable loads is as follows: The names and types of various adjustable loads are numbered, the usage time of adjustable loads is used as time characteristics, the name, type and power of adjustable loads are used as load characteristics, the outdoor environmental information when using adjustable loads is used as environmental characteristics, and the historical operation behavior of users using various adjustable loads is transformed into three-dimensional time series samples of users using various adjustable loads. Analyze the probability distribution of user usage patterns for different adjustable loads at different time periods; Select a prediction model, divide the 3D time series samples into training and test sets proportionally, and train the prediction model: In the first stage of model training, time features and environmental features are used as inputs, and the probability of a user using a certain adjustable load during a certain time period is output based on the outdoor environment. In the second stage of model training, time features, load features and environmental features are used as inputs to output the usage duration and power of a user using an adjustable load at a certain time in the future. During model training, the prediction results generated by the model based on the training set and the real results in the test set are obtained. The prediction error between the prediction results and the real results is calculated, and the weights of the adaptive features of the model for time features, load features and environmental features are updated until the model reaches the maximum number of iterations. Based on the user's outdoor weather on that day, a predictive model is used to predict the usage time and electricity consumption of each adjustable load from time ti to 24:

00.

7. The household energy storage and smart home collaborative scheduling system according to claim 1, characterized in that, The procedure for adjusting the adjustable load operating time is as follows: Obtain the power consumption ue of the first to the second adjustable loads. （1） ~ue （ad） ; Calculate the user's electricity consumption (lo) for the fixed load from time ti to 24 hours based on the actual power consumption of the fixed load. （ti） ~lo （24） ; Get the user's expected electricity consumption ex from time ti to 24 hours. （ti） ~ex （24） ; Use ex （ti） ~ex （24） Subtract lo in sequence （ti） ~lo （24） The allocable electricity ca from time ti to 24 hours is obtained. （ti） ~ca （24） ; Compare the stored capacity of energy storage devices with The size of the energy storage device determines whether it has sufficient electrical energy. If sufficient, obtain the time-of-use electricity price (PR) for the user's location from ti to 24:

00. （ti） ~pr （24） ; ca （ti） ~ca （24） Press pr （ti） ~pr （24） Arranged in ascending order, we get aa （1） ~aa （td） Where td represents the time difference from time ti to 24:00, and is aa （1） Adjustable loads are allocated to the corresponding time periods; If the amount is insufficient, calculate ca. （ti） ~ca （24） The sum of the charges and the additional charging capacity of the energy storage device are used to revise the charging scheme for the energy storage device.

8. The household energy storage and smart home collaborative scheduling system according to claim 7, characterized in that, for aa （1） The process for allocating adjustable load to the corresponding time period is as follows: Set a binary variable s （r） s （r） Indicates whether the r-th adjustable load is assigned to aa （1） The corresponding time period, 1 indicates allocation, 0 indicates no allocation; Define the objective function M, and let ue （1） ~ue （ad） Sort in descending order to obtain the sequence se （1） ~se （ad） ; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The front b （u） Item and ma （1） , which serves as the upper bound of the objective function; Calculate the sequence se （1） ~se （ad） First time greater than aa （1） The after c （u） Item and ma （2） , which serves as a lower bound for the objective function; If ma （1） with ma （2） If they are equal, then ma （2） Change to: sequence se （1） ~se （ad） After (c) （u） -1) terms and; According to ma （1） with ma （2） Define constraints; Based on the constraints, the adjustable load that maximizes the objective function M is used as the alternative load.

9. The household energy storage and smart home collaborative scheduling system according to claim 1, characterized in that, for aa （1） The subsequent process for allocating adjustable load to the corresponding time period is as follows: Obtain the usage time corresponding to each alternative load; In the historical operation behavior of users using various adjustable loads, the usage time with a basic usage probability greater than 0.5 corresponding to each alternative load is extracted and used as the reference time for each alternative load; If the usage time of a candidate load is within the reference time of that load, then that load is retained; If the usage time of a candidate load does not fall within the reference time of that load, then that load is removed. The selected alternative loads are compiled and used as the final loads; Calculate and determine the total power consumption taa of the load, and then use aa （1） The value becomes Continue selecting and choosing adjustable loads from the other adjustable loads; Adjust the usage time corresponding to the determined load to aa （1） The corresponding time period.

10. The household energy storage and smart home collaborative scheduling system according to claim 6, characterized in that, The process for analyzing the probability distribution of user usage patterns for different adjustable loads at different time periods is as follows: Count the historical usage frequency of each adjustable load in each time period, and divide by the total number of days to obtain the basic usage probability of each adjustable load in each time period; By designing sliding windows with a weekly or monthly timeframe, we can identify the cyclical patterns of users' use of various adjustable loads; analyze the usage duration and power consumption of users when using adjustable loads in different time periods to identify the usage duration and power consumption patterns of users when using various adjustable loads; and analyze the environmental information of users when using various adjustable loads to identify the outdoor environmental patterns of users when using various adjustable loads. We select a machine learning model to define the outdoor environmental patterns of users using various adjustable loads as usage rules. We then check whether the usage conditions corresponding to each adjustable load meet the established rules and adjust the basic usage probability of each adjustable load in different outdoor environments.

Images