Household energy management method based on time-of-use electricity price

Abstract

The invention discloses a household energy management method based on time-of-use electricity price, and the method comprises the following steps: S1, building a household energy system model which comprises a photovoltaic unit model, an energy storage unit model and an electric vehicle model; s2, constructing an electricity price time-of-use neural network model, obtaining historical electricity consumption and electricity price of the user per minute in N years, training the electricity price time-of-use neural network model, and obtaining an electricity price time period based on the trained electricity price time-of-use neural network model; and S3, based on the household energy system model and the electricity price time period, making an optimal control scheme of the electricity utilization cost. According to the method, one day can be divided into a plurality of time periods according to the power utilization habit and the time-of-use electricity price of the user, the operation mode with the lowest cost is selected, meanwhile, the V2H technology is considered, the electric vehicle is added in the operation mode, and charging and discharging of the electric vehicle are reasonably set on the basis of the purpose of the electric vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and more particularly to a household energy management method based on time-of-use electricity prices. Background Art

[0002] With the application of smart grids in households, home energy management systems are developing rapidly. Home energy management systems use user resources to manage electricity consumption, which can effectively improve energy utilization and promote energy conservation and emission reduction. However, with the development of electric vehicles, the current home energy management system solution comprehensively considers the comprehensive application of V2H technology and time-of-use electricity prices, but still has the problem of high electricity costs.

[0003] The publication number is CN118297631A, and the name is a time-of-use electricity price adjustment system, which belongs to the field of electric power technology. The time-of-use electricity price adjustment system includes a data processing device, an operation optimization device and a response feedback device. The data processing device is connected to the operation optimization device and the response feedback device respectively, and the operation optimization device is also connected to the response feedback device; the data processing device is used to pre-process the power load information of the power grid side device and the power consumption information of the target user obtained by the response feedback device, and obtain the corresponding comprehensive power consumption information. The data processing device is also used to send the comprehensive power consumption information to the operation optimization device; the operation optimization device is used to analyze and process the comprehensive power consumption information, and obtain the corresponding time-of-use electricity price strategy. The operation optimization device is also used to send the time-of-use electricity price strategy to the response feedback device; the response feedback device is used to obtain the power consumption information of the target user and feedback the time-of-use electricity price strategy to the target user. This can effectively improve the allocation effect of power resources.

[0004] The publication number is CN111598612A, and the name is a method for formulating time-of-use electricity prices, including: predicting the wind power interval at the time to be predicted; randomly sampling 100 groups of wind power scenarios within the wind power interval at the time to be predicted; reducing the scenarios to 10 groups of wind power scenarios; adding wind power scenarios to user loads to generate a net load scenario diagram; dividing the peak and valley periods in the net load scenario into H time periods, and then dividing the H time periods into peak, flat, and valley periods; calculating the wind power scenario expectations and generating an equivalent net load curve; constructing a time-of-use electricity price model based on the equivalent net load; and using the NSGA-II algorithm to perform multi-objective solutions to the time-of-use electricity price model. Compared with the method that does not consider the uncertainty of wind power, it fully considers the randomness of wind power and effectively guides users to respond to time-of-use electricity prices, achieving the goals of balancing system loads and shaving peaks and filling valleys, while ensuring that users' electricity usage habits are not affected. However, none of the above-mentioned public technologies can achieve accurate time-sharing of electricity prices and energy management based on time-of-use electricity prices.

[0005] Therefore, how to provide a method that fully considers V2H technology and time-of-use electricity prices, minimizes electricity costs while meeting the needs of users themselves, and achieves optimal electricity costs is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0006] In view of this, the present invention provides a household energy management method based on time-of-use electricity prices, aiming to solve the above technical problems.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A household energy management method based on time-of-use electricity price comprises the following steps:

[0009] S1. Construct a home energy system model, including a photovoltaic unit model, an energy storage unit model, and an electric vehicle model;

[0010] S2. Build a time-sharing neural network model for electricity prices, obtain the historical electricity consumption and electricity prices of the user every minute for N years, train the time-sharing neural network model for electricity prices, and obtain the electricity price period based on the trained time-sharing neural network model for electricity prices;

[0011] S3. Based on the household energy system model and electricity price period, formulate the optimal control plan for electricity cost.

[0012] Furthermore, the photovoltaic unit model is:

[0013]

[0014] A=T+(0.0256×I s (t))

[0015] Where P pv (t) is the photovoltaic output power, P pv_0 is the photovoltaic rated power, I s (t) is the solar radiation intensity, I s_ref is the solar radiation intensity under standard conditions, T ref is the reference temperature under standard conditions, δ is the temperature coefficient, and T is the ambient temperature;

[0016] The energy storage unit model includes the charging process and the discharging process, specifically:

[0017] The charging process is:

[0018]

[0019] The discharge process is:

[0020]

[0021] To restrict overcharge and overdischarge:

[0022] SOC min ≤SOC≤SOC max

[0023] In the formula, E bss (t) is the remaining power of the energy storage unit at time t, δ e is the self-discharge efficiency of the energy storage unit, P in (t) is the charging power, P ex (t) is the discharge power, η in is the charging efficiency, η ex is the discharge efficiency, is the maximum capacity of the energy storage unit, is the minimum value of the energy storage unit, is the rated charging power of the energy storage unit at time t, is the rated discharge power of the energy storage unit at time t, SOC min The minimum state of charge, SOC max is the maximum value of the state of charge;

[0024] The electric vehicle model includes the charging process and the discharging process, specifically:

[0025] Charging process:

[0026]

[0027] Discharge process:

[0028]

[0029] To restrict overcharge and overdischarge:

[0030]

[0031] In the formula, Q EV (t) is the remaining power of the electric vehicle at time t, P EV in (t) is the charging power of the electric vehicle, P EV ex (t) is the discharge power of the electric vehicle, is the charging efficiency of the electric vehicle, is the discharge efficiency of the electric vehicle, is the maximum energy of the electric vehicle, is the minimum energy of the electric vehicle, is the rated charging power of the electric vehicle at time t, is the rated discharge power of the electric vehicle at time t, is the minimum value of the electric vehicle's state of charge, is the maximum value of the electric vehicle's state of charge, and is the power demand at time t.

[0032] Furthermore, the S2 is specifically:

[0033] S21. Obtain the historical electricity consumption and electricity price of the user per minute for N years;

[0034] S22. Obtain the number of days in each quarter within N years, preprocess it to make it the same as the number of days in each quarter of the year to be predicted, and construct a training set;

[0035] S23. Construct a neural network, train the neural network with the training set to obtain a prediction network model, input a data set into the prediction network model, and obtain a prediction result;

[0036] S24. Process the prediction result to obtain peak electricity consumption periods, and perform integerization processing on the peak electricity consumption periods of each day in each quarter to obtain standard peak electricity consumption periods and normal periods;

[0037] S25. Obtain peak-valley electricity prices and corresponding standard periods, and divide the standard peak electricity consumption periods and normal periods based on the standard periods of peak-valley electricity prices to obtain electricity consumption periods.

[0038] Further, the specific content of S22 is as follows:

[0039] Judge whether the number of days in each quarter of each year in the nearly N years is equal to the number of days in each quarter of the year for which time-of-use electricity price will be implemented. If not, for the quarters with unequal numbers of days, randomly duplicate or delete within the existing quarters with unequal numbers of days so that the number of days in each quarter of each year in the N years is equal to the number of days in each quarter of the year for which time-of-use electricity price will be implemented;

[0040] Make the number of days equal by using differences, and delete the days with similar electricity consumption per minute and electricity price to make the number of days equal.

[0041] Further, the specific content of S23 is as follows:

[0042] If predicting the time-of-use electricity price for next year this year, input the historical electricity consumption and electricity price of each minute of each day in the N years as training data into the neural network to obtain a trained neural network. Use the weighted average of the N years as verification data to verify the reliability of the model. The weighted average of the electricity consumption and electricity price of each minute in the past 4 years is a*(N - 3)+b*(N - 2)+c*(N - 1)+d*N, where N is the previous year, a < b < c < d, and a + b + c + d = 1. Input the verification data into the model to obtain the predicted electricity consumption and electricity price of each minute of each day in 2024;

[0043] If predicting the time-of-use electricity price for the 2nd - 4th quarters of this year in the first quarter of this year, input the historical electricity consumption and electricity price of each minute of each day in the N years as training data into the neural network to obtain a trained neural network. Use the electricity consumption and electricity price of each minute of each day in the first quarter of this year as verification data to verify the reliability of the model. Input the verification data to obtain the predicted electricity consumption and electricity price of each minute of each day in the 2nd - 4th quarters of this year.

[0044] Furthermore, the S24 is specifically as follows:

[0045] The integral of each day of each predicted quarter is calculated M minutes apart to obtain the integral value within the corresponding interval, where the integral value of electricity consumption within the interval is greater than the preset value, which is the peak period of electricity consumption; or a power consumption threshold Q is set, and the moment when it is greater than Q is the peak period.

[0046] Furthermore, the S25 is specifically:

[0047] According to the predicted peak electricity consumption period of each day in each quarter, the union is calculated and rounded to decimal. The rounding and rounding cannot exceed the earliest time and the latest time of the peak electricity consumption period of each day in each quarter, so as to obtain the peak electricity consumption period of the quarter, and the rest of the time is the general time period;

[0048] According to the peak and normal periods of electricity consumption, and the time corresponding to the current peak and valley electricity prices where the user is located, a day is divided into multiple time periods, and there is no overlap between the peak and valley electricity prices.

[0049] Furthermore, the S3 is specifically:

[0050] S31, determining the current time period;

[0051] S32, collecting current equipment information, including but not limited to photovoltaic power generation, energy storage SOC, power consumption, and electric vehicle SOC;

[0052] S33, on the basis of meeting the power demand, taking the minimum cost as the goal, and formulating the control strategy for the current time period based on the constraint conditions;

[0053] S34, issuing control strategies and making fine adjustments to the control according to the data collected in real time;

[0054] S35. Determine whether the current time is still in the current time period. If it is still in the current time period, continue to execute the control strategy of the time period. Otherwise, end this time period and enter the next time period.

[0055] Furthermore, the minimum cost in S33 is specifically:

[0056] Construct the minimum cost objective:

[0057] minC=min(C PV +C bss +C EV -C grid )

[0058] Where C is the system cost, C PV is the photovoltaic unit cost, C bss is the energy storage unit cost, C EVis the cost of electric vehicles, C grid is the grid cost;

[0059] The specific cost of photovoltaic units is:

[0060] C PV =C o +C yun

[0061] In the formula, C PV is the photovoltaic unit cost, C o is the initial investment in photovoltaic power generation, C yun For operation and maintenance costs;

[0062] The specific cost of energy storage unit is:

[0063]

[0064] In the formula, C bss is the cost of the energy storage unit, P bss is the charging or discharging power of the energy storage unit, is the power upper limit, α bss , β bss , γ bss is the energy storage unit cost coefficient, is the discharge depth;

[0065] The specific grid costs are:

[0066]

[0067] Where ξ(t) is the limit value that cannot simultaneously purchase and sell electricity from the power grid in the same time period, and its value is 0 or 1. grid_i (t) is the power purchased from the power grid during period t, C grid_t (t) is the price of electricity purchased from the power grid during period t, P grid_o (t) is the power sold to the grid during period t, C grid_o (t) is the price of electricity sold to the grid during period t.

[0068] Furthermore, the constraints in S33 specifically include power balance, photovoltaic constraints, energy storage constraints, and electric vehicle constraints;

[0069] Power Balance:

[0070] P PV +P bss +P grid =P EV +P load

[0071] Where P PV is the photovoltaic power, P bss is the energy storage power, P gridis the grid power, P EV is the electric vehicle power, P load is the load power;

[0072] The photovoltaic constraint is:

[0073] P PV Min ≤P PV ≤P PV Max

[0074] Where P PV Min is the lower limit of photovoltaic output, P PV Max is the upper limit of photovoltaic output;

[0075] The energy storage constraint is:

[0076]

[0077]

[0078] SOC min ≤SOC≤SOC max

[0079] The electric vehicle constraints are:

[0080]

[0081]

[0082]

[0083] The above technical solution includes at least the following technical effects:

[0084] It can be known from the above technical solutions that the present invention discloses a method for household energy management based on time-of-use electricity prices. Compared with the prior art, the present invention can divide a day into several time periods according to the user's electricity usage habits and time-of-use electricity prices, select the lowest cost operation mode, and consider V2H technology at the same time, add electric vehicles to the operation mode, and reasonably set the charging and discharging of electric vehicles based on their uses. In addition, the household energy management system also takes seasonal factors into consideration, that is, in different seasons, the energy scheduling scheme is different, so as to achieve the lowest comprehensive annual electricity cost. Compared with the existing household energy management methods, the present invention fully considers V2H technology and time-of-use electricity prices in energy management, minimizes electricity costs, and achieves optimal electricity costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0086] Figure 1 The present invention is a schematic flow chart of a household energy management method based on time-of-use electricity prices. DETAILED DESCRIPTION

[0087] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] See attached Figure 1 The embodiment of the present invention discloses a household energy management method based on time-of-use electricity price, comprising the following steps:

[0089] S1. Construct a home energy system model, including a photovoltaic unit model, an energy storage unit model, and an electric vehicle model;

[0090] S2. Build a time-sharing neural network model for electricity prices, obtain the historical electricity consumption and electricity prices of the user every minute for N years, train the time-sharing neural network model for electricity prices, and obtain the electricity price period based on the trained time-sharing neural network model for electricity prices;

[0091] S3. Based on the household energy system model and electricity price period, formulate the optimal control plan for electricity cost.

[0092] To further optimize the above technical solution, the photovoltaic unit model is:

[0093]

[0094] A=T+(0.0256×I s (t))

[0095] Where P pv (t) is the photovoltaic output power, P pv_0 is the photovoltaic rated power, I s (t) is the solar radiation intensity, I s_ref is the solar radiation intensity under standard conditions, T ref is the reference temperature under standard conditions, δ is the temperature coefficient, and T is the ambient temperature;

[0096] The energy storage unit model includes the charging process and the discharging process, specifically:

[0097] The charging process is:

[0098]

[0099] The discharge process is:

[0100]

[0101] To restrict overcharge and overdischarge:

[0102] SOC min ≤SOC≤SOC max

[0103] In the formula, E bss (t) is the remaining power of the energy storage unit at time t, δ e is the self-discharge efficiency of the energy storage unit, P in (t) is the charging power, P ex (t) is the discharge power, η in is the charging efficiency, η ex is the discharge efficiency, is the maximum capacity of the energy storage unit, is the minimum value of the energy storage unit, is the rated charging power of the energy storage unit at time t, is the rated discharge power of the energy storage unit at time t, SOC min The minimum state of charge, SOC max is the maximum value of the state of charge;

[0104] The electric vehicle model includes the charging process and the discharging process, specifically:

[0105] Charging process:

[0106]

[0107] Discharge process:

[0108]

[0109] To restrict overcharge and overdischarge:

[0110]

[0111] In the formula, Q EV (t) is the remaining power of the electric vehicle at time t, P EV in (t) is the charging power of the electric vehicle, P EV ex(t) is the discharge power of the electric vehicle, is the charging efficiency of the electric vehicle, is the discharge efficiency of the electric vehicle, is the maximum energy of the electric vehicle, is the minimum energy of the electric vehicle, is the rated charging power of the electric vehicle at time t, is the rated discharge power of the electric vehicle at time t, is the minimum value of the electric vehicle's state of charge, is the maximum value of the electric vehicle's state of charge, and is the power demand at time t.

[0112] To further optimize the above technical solution, S2 is specifically:

[0113] S21. Obtain the historical electricity consumption and electricity price of the user per minute for N years;

[0114] S22, obtaining the number of days in each quarter of N years, preprocessing them to make them the same as the number of days in each quarter of the year to be predicted, and constructing a training set;

[0115] S23, constructing a neural network, training the neural network through a training set to obtain a prediction network model, inputting a data set into the prediction network model to obtain a prediction result;

[0116] S24, processing the prediction results to obtain the peak power consumption period, performing integer processing on the peak power consumption period of each day in each quarter to obtain the standard peak power consumption period and the general period;

[0117] S25. Obtain peak-valley electricity prices and corresponding standard time periods, and divide standard electricity consumption peak time periods and general time periods based on the standard time periods of peak-valley electricity prices to obtain electricity consumption time periods.

[0118] To further optimize the above technical solution, S22 is specifically:

[0119] Determine whether the number of days in each quarter of each year in the past N years is equal to the number of days in each quarter of the year in which the time-of-use electricity price will be implemented. If not, for the quarters with different numbers of days, randomly copy or delete the existing quarters with different numbers of days so that the number of days in each quarter of each year in the past N years is equal to the number of days in each quarter of the year in which the time-of-use electricity price will be implemented;

[0120] The difference is used to make the days equal, and the days with similar electricity consumption and electricity prices per minute are deleted to make the days equal.

[0121] To further optimize the above technical solution, S23 is specifically:

[0122] If predicting the time-of-use electricity price for next year this year, the historical electricity consumption and electricity price for each minute of each day in N years are used as training data to input into the neural network to obtain a trained neural network. The weighted average of N years is used as verification data to verify the reliability of the model. The weighted average of the electricity consumption and electricity price for each minute in the past 4 years is a*(N - 3)+b*(N - 2)+c*(N - 1)+d*N, where N is the previous year, a < b < c < d, and a + b + c + d = 1. The verification data is input into the model to obtain the predicted electricity consumption and electricity price for each minute of each day in 2024.

[0123] If predicting the time-of-use electricity price for the 2nd - 4th quarters of this year in the first quarter of this year, the historical electricity consumption and electricity price for each minute of each day in N years are used as training data to input into the neural network to obtain a trained neural network. The electricity consumption and electricity price for each minute of each day in the first quarter of this year are used as verification data to verify the reliability of the model. The verification data is input to obtain the predicted electricity consumption and electricity price for each minute of each day in the 2nd - 4th quarters of this year.

[0124] To further optimize the above technical solution, S24 is specifically as follows:

[0125] For each day of each quarter of the prediction, take the integral at intervals of M minutes to obtain the integral value within the corresponding interval. If the integral value of electricity consumption within the interval is greater than the preset value, it is the peak electricity consumption period; or set an electricity consumption threshold Q, and the time when it is greater than Q is the peak period.

[0126] To further optimize the above technical solution, S25 is specifically as follows:

[0127] Take the union of the peak electricity consumption periods for each day of each quarter of the prediction, and perform decimal rounding. The rounding cannot exceed the earliest time and the latest time of the peak electricity consumption period for each day of each quarter, so as to obtain the peak electricity consumption period for this quarter, and the remaining periods are general periods.

[0128] According to the peak electricity consumption period and the general period, and the moments corresponding to the current peak - valley electricity price in the place where the user is located, divide a day into multiple time periods, and there is no overlap between the peak electricity consumption period and the peak - valley electricity price.

[0129] To further optimize the above technical solution, S3 is specifically as follows:

[0130] S31. Judge the current time period;

[0131] S32. Collect the current device information, including but not limited to photovoltaic power generation, energy storage SOC, power consumption, and electric vehicle SOC;

[0132] S33. On the basis of meeting the power demand, with the goal of minimizing cost, formulate the control strategy for the current time period based on the constraint conditions;

[0133] S34, issuing control strategies and making fine adjustments to the control according to the data collected in real time;

[0134] S35. Determine whether the current time is still in the current time period. If it is still in the current time period, continue to execute the control strategy of the time period. Otherwise, end this time period and enter the next time period.

[0135] To further optimize the above technical solution, the minimum cost in S33 is specifically:

[0136] Construct the minimum cost objective:

[0137] minC=min(C PV +C bss +C EV -C grid )

[0138] Where C is the system cost, C PV is the photovoltaic unit cost, C bss is the energy storage unit cost, C EV is the cost of electric vehicles, C grid is the grid cost;

[0139] The specific cost of photovoltaic units is:

[0140] C PV =C o +C yun

[0141] In the formula, C PV is the photovoltaic unit cost, C o is the initial investment in photovoltaic power generation, C yun For operation and maintenance costs;

[0142] The specific cost of energy storage unit is:

[0143]

[0144] In the formula, C bss is the cost of the energy storage unit, P bss is the charging or discharging power of the energy storage unit, is the power upper limit, α bss , β bss , γ bss is the energy storage unit cost coefficient, is the discharge depth;

[0145] The specific grid costs are:

[0146]

[0147] Where ξ(t) is the limit value that cannot simultaneously purchase and sell electricity from the power grid in the same time period, and its value is 0 or 1.grid_i (t) is the power purchased from the power grid during period t, C grid_t (t) is the price of electricity purchased from the power grid during period t, P grid_o (t) is the power sold to the grid during period t, C grid_o (t) is the price of electricity sold to the grid during period t.

[0148] To further optimize the above technical solution, the constraints in S33 specifically include power balance, photovoltaic constraints, energy storage constraints, and electric vehicle constraints;

[0149] Power Balance:

[0150] P PV +P bss +P grid =P EV +P load

[0151] Where P PV is the photovoltaic power, P bss is the energy storage power, P grid is the grid power, P EV is the electric vehicle power, P load is the load power;

[0152] The photovoltaic constraint is:

[0153] P PV Min ≤P PV ≤P PV Max

[0154] Where P PV Min is the lower limit of photovoltaic output, P PV Max is the upper limit of photovoltaic output;

[0155] The energy storage constraint is:

[0156]

[0157]

[0158] SOC min ≤SOC≤SOC max

[0159] The electric vehicle constraints are:

[0160]

[0161]

[0162]

[0163] First, a home energy management model is established, including rooftop photovoltaic system, AC and DC power loads, electric vehicles, energy storage system, power grid, energy management system and integrated photovoltaic storage and direct-flexible machine. The rooftop photovoltaic system is the main energy source of the entire system; electric vehicles, energy storage system and power grid can all be used as energy sources or "loads"; the energy management system formulates the optimal control strategy for the system; the integrated photovoltaic storage and direct-flexible machine has the function of energy routing, realizing voltage changes and two-way flow of energy.

[0164] The photovoltaic unit model is:

[0165]

[0166] A=T+(0.0256×I s (t))

[0167] Where P pv (t) is the photovoltaic output power, P pv_0 is the photovoltaic rated power, I s (t) is the solar radiation intensity, I s_ref is the solar radiation intensity under standard conditions, T ref is the reference temperature under standard conditions, δ is the temperature coefficient, and T is the ambient temperature;

[0168] The energy storage unit model includes the charging process and the discharging process, specifically:

[0169] The charging process is:

[0170]

[0171] The discharge process is:

[0172]

[0173] To restrict overcharge and overdischarge:

[0174] SOC min ≤SOC≤SOC max

[0175] In the formula, E bss (t) is the remaining power of the energy storage unit at time t, δ e is the self-discharge efficiency of the energy storage unit, P in (t) is the charging power, P ex (t) is the discharge power, η in is the charging efficiency, η ex is the discharge efficiency, is the maximum capacity of the energy storage unit, is the minimum value of the energy storage unit, is the rated charging power of the energy storage unit at time t, is the rated discharge power of the energy storage unit at time t, SOC min The minimum state of charge, SOC max is the maximum value of the state of charge;

[0176] The electric vehicle model includes the charging process and the discharging process, specifically:

[0177] Charging process:

[0178]

[0179] Discharge process:

[0180]

[0181] To restrict overcharge and overdischarge:

[0182]

[0183] In the formula, Q EV (t) is the remaining power of the electric vehicle at time t, P EV in (t) is the charging power of the electric vehicle, P EV ex (t) is the discharge power of the electric vehicle, is the charging efficiency of the electric vehicle, is the discharge efficiency of the electric vehicle, is the maximum energy of the electric vehicle, is the minimum energy of the electric vehicle, is the rated charging power of the electric vehicle at time t, is the rated discharge power of the electric vehicle at time t, is the minimum value of the electric vehicle's state of charge, is the maximum value of the electric vehicle's state of charge, and is the power demand at time t.

[0184] Then, the user's electricity consumption habits, electric vehicle usage, electricity prices and other needs are analyzed and divided into time periods. Taking the spring electricity consumption of ordinary office workers as an example, a day is divided into the following time periods: t1 normal electricity consumption at off-peak prices, t2 peak electricity consumption at off-peak prices, t3 peak electricity consumption at peak prices, t4 normal electricity consumption at peak prices, t5 normal electricity consumption at normal prices, t6 peak electricity consumption at normal prices, and t7 peak electricity consumption at peak prices.

[0185] Then, according to the photovoltaic power generation and user electricity consumption in each time period, a control plan with the best electricity cost is formulated. The specific process is as follows:

[0186] Get the user's minute-by-minute historical electricity consumption and electricity price for N years;

[0187] Determine whether the number of days in each quarter of each year in the past N years is equal to the number of days in each quarter of the year in which the time-of-use electricity price will be implemented. If not, for the quarters with different numbers of days, randomly copy or delete the existing quarters with different numbers of days so that the number of days in each quarter of each year in the past N years is equal to the number of days in each quarter of the year in which the time-of-use electricity price will be implemented;

[0188] Use the difference to make the number of days equal, and delete the days with similar electricity consumption per minute and electricity price, so as to make the number of days equal.

[0189] When predicting the time-of-use electricity price for next year this year, the historical electricity consumption and electricity price of each minute of each day in N years are used as training data and input into the neural network to obtain a trained neural network. The weighted average of N years is used as verification data to verify the reliability of the model. The weighted average of the electricity consumption and electricity price of each minute in the past 4 years is a*(N - 3)+b*(N - 2)+c*(N - 1)+d*N, where N is the previous year, a < b < c < d, and a + b + c + d = 1; the verification data is input into the model to obtain the predicted electricity consumption and electricity price of each minute of each day in 2024.

[0190] If predicting the time-of-use electricity price for the 2nd - 4th quarters of this year in the first quarter of this year, the historical electricity consumption and electricity price of each minute of each day in N years are used as training data and input into the neural network to obtain a trained neural network; the electricity consumption and electricity price of each minute of each day in the first quarter of this year are used as verification data to verify the reliability of the model; input the verification data to obtain the predicted electricity consumption and electricity price of each minute of each day in the 2nd - 4th quarters of this year.

[0191] Integrate every M minutes of each day in the predicted each quarter to obtain the integral value within the corresponding interval. If the electricity consumption integral value within the interval is greater than the preset value, it is the peak electricity consumption period; or set an electricity consumption threshold Q, and the time when it is greater than Q is the peak period.

[0192] Take the union according to the predicted peak electricity consumption period of each day in each quarter, and perform decimal rounding. The rounding and truncation cannot exceed the earliest time and the latest time of the peak electricity consumption period of each day in each quarter, so as to obtain the peak electricity consumption period of this quarter, and the rest of the time is the normal period.

[0193] According to the peak electricity consumption period and the normal period, and the corresponding time of the current peak-valley electricity price where the user is located, divide a day into multiple time periods, and there is no overlap between the peak electricity consumption period and the peak-valley electricity price.

[0194] 1) Judge the current time period. Taking the electricity consumption of ordinary office workers in spring as an example, it can be divided into t1 off-peak electricity price for ordinary electricity use, t2 off-peak electricity price for peak electricity consumption, t3 peak electricity price for peak electricity consumption, t4 peak electricity price for ordinary electricity use, t5 normal electricity price for ordinary electricity use, t6 normal electricity price for peak electricity consumption, t7 peak electricity price for peak electricity consumption.

[0195] 2) Set the type of electric vehicle, which can be divided into work electric vehicles and leisure electric vehicles. Work electric vehicles are mainly used by office workers to go to and from get off work, and are parked in the company parking lot during working hours; leisure electric vehicles are used for daily activities other than work, such as playing, shopping, etc., and can be left at home during the day.

[0196] 3) Collect current equipment information, including but not limited to photovoltaic power generation, energy storage SOC, power consumption, and electric vehicle SOC.

[0197] 4) On the basis of meeting the power demand, formulate the control strategy for the current time period with the minimum cost as the goal.

[0198] Construct the minimum cost objective:

[0199] minC=min(C PV +C bss +C EV -C grid )

[0200] Where C is the system cost, C PV is the photovoltaic unit cost, C bss is the energy storage unit cost, C EV is the cost of electric vehicles, C grid is the grid cost;

[0201] The specific cost of photovoltaic units is:

[0202] C PV =C o +C yun

[0203] In the formula, C PV is the photovoltaic unit cost, C o is the initial investment in photovoltaic power generation, C yun For operation and maintenance costs;

[0204] The specific cost of energy storage unit is:

[0205]

[0206] In the formula, C bss is the cost of the energy storage unit, P bss is the charging or discharging power of the energy storage unit, is the power upper limit, α bss , β bss , γ bss is the energy storage unit cost coefficient, is the discharge depth;

[0207] The specific grid costs are:

[0208]

[0209] Where ξ(t) is the limit value that cannot simultaneously purchase and sell electricity from the power grid in the same time period, and its value is 0 or 1. grid_i (t) is the power purchased from the power grid during period t, C grid_t (t) is the price of electricity purchased from the power grid during period t, P grid_o (t) is the power sold to the grid during period t, C grid_o (t) is the price of electricity sold to the grid during period t.

[0210] The constraints specifically include power balance, photovoltaic constraints, energy storage constraints, and electric vehicle constraints;

[0211] Power Balance:

[0212] P PV +P bss +P grid =P EV +P load

[0213] Where P PV is the photovoltaic power, P bss is the energy storage power, P grid is the grid power, P EV is the electric vehicle power, P load is the load power;

[0214] The photovoltaic constraint is:

[0215] P PV Min ≤P PV ≤P PV Max

[0216] Where P PV Min is the lower limit of photovoltaic output, P PV Max is the upper limit of photovoltaic output;

[0217] The energy storage constraint is:

[0218]

[0219]

[0220] SOC min ≤SOC≤SOC max

[0221] The electric vehicle constraints are:

[0222]

[0223]

[0224]

[0225] 5) Issue control strategies and make fine adjustments based on real-time collected data.

[0226] Determine whether the current time is still in the current time period. If it is still in the current time period, continue to execute the control strategy of the time period. Otherwise, end this time period and enter the next time period.

[0227] Specifically, take the spring as an example of an ordinary office worker user, who has only one working electric car: t1 is ordinary electricity consumption at off-peak electricity prices. During this period, the electricity consumption of t2 and t3 users can be reserved, and the energy storage system can be used to charge the electric car, and the insufficient power is supplied by the power grid; t2 is the peak electricity consumption at off-peak electricity prices. During this period, the energy storage system can be used to supply power to users to meet their daily needs; t3 is the peak electricity consumption at peak electricity prices. During this period, the photovoltaic system can be used to supply power to users, and the excess part is used to charge the energy storage system or the insufficient part is supplemented by the energy storage system; t4 is ordinary electricity consumption at peak electricity prices. During this period, the photovoltaic system can be used to supply power to users, and the excess part is connected to the Internet; t5 is ordinary electricity consumption at normal electricity prices. During this period, the photovoltaic system can be used to supply power to users, and the excess part is supplied by energy storage. After charging is completed, the remaining part is connected to the Internet; t6 is the peak electricity consumption at normal electricity prices. During this period, the energy storage system can be used to supply power to users; t7 is the peak electricity consumption at peak electricity prices. During this period, the energy storage system can be used to supply power to users, and the insufficient part is supplied by electric vehicles through V2H technology.

[0228] According to a household energy management method based on time-of-use electricity prices designed by the present invention, when designing energy management, a day is divided into several time periods according to the user's electricity usage habits and time-of-use electricity prices, and the lowest cost operation mode is selected. At the same time, V2H technology is considered, and electric vehicles are added to the operation mode. Based on the purpose of the electric vehicle, its charging and discharging are reasonably set. In addition, the household energy management system also takes seasonal factors into consideration, that is, in different seasons, the energy scheduling scheme is different, and the lowest comprehensive annual electricity cost is achieved. Compared with the existing household energy management system, the present invention fully considers V2H technology and time-of-use electricity prices in energy management, minimizes electricity costs, and achieves optimal electricity costs.

[0229] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0230] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A household energy management method based on time-of-use electricity price, characterized in that: It includes the following steps: S1. Build a home energy system model, including a photovoltaic unit model, an energy storage unit model, and an electric vehicle model; S2. Build a time-of-use electricity price neural network model, obtain the historical electricity consumption and electricity price of the user per minute for N years, train the time-of-use electricity price neural network model, and obtain the electricity price periods based on the trained time-of-use electricity price neural network model; S3. Based on the home energy system model and the electricity price periods, formulate an optimal control plan for electricity consumption costs.

2. A household energy management method based on time-of-use electricity price according to claim 1, characterized in that: The photovoltaic unit model is: A=T+(0.0256×I s (t)) Where P pv (t) is the photovoltaic output power, P pv_0 is the photovoltaic rated power, I s (t) is the solar radiation intensity, I s_ref is the solar radiation intensity under standard conditions, T ref is the reference temperature under standard conditions, δ is the temperature coefficient, and T is the ambient temperature; The energy storage unit model includes a charging process and a discharging process. Specifically: The charging process is: The discharging process is: Constrain overcharging and over-discharging: SOC min ≤SOC≤SOC max In the formula, E bss (t) is the remaining power of the energy storage unit at time t, δ e is the self-discharge efficiency of the energy storage unit, P in (t) is the charging power, P ex (t) is the discharge power, η in is the charging efficiency, η ex is the discharge efficiency, is the maximum capacity of the energy storage unit, is the minimum value of the energy storage unit, is the rated charging power of the energy storage unit at time t, is the rated discharge power of the energy storage unit at time t, SOC min The minimum state of charge, SOC max is the maximum value of the state of charge; The electric vehicle model includes a charging process and a discharging process. Specifically: The charging process: The discharging process: Constrain overcharging and over-discharging: In the formula, Q EV (t) is the remaining power of the electric vehicle at time t, P EV in (t) is the charging power of the electric vehicle, P EV ex (t) is the discharge power of the electric vehicle, is the charging efficiency of the electric vehicle, is the discharge efficiency of the electric vehicle, is the maximum energy of the electric vehicle, is the minimum energy of the electric vehicle, is the rated charging power of the electric vehicle at time t, is the rated discharge power of the electric vehicle at time t, is the minimum value of the electric vehicle's state of charge, is the maximum value of the electric vehicle's state of charge, and is the power demand at time t.

3. A household energy management method based on time-of-use electricity price according to claim 1, characterized in that: The specific content of S2 is: S21. Obtain the historical electricity consumption and electricity price of the user per minute for N years; S22. Obtain the number of days in each quarter within N years, preprocess it to make it the same as the number of days in each quarter of the year to be predicted for time-of-use electricity price, and build a training set; S23. Build a neural network, train the neural network with the training set to obtain a prediction network model, input the data set into the prediction network model, and obtain a prediction result; S24. Process the prediction result to obtain the peak electricity consumption periods, integerize the peak electricity consumption periods of each day in each quarter to obtain the standard peak electricity consumption periods and normal periods; S25. Obtain the peak-valley electricity prices and the corresponding standard periods, and divide the standard peak electricity consumption periods and normal periods based on the standard periods of the peak-valley electricity prices to obtain the electricity price periods.

4. A household energy management method based on time-of-use electricity price according to claim 3, characterized in that: The specific content of S22 is: Judge whether the number of days in each quarter of each of the nearly N years is equal to the number of days in each quarter of the year to implement time-of-use electricity price. If not, for the quarters with unequal numbers of days, randomly duplicate or delete within the existing quarters with unequal numbers of days to make the number of days in each quarter of each of the N years equal to the number of days in each quarter of the year to implement time-of-use electricity price; Use the difference to make the number of days equal, and delete the days with similar electricity consumption and electricity price per minute to make the number of days equal.

5. A household energy management method based on time-of-use electricity price according to claim 3, characterized in that: The specific content of S23 is: When predicting the time-of-use electricity price next year based on this year, input the historical electricity consumption and electricity price of each minute of each day in N years as training data into the neural network to obtain a trained neural network. Use the weighted average of N years as verification data to verify the reliability of the model. The weighted average of the electricity consumption and electricity price per minute in the past 4 years is a*(N - 3)+b*(N - 2)+c*(N - 1)+d*N, where N is the previous year, a < b < c < d, and a + b + c + d = 1; input the verification data into the model to obtain the predicted electricity consumption and electricity price of each minute of each day in 2024; If the time-of-use electricity price for the second to fourth quarters of this year is predicted in the first quarter of this year, the historical electricity consumption and electricity price every minute of every day for N years will be input into the neural network as training data to obtain a trained neural network; the electricity consumption and electricity price every minute of every day in the first quarter of this year will be used as verification data to verify the reliability of the model; the verification data will be input to obtain the predicted electricity consumption and electricity price every minute of every day for the second to fourth quarters of this year.

6. A household energy management method based on time-of-use electricity price according to claim 3, characterized in that: The S24 is specifically: The integral of each day of each predicted quarter is calculated M minutes apart to obtain the integral value within the corresponding interval, where the integral value of electricity consumption within the interval is greater than the preset value, which is the peak period of electricity consumption; or a power consumption threshold Q is set, and the moment when it is greater than Q is the peak period.

7. A household energy management method based on time-of-use electricity price according to claim 3, characterized in that: The S25 is specifically: According to the predicted peak electricity consumption period of each day in each quarter, the union is calculated and rounded to decimal. The rounding and rounding cannot exceed the earliest time and the latest time of the peak electricity consumption period of each day in each quarter, so as to obtain the peak electricity consumption period of the quarter, and the rest of the time is the general time period; According to the peak and normal periods of electricity consumption, and the time corresponding to the current peak and valley electricity prices where the user is located, a day is divided into multiple time periods, and there is no overlap between the peak and valley electricity prices.

8. A household energy management method based on time-of-use electricity price according to claim 1, characterized in that: The S3 is specifically: S31, determining the current time period; S32, collecting current equipment information, including but not limited to photovoltaic power generation, energy storage SOC, power consumption, and electric vehicle SOC; S33, on the basis of meeting the power demand, taking the minimum cost as the goal, and formulating the control strategy for the current time period based on the constraint conditions; S34, issuing control strategies and making fine adjustments to the control according to the data collected in real time; S35. Determine whether the current time is still in the current time period. If it is still in the current time period, continue to execute the control strategy of the time period. Otherwise, end this time period and enter the next time period.

9. A household energy management method based on time-of-use electricity price according to claim 8, characterized in that: The minimum cost in S33 is specifically: Construct the minimum cost objective: minC=min(C PV +C bss +C EV -C grid ) Where C is the system cost, C PV is the photovoltaic unit cost, C bss is the energy storage unit cost, C EV is the cost of electric vehicles, C grid is the grid cost; The specific cost of photovoltaic units is: C PV =C o +C yun In the formula, C PV is the photovoltaic unit cost, C o is the initial investment in photovoltaic power generation, C yun For operation and maintenance costs; The specific cost of energy storage unit is: In the formula, C bss is the cost of the energy storage unit, P bss is the charging or discharging power of the energy storage unit, is the power upper limit, α bss , β bss , γ bss is the energy storage unit cost coefficient, is the discharge depth; The specific grid costs are: Where ξ(t) is the limit value that cannot simultaneously purchase and sell electricity from the power grid in the same time period, and its value is 0 or 1. grid_i (t) is the power purchased from the power grid during period t, C grid_t (t) is the price of electricity purchased from the power grid during period t, P grid_o (t) is the power sold to the grid during period t, C grid_o (t) is the price of electricity sold to the grid during period t.

10. A household energy management method based on time-of-use electricity price according to claim 8, characterized in that: The constraints in S33 specifically include power balance, photovoltaic constraints, energy storage constraints, and electric vehicle constraints; Power Balance: P PV +P bss +P grid =P EV +P load Where P PV is the photovoltaic power, P bss is the energy storage power, P grid is the grid power, P EV is the electric vehicle power, P load is the load power; The photovoltaic constraint is: P PV Min ≤P PV ≤P PV Max Where P PV Min is the lower limit of photovoltaic output, P PV Max is the upper limit of photovoltaic output; The energy storage constraint is: SOC min ≤SOC≤SOC max The electric vehicle constraints are:

Images