Method, device and storage medium for simulating annual power load curve

Abstract

The embodiments of this specification provide a method, device and storage medium for simulating a full-year power load curve, the method comprising: simulating the power consumption of each week of the target year according to historical data, and determining the power relationship between the year, month and week; selecting a typical weekly power curve for each month, determining the power and load relationship between the week and day, and obtaining an initial full-year power load curve; establishing an optimization model to optimize the maximum load of each month, determining the load relationship between the year, month and week, and correcting the initial full-year power load curve; correcting the intraday power load curve week by week and day by day according to the power consumption difference, determining the load-power relationship between the day and hour, and obtaining a simulated final full-year power load curve.

Description

Technical Field

[0001] This document relates to the technical field of power system planning, and in particular to a method, device and storage medium for simulating annual power load curves. Background Art

[0002] The 8760-hour power load curve is a time series of the average load hour by hour for 8760 hours throughout the year. The annual 8760-hour power load curve simulation is the basis for power and electricity balance and power system production simulation, and is of great significance for making power planning and dispatching operation decisions. Its purpose is to simulate the target year 8760-hour power load curve of a region based on the historical power load curve data of a region, combined with the prediction of factors such as the maximum load and annual power consumption in the target year. It does not pursue accurate prediction of hourly power load, but the load characteristics presented by the power load curve in multiple time dimensions such as year, month, week, and day should be reasonable, that is, in line with the laws of historical data. Because the power is the integral of the load, the 8760-hour power load curve has the same value of load and power in each hour.

[0003] The difficulty of simulating the annual 8760-hour power load curve lies in two aspects: first, the load characteristics in different time dimensions of year, month, week and day are coupled with each other. How to coordinate different load characteristics in the simulation? Second, the maximum load and annual power consumption as boundary conditions of the load curve are usually given by other methods. How to correct the load curve based on the predicted results of the maximum load and annual power consumption in the target year.

[0004] In view of this, there is an urgent need to provide a 8760-hour power load curve simulation method that coordinates multi-time dimension load characteristic indicators. Summary of the invention

[0005] The purpose of the present invention is to provide a method, device and storage medium for simulating annual power load curve, aiming to solve the above-mentioned problems in the prior art.

[0006] The present invention provides a method for simulating an annual power load curve, comprising:

[0007] Simulate the electricity consumption of each week in the target year based on historical data and determine the relationship between electricity consumption in the year, month and week;

[0008] Select the typical weekly electricity curve of each month, determine the relationship between electricity and load between week and day, and obtain the initial annual power load curve;

[0009] Establishing an optimization model to optimize the maximum load of each month, determining the load relationship between the year, month and week, and revising the initial annual power load curve;

[0010] The daily power load curve is corrected week by week and day by day according to the power consumption difference, the load-power relationship between day and hour is determined, and the final simulated annual power load curve is obtained.

[0011] The present invention provides a year-round power load curve simulation device, comprising:

[0012] The first determination module is used to simulate the power consumption of each week in the target year based on historical data and determine the relationship between the power consumption of the year, month and week;

[0013] The second determination module is used to select the typical weekly power curve of each month, determine the power and load relationship between the week and the day, and obtain the initial annual power load curve;

[0014] The first correction module is used to establish an optimization model to optimize the maximum load of each month, determine the load relationship between the year, month and week, and correct the initial annual power load curve;

[0015] The second correction module is used to correct the daily power load curve week by week and day by day according to the power consumption difference, determine the load and power relationship between the day and the hour, and obtain the simulated final annual power load curve.

[0016] An embodiment of the present invention also provides a device for simulating a year-round power load curve, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the above-mentioned method for simulating a year-round power load curve when executed by the processor.

[0017] An embodiment of the present invention further provides a computer-readable storage medium, on which a program for implementing information transmission is stored, and when the program is executed by a processor, the steps of the above-mentioned method for simulating the annual power load curve are implemented.

[0018] By adopting the embodiments of the present invention, it is possible to coordinate the load characteristics of different time dimensions such as year, month, week and day in the annual 8760-hour power load curve simulation, and the indicators such as the maximum load, annual power consumption, and seasonal imbalance coefficient of the simulated 8760-hour power load curve meet the boundary conditions given in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0020] Figure 1is a flow chart of a method for simulating an annual power load curve according to an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram of the proportion of weekly electricity consumption to annual electricity consumption according to an embodiment of the present invention;

[0022] Figure 3 1 is a schematic diagram of a typical weekly electricity consumption curve for January selected according to an embodiment of the present invention;

[0023] Figure 4 is a schematic diagram of the results of the optimization of the maximum loads of each month in an embodiment of the present invention;

[0024] Figure 5 is a schematic diagram of daily load curve adjustment when daily power consumption is increased according to an embodiment of the present invention;

[0025] Figure 6 is a schematic diagram of daily load curve adjustment when daily power consumption is reduced in an embodiment of the present invention;

[0026] Figure 7 This is a schematic diagram of a 8760-hour power load curve obtained by the final simulation of an embodiment of the present invention;

[0027] Figure 8 It is a schematic diagram of a device for simulating the annual power load curve according to the first embodiment of the device of the present invention;

[0028] Fig. 9 It is a schematic diagram of a device for simulating the annual power load curve according to the second embodiment of the device of the present invention. DETAILED DESCRIPTION

[0029] In order to solve the problems in the prior art, the embodiment of the present invention discloses a method for simulating an 8760-hour power load curve: simulating the power consumption of each week of the target year according to historical data; selecting the typical weekly power curve of each month based on the principle of "shortest average Euclidean distance" to preliminarily obtain the 8760-hour power load curve; establishing an optimization model to optimize the maximum load of each month and correct the 8760-hour power load curve; correcting the intraday power load curve week by week and day by day according to the power consumption difference to obtain the final simulated 8760-hour power load curve. The present invention can coordinate the load characteristics of different time dimensions of year, month, week and day in the annual 8760-hour power load curve simulation, and the indicators such as the maximum load, annual power consumption, and seasonal imbalance coefficient of the simulated 8760-hour power load curve meet the boundary conditions given in advance.

[0030] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0031] Example 1

[0032] According to an embodiment of the present invention, a method for simulating an annual power load curve is provided. Figure 1 is a flow chart of the annual power load curve simulation method according to an embodiment of the present invention. Figure 1 As shown, the annual power load curve simulation method according to an embodiment of the present invention specifically includes:

[0033] Step 101, simulating the power consumption of each week in the target year based on historical data, and determining the power consumption relationship between the year, month and week;

[0034] Step 102, based on the principle of shortest average Euclidean distance, select the typical weekly electricity curve of each month, determine the relationship between electricity and load between week and day, and obtain the initial annual power load curve;

[0035] Step 103, establishing an optimization model to optimize the maximum load of each month, determining the load relationship between the year, month and week, and correcting the initial annual power load curve;

[0036] Step 104, correct the daily power load curve on a weekly and daily basis according to the power consumption difference, determine the load-power relationship between the day and the hour, and obtain the simulated final annual power load curve.

[0037] Example 2

[0038] The present invention needs to coordinate the load characteristics of different time dimensions of year, month, week and day, and simulate the maximum load, annual power consumption, seasonal imbalance coefficient and other indicators of the 8760-hour power load curve to meet the boundary conditions given in advance. The present invention is described in detail below in conjunction with the specific implementation methods and the drawings of the specification.

[0039] like Figure 1 As shown, the present invention provides a method for simulating an 8760-hour power load curve, comprising the following steps:

[0040] This embodiment first simulates the power consumption of each week of the target year based on historical data to determine the "year-month-week" power relationship; based on the "shortest average Euclidean distance" principle, selects the typical weekly power curve of each month, determines the "week-day" power and load relationship, and preliminarily obtains the 8760-hour power load curve; establishes an optimization model to optimize the maximum load of each month, determines the "year-month-week" load relationship, and corrects the 8760-hour power load curve; corrects the intraday power load curve week by week and day by day according to the power consumption difference, determines the "day-hour" load-power relationship, and obtains the final simulated 8760-hour power load curve. See Table 1 for details.

[0041] Table 1 The relationship between power and load in different time dimensions determined in each step

[0042]

[0043] The embodiment of the present invention takes the 8760-hour power load curve of a regional power grid as an example to demonstrate the specific implementation method.

[0044] S1: The historical annual 8760-hour power load sequence is supplemented with a power load sequence containing 53 full weeks (8904 hours); according to the proportion of historical weekly power consumption to the annual (53 full weeks) power consumption, the proportion of power consumption in each week of the target year is simulated, and combined with the power consumption forecast for the target year, the power consumption in each week of the target year is obtained.

[0045] That is, based on the data of the previous year and the next year of historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve containing 53 full weeks of historical year j is obtained, and the weekly power consumption is calculated.

[0046] Based on weekly electricity consumption Calculate the proportion of electricity consumption in week w to the annual electricity consumption

[0047] Based on N years of historical data, calculate the average proportion of electricity consumption in week w to annual electricity consumption and will As the proportion of electricity consumption in the wth week of the target year to the annual electricity consumption;

[0048] According to the target annual electricity consumption E f Get the simulated electricity consumption for each week of the target year.

[0049] Specifically:

[0050] make represents the average power load in the tth hour of historical year j, then The 8760-hour power load curve for historical year j is shown. Based on the data of the previous year and the next year of historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve for historical year j containing 53 full weeks is obtained, and the weekly power consumption is calculated.

[0051]

[0052] In the above formula, t∈w means that the tth hour belongs to the wth week. On this basis, calculate the proportion of electricity consumption in the wth week to the annual electricity consumption

[0053]

[0054] Based on N years of historical data, calculate the average proportion of electricity consumption in week w to annual electricity consumption

[0055] Will As the proportion of the power consumption in the wth week of the target year to the annual power consumption. Figure 2 shown.

[0056] The target annual electricity consumption E is obtained through traditional methods such as department analysis method and output value unit consumption method. f (It should be considered that 8760 hours has been expanded to 8904 hours). Further, the simulated power consumption of each week in the target year is obtained:

[0057]

[0058] It is required that the weekly electricity consumption of the final simulated target year 8760-hour power load curve is equal to this value.

[0059] S2: Considering the differences in load characteristics between months, select the typical weekly electricity curve of each month, calculate the proportion of electricity consumption in each hour of the week to the weekly electricity consumption, and based on the simulated electricity consumption in each week of the target year in the previous step and the integral relationship between electricity and load, preliminarily obtain the 8760-hour power load curve of the target year.

[0060] For month m, find all the historical weekly load curves corresponding to month m Taking January 2021 as an example, there are 5 weeks in this month (December 28, 2020 to January 3, 2021, January 4, 2021 to January 10, 2021 to January 11, 2021 to January 17, 2021 to January 18, 2021 to January 24, 2021 to January 18, 2021 to January 31), so there are 5 weekly load curves, each of which contains 168 points (hours).

[0061] For each weekly load curve, calculate the daily electricity consumption

[0062]

[0063] In the above formula, t∈d, d∈w represent all hours of the dth day of the wth week. Thus, the daily scale weekly electricity curve of the mth month is That is, the weekly electricity consumption curve includes the electricity consumption of 7 days in a week. From all the historical weekly electricity consumption curves in the mth month, select a curve that is most representative in form. The most representative curve is characterized by the "shortest average distance" from other curves. The selection method includes the following three steps:

[0064] (1) The weekly power curve Taking the largest element as the reference value, normalized to

[0065]

[0066] Where max() is the maximum value. According to the actual situation, we should first remove the obviously "atypical" weekly electricity consumption curve, such as holidays falling within the week, which results in higher weekend electricity consumption than weekday electricity consumption. Figure 3 As shown, there are 9 weekly electricity consumption curves for January in this embodiment.

[0067] (2) If there are n weekly electricity curves, the Euclidean distance between every two weekly electricity curves is calculated to form a Euclidean distance correlation matrix (symmetric matrix):

[0068]

[0069] Where d Euler is the Euclidean distance between the two curves, with the Euclidean distance d between the first and second curves 12 Euler For example,

[0070]

[0071] Where x1, x2, ... x7 represent the elements of the first curve, and y1, y2, ... y7 represent the elements of the second curve. The symmetric matrix of the monthly weekly electricity consumption curve of this embodiment is shown in Table 2:

[0072] Table 21 Symmetrical matrix table of monthly weekly electricity curve

[0073]

[0074]

[0075] (3) Select the curve with the shortest average Euclidean distance to other curves as the typical weekly electricity curve:

[0076]

[0077] In this embodiment, the average Euclidean distances of the weekly electricity curves in January to other curves are 0.057, 0.046, 0.047, 0.039, 0.038, 0.070, 0.050, and 0.047, respectively. Since curve 5 has the shortest average Euclidean distance, curve 5 is selected as the typical weekly electricity curve in January. Figure 3 Shown by the solid line.

[0078] The typical weekly electricity curve of month m is recorded as j* in the year and w* in the week. Extract the historical load curve of the week Since the integral of the load on an hourly scale is equal to the power consumption, the proportion of power consumption at time t in the week to the total power consumption of the whole week is:

[0079]

[0080] At this point, the simulated electricity consumption for each week of the target year has been obtained. And the proportion of electricity consumption at each hour of each week (i.e. average load within the hour) to the weekly electricity consumption h wt By multiplying the two and excluding the days added in the first and last week of the year, we can get the target year 8760-hour power load curve {D f(1) t , t=1,2,…8760}:

[0081]

[0082] The reason why the proposed method measures electricity consumption and determines the typical curve on a weekly basis rather than on a monthly or daily basis is that there is an obvious rule in the weekly load curve that the load on weekdays is greater than the load on weekends, and the loads on each weekday are also different. This factor cannot be taken into account on a monthly or daily basis.

[0083] Although the target annual power load curve is obtained here, the maximum load may not correspond to the given boundary conditions, and load characteristic indicators such as the seasonal imbalance coefficient (the ratio of the average maximum load of each month throughout the year to the annual maximum load) are also different from the predicted values, so the simulated power load curve needs to be further adjusted.

[0084] S3: Taking the prediction results of the annual maximum load, the monthly maximum load, and the seasonal imbalance coefficient as constraints, an optimization model is established to optimize the monthly maximum load. The load curve is amplified in the same proportion according to the load magnification factor before and after optimization, and the power load curve is corrected. The specific approach is:

[0085] First, the current target annual power load curve {D f(1) t , the maximum load of each month at t = 1, 2, ... 8760} is expressed as {D f(1) m , m=1,2,…12}:

[0086]

[0087] Then, with the constraints that the seasonal imbalance coefficient is in a reasonable range, the maximum loads of each month satisfy a certain relationship, and the maximum load is consistent with the given boundary conditions, and the goal is to minimize the sum of the squares of the maximum load adjustments of each month, an optimization model is established to optimize the new maximum load of each month {D f(2) m , m=1,2,…12}. The optimization model objective function is:

[0088]

[0089] The constraints are:

[0090] (1) Seasonal imbalance coefficient constraint

[0091]

[0092] (2) Relationship constraints between the maximum loads of each month

[0093]

[0094] (3) Annual maximum load constraints

[0095]

[0096] In the above formula, η min and η max are the minimum and maximum values ​​of the seasonal imbalance coefficient respectively; Respectively represent the proportional relationship between the maximum loads of month m and m' (as boundary conditions, if not, this constraint is not included); m* is the month in which the maximum load is expected to occur, is the expected annual maximum load.

[0097] Solve the above optimization model to obtain the optimized maximum load of each month {D f(2) m , m=1,2,…12}. Before and after optimization, the magnification of the maximum load of each month {α m ,m=1,2,…12} is

[0098]

[0099] According to the magnification factor, the load curves of each month are magnified in the same proportion to obtain the new 8760-hour power load curve {D f(2) t , t=1,2,…8760}

[0100]

[0101] The above method is used in this embodiment to obtain the maximum load of each month as follows: Figure 4 shown.

[0102] S4: Under the condition that the daily maximum load remains unchanged and the weekly power consumption is equal to the simulated value in S1, the daily power load curve is corrected week by week and day by day according to the power consumption difference, and the final simulated 8760-hour power load curve is obtained, and the power consumption and maximum load are equal to the given values ​​of the boundary conditions. The specific method is:

[0103] Calculate the current 8760-hour power load curve {D f(2) t , t=1,2,…8760}, the maximum load of each day {D f(2) d , d = 1, 2, ... 365}

[0104]

[0105] Weekly electricity consumption E f(2) w :

[0106]

[0107] Daily electricity consumption E f(2) d :

[0108]

[0109] Taking the wth week as an example, the modified power load curve {D f(2) t , t∈w}, so that the weekly electricity consumption is equal to The maximum daily load is maintained at {D f(2) d , d∈w}. The correction method is as follows:

[0110] Calculate the target amount of weekly electricity consumption adjustment With the current amount E f(2) w The difference E gap w :

[0111]

[0112] According to the ratio of daily electricity consumption, the difference is divided into each day, that is, the change of electricity consumption on day d in a week is E gap wd :

[0113]

[0114] According to E gap w The symbol is divided into two cases:

[0115] Case 1: If E gap w If it is greater than 0, it means that the power load curve needs to be enlarged. The principle of adjusting the power load curve on the d day is: the power consumption increases by E gap wd ; The maximum load is still D f(2) d ; The relative order of the load size in each hour remains unchanged. The adjustment method is:

[0116] (1) Calculate the hourly load adjustment space D δ t and daily electricity consumption adjustment space E δ t

[0117]

[0118]

[0119] Load adjustment space D δ t It is the difference between the current hourly load and the maximum load; the daily electricity consumption adjustment space is the integral of the difference within the day.

[0120] (2) Calculation of adjustment coefficient β d and hourly load adjustment D gap t

[0121]

[0122]

[0123] The adjustment coefficient is the ratio of the daily electricity consumption increase to the daily electricity consumption adjustment space, that is, how much of the daily electricity consumption adjustment space is used as the daily electricity consumption increase. The hourly load adjustment is to calculate the hourly load increment.

[0124] (3) Daily hourly load

[0125]

[0126] The load at hour t on day d is equal to the basic D f(2) t Superimposed load adjustment D gap t .

[0127] If the daily power consumption is increased by 2 million kWh in this embodiment, the daily load curves before and after the adjustment are as follows: Figure 5 As shown, the lower curve is before adjustment, and the upper curve is after adjustment.

[0128] Case 2: If E gap w If it is less than 0, it means that the power load curve needs to be reduced. The principle of adjusting the power load curve on the d day is: the power consumption increases by E gap wd (E gap wd is a negative value, the power consumption is actually reduced); the maximum load is still D f(2) d ; The relative order of the load size in each hour remains unchanged. The adjustment method is:

[0129] (1) Calculate the hourly load adjustment D except for the time of the maximum daily load gap t

[0130]

[0131] Divide the daily electricity consumption change evenly into 23 hours excluding the time of maximum daily load.

[0132] (2) Load per hour on day

[0133]

[0134] The load at hour t on day d is equal to the basic D f(2) t Superimposed load adjustment D gap t ;t max d It is the hour with the maximum load of the day.

[0135] If the daily power consumption is reduced by 2 million kWh in this embodiment, the daily load curves before and after the adjustment are as follows: Figure 6 As shown, the upper curve is before adjustment, and the lower curve is after adjustment.

[0136] According to the above method, the 8760-hour power load curve {D f(2) t, t=1,2,…8760} correct the daily power load curve week by week and day by day to obtain the final simulated 8760-hour power load curve

[0137] The 8760-hour power load curve finally simulated by the embodiment of the present invention is as follows: Figure 7 shown.

[0138] Example 3

[0139] According to an embodiment of the present invention, a device for simulating a year-round power load curve is provided. Figure 8 Schematic diagram of the annual power load curve simulation device according to an embodiment of the present invention. Figure 8 As shown, the annual power load curve simulation device according to an embodiment of the present invention specifically includes:

[0140] The first determination module 80 is used to simulate the power consumption of each week of the target year based on historical data and determine the relationship between the power consumption of the year, month and week;

[0141] The second determination module 82 is used to select a typical weekly power curve of each month based on the principle of the shortest average Euclidean distance, determine the power and load relationship between the week and the day, and obtain an initial annual power load curve;

[0142] The first correction module 84 is used to establish an optimization model to optimize the maximum load of each month, determine the load relationship between the year, month and week, and correct the initial annual power load curve;

[0143] The second correction module 86 is used to correct the daily power load curve according to the power consumption difference on a weekly and daily basis, determine the load power relationship between the day and the hour, and obtain the simulated final annual power load curve.

[0144] The first determination module 80 is specifically used to: supplement the first and last week of the year with a full week starting from Monday and ending on Sunday based on the data of the previous year and the next year of the historical year j, obtain an 8904-hour power load curve containing 53 full weeks of the historical year j, and calculate the weekly power consumption Based on weekly electricity consumption Calculate the proportion of electricity consumption in week w to the annual electricity consumption Based on N years of historical data, calculate the average proportion of electricity consumption in week w to annual electricity consumption and will As the proportion of electricity consumption in the wth week of the target year to the annual electricity consumption; according to the target annual electricity consumption E f Get the simulated electricity consumption for each week of the target year.

[0145] Specifically: represents the average power load in the tth hour of historical year j, then The 8760-hour power load curve of historical year j is represented. According to the data of the previous year and the next year of historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve of historical year j containing 53 full weeks is obtained. The weekly power consumption is calculated according to formula 1

[0146]

[0147] Among them, t∈w means that the tth hour belongs to the wth week;

[0148] According to formula 2, calculate the proportion of electricity consumption in week w to the annual electricity consumption

[0149]

[0150] Based on N years of historical data, the average proportion of electricity consumption in week w to annual electricity consumption is calculated based on formula 3. and will As the proportion of electricity consumption in the wth week of the target year to the annual electricity consumption:

[0151]

[0152] Based on Formula 4, the target annual electricity consumption E f Get the simulated electricity consumption for each week of the target year:

[0153]

[0154] The second determining module 82 is specifically configured to:

[0155] For month m, determine all historical weekly load curves corresponding to month m For each weekly load curve, calculate the daily power consumption according to Formula 5

[0156] Among them, t∈d, d∈w represents all hours of the dth day of the wth week;

[0157] Weekly power curve Taking the largest element as the reference value, it is converted into

[0158]

[0159] Among them, max() is to find the maximum value;

[0160] If there are n weekly electricity curves, the Euclidean distance between every two weekly electricity curves is calculated to form the Euclidean distance association matrix shown in Formula 7:

[0161]

[0162] Among them, d Euler is the Euclidean distance between the two curves;

[0163] Calculate the Euclidean distance d between two curves based on their elements Euler ;

[0164] According to Formula 8, the curve with the shortest average Euclidean distance to other curves is selected as the typical weekly power curve:

[0165]

[0166] The first correction module 84 is specifically used for:

[0167] According to the typical weekly electricity consumption curve of each month, calculate the proportion of electricity consumption in each hour of the week to the weekly electricity consumption. h wt , based on the simulated electricity consumption in each week of the target year As well as the proportion of electricity consumption at each hour of each week to the weekly electricity consumption, the target year 8760-hour power load curve is preliminarily obtained according to formula 9 and the maximum load of each month {D f(1) m , m=1,2,…12}:

[0168]

[0169] With the constraints that the seasonal imbalance coefficient is in a reasonable range, the maximum loads of each month satisfy a certain relationship, and the maximum load is consistent with the given boundary conditions, and the goal is to minimize the sum of the squares of the maximum load adjustments of each month, a monthly maximum load optimization model is established, and the new monthly maximum load {D f(2) m , m=1,2,…12}, before and after optimization, the magnification factor of the maximum load of each month is determined according to formula 10 {α m , m=1,2,…12}:

[0170]

[0171] According to the magnification factor, the load curves of each month are magnified in the same proportion, and the new 8760-hour power load curve {D f(2) t , t=1,2,…8760}:

[0172]

[0173] The first correction module 84 is specifically used for:

[0174] Determine the objective function according to formula 12:

[0175]

[0176] Among them, the maximum load of each month {D f(2) m , m=1,2,…12} are optimization variables;

[0177] Determine the seasonal imbalance coefficient constraint according to formula 12:

[0178]

[0179] Among them, η min and η max are the minimum and maximum values ​​of the seasonal imbalance coefficient respectively;

[0180] Determine the relationship constraints between the maximum loads of each month according to formula 14:

[0181]

[0182] in, Respectively represent the proportional relationship between the maximum loads of month m and m';

[0183] Determine the annual maximum load constraint according to Formula 15:

[0184]

[0185] Among them, m * is the month in which the maximum load is expected to occur. is the estimated annual maximum load;

[0186] The second correction module 86 is specifically used for:

[0187] According to formula 16, the current 8760-hour power load curve {D f(2) t , t=1,2,…8760}, the maximum load of each day {D f(2) d , d = 1, 2, ... 365}:

[0188]

[0189] Calculate weekly electricity consumption E according to formula 17 f(2) w :

[0190]

[0191] Calculate the daily electricity consumption E according to formula 18 f(2) d :

[0192]

[0193] Calculate the adjusted target amount of weekly electricity consumption for week w according to formula 19 With the current amount E f(2) w The difference E gap w :

[0194]

[0195] According to formula 20, the difference is divided into each day according to the ratio of daily electricity consumption, that is, the change of electricity consumption on day d in a week is E gap wd :

[0196]

[0197] If E gap w Greater than 0, increase the power consumption by E gap wd , the maximum load is still D f(2) d , the relative order of the load size of each hour remains unchanged, and the hourly load adjustment space D is calculated according to formula 21 and formula 22 δ t and daily electricity consumption adjustment space E δ t :

[0198]

[0199]

[0200] Calculate the adjustment factor β according to Formula 23 and Formula 24 d and hourly load adjustment D gap t :

[0201]

[0202]

[0203] Calculate the load for each hour on day d according to formula 25

[0204]

[0205] The load at hour t on day d is equal to the basic D f(2) t Superimposed load adjustment D gap t ;

[0206] If E gap w Less than 0, power consumption increases by E gap wd , the maximum load is D f(2) d , the relative order of the load size of each hour remains unchanged, and the hourly load adjustment D is calculated according to formula 26 except for the time of the maximum load of the day. gap t :

[0207]

[0208] Divide the daily electricity consumption change evenly into 23 hours except the time of maximum daily load;

[0209] Calculate the load for each hour on day d according to formula 27

[0210]

[0211] Adjust the load at hour t on day d to be equal to the basic D f(2) t Superimposed load adjustment D gap t , adjust t max d The hour with the maximum load of the day;

[0212] Corrected power load curve {D f(2) t , t∈w}, so that the weekly electricity consumption is equal to E f w , the maximum daily load is maintained at {D f(2) d , d∈w};

[0213] For the 8760-hour power load curve {D f(2) t , t=1,2,…8760} correct the daily power load curve week by week and day by day to obtain the final simulated 8760-hour power load curve

[0214] The embodiment of the present invention is a device embodiment corresponding to the above method embodiment. The specific operations of each module can be understood by referring to the description of the method embodiment, which will not be repeated here.

[0215] Example 4

[0216] The embodiment of the present invention provides a device for simulating the annual power load curve. Fig. 9 As shown, it includes: a memory 90, a processor 92, and a computer program stored in the memory 90 and executable on the processor 92. When the computer program is executed by the processor 92, the steps described in the method embodiment are implemented.

[0217] Example 5

[0218] An embodiment of the present invention provides a computer-readable storage medium, on which a program for implementing information transmission is stored. When the program is executed by the processor 92, the steps described in the method embodiment are implemented.

[0219] The computer-readable storage medium in this embodiment includes, but is not limited to, ROM, RAM, magnetic disk or optical disk, etc.

[0220] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating an annual power load curve, characterized in that: include: Simulate the electricity consumption of each week in the target year based on historical data and determine the relationship between electricity consumption in the year, month and week; Select the typical weekly electricity curve of each month, determine the relationship between electricity and load between week and day, and obtain the initial annual power load curve; specifically, for the mth month, determine all the historical weekly load curves corresponding to the mth month {D h jt ,t∈w|w∈m,j=1,2,…N}, for each historical weekly load curve, calculate the daily power consumption E according to Formula 5 h jwd : Among them, t∈d, d∈w represents all hours of the dth day of the wth week; The weekly power curve {E h jwd ,d=1,2,…7} takes the largest element as the reference value and converts it into {e h jwd ,d=1,2,…7}: Among them, max() is to find the maximum value; If there are n weekly electricity curves, the Euclidean distance between every two weekly electricity curves is calculated to form the Euclidean distance association matrix shown in Formula 7: Among them, d Euler is the Euclidean distance between the two curves; Calculate the Euclidean distance d between two curves based on their elements Euler ; According to Formula 8, the curve with the shortest average Euclidean distance to other curves is selected as the typical weekly power curve: Establish an optimization model to optimize the maximum load of each month, determine the load relationship between the year, month and week, and modify the initial annual power load curve; specifically, calculate the proportion of each hour of electricity consumption in the week to the weekly electricity consumption according to the typical weekly electricity consumption curve of each month. h wt , based on the simulated electricity consumption E in each week of the target year f w As well as the proportion of electricity consumption at each hour of each week to the weekly electricity consumption, the target year 8760-hour power load curve {D f(1) t |D f(1) t =e h wt *E f w , t∈w, t=1,2,…8760} and the maximum load of each month {D f(1) m , m=1,2,…12}: With the constraints that the seasonal imbalance coefficient is in a reasonable range, the maximum loads of each month satisfy a certain relationship, and the maximum load is consistent with the given boundary conditions, and the goal is to minimize the sum of the squares of the maximum load adjustments of each month, a monthly maximum load optimization model is established, and the new monthly maximum load {D f(2) m , m=1,2,…12}, before and after optimization, the magnification factor of the maximum load of each month is determined according to formula 10 {α m , m=1,2,…12}: According to the magnification factor, the load curves of each month are magnified in the same proportion, and the new 8760-hour power load curve {D f(2) t , t=1,2,…8760}: The daily power load curve is corrected week by week and day by day according to the power consumption difference, the load-power relationship between day and hour is determined, and the final simulated annual power load curve is obtained.

2. The method according to claim 1, characterized in that According to the data of the previous year and the next year of historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve containing 53 full weeks of historical year j is obtained, and the weekly power consumption E is calculated. h jw Specifically include: According to the data of the previous year and the next year of historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve containing 53 full weeks of historical year j is obtained. The weekly power consumption E is calculated according to formula 1 h jw : Among them, t∈w means that the tth hour belongs to the wth week, D h jt represents the average power load in the tth hour of historical year j, {D h jt , t=1,2,…8760} represents the 8760-hour power load curve of historical year j; Based on weekly electricity consumption E h jw Calculate the proportion of electricity consumption in week w to the annual electricity consumption h jw Specifically include: According to formula 2, calculate the proportion of electricity consumption in week w to the annual electricity consumption h jw : Based on N years of historical data, calculate the average proportion of electricity consumption in week w to annual electricity consumption and will The proportion of electricity consumption in the wth week of the target year to the annual electricity consumption specifically includes: Based on N years of historical data, the average proportion of electricity consumption in week w to annual electricity consumption is calculated based on formula 3. and will As the proportion of electricity consumption in the wth week of the target year to the annual electricity consumption: According to the target annual electricity consumption E f The simulated electricity consumption for each week of the target year includes: Based on Formula 4, the target annual electricity consumption E f Get the simulated electricity consumption for each week of the target year:

3. The method according to claim 1, characterized in that With the constraints that the seasonal imbalance coefficient is in a reasonable range, the maximum loads of each month satisfy a certain relationship, and the maximum load is consistent with the given boundary conditions, and the goal is to minimize the sum of the squares of the maximum load adjustments of each month, the monthly maximum load optimization model is established, which specifically includes: Determine the objective function according to formula 12: Among them, the maximum load of each month {D f(2) m , m=1,2,…12} are optimization variables; Determine the seasonal imbalance coefficient constraint according to formula 12: Among them, η min and η max are the minimum and maximum values ​​of the seasonal imbalance coefficient respectively; Determine the relationship constraints between the maximum loads of each month according to formula 14: in, Respectively represent the proportional relationship between the maximum loads of month m and m'; Determine the annual maximum load constraint according to formula 15: Among them, m * is the month in which the maximum load is expected to occur. is the expected annual maximum load.

4. The method according to claim 1, characterized in that: According to the difference in electricity consumption, the daily power load curve is corrected week by week and day by day, and the load and electricity relationship between the day and the hour is determined. The final annual power load curve simulated specifically includes: According to formula 16, calculate the current 8760-hour power load curve {D f(2) t , t=1,2,…8760}, the maximum load of each day {D f(2) d , d = 1, 2, ... 365}: Calculate weekly electricity consumption E according to formula 17 f(2) w : Calculate the daily electricity consumption E according to formula 18 f(2) d : According to formula 19, the adjustment target amount E of weekly electricity consumption in week w is calculated. f w With the current amount E f(2) w The difference E gap w : According to formula 20, the difference is divided into each day according to the ratio of daily electricity consumption, that is, the change of electricity consumption on day d in a week is E gap wd : If E gap w Greater than 0, increase the power consumption by E gap wd , the maximum load is still D f(2) d , the relative order of the load size of each hour remains unchanged, and the hourly load adjustment space D is calculated according to formula 21 and formula 22 δ t and daily electricity consumption adjustment space E δ t : Calculate the adjustment factor β according to Formula 23 and Formula 24 d and hourly load adjustment D gap t : Calculate the load D for each hour on day d according to formula 25 f* t : The load at hour t on day d is equal to the basic D f(2) t Superimposed load adjustment D gap t ; If E gap w Less than 0, power consumption increases by E gap wd , the maximum load is D f(2) d , the relative order of the load size of each hour remains unchanged, and the hourly load adjustment D is calculated according to formula 26 except for the time of the maximum load of the day. gap t : Divide the daily electricity consumption change evenly into 23 hours except the time of maximum daily load; Calculate the load D for each hour on day d according to formula 27 f* t : The load at hour t on day d after adjustment is equal to the basic D f(2) t Superimposed load adjustment D gap t , t max d The hour with the maximum load of the day; Corrected power load curve {D f(2) t , t∈w}, so that the weekly electricity consumption is equal to E f w , the maximum daily load is maintained at {D f(2) d , d∈w}; For the 8760-hour power load curve {D f(2) t , t=1,2,…8760} correct the daily power load curve week by week and day by day to obtain the final simulated 8760-hour power load curve {D f* t , t=1,2,…8760}.

5. A device for simulating the annual power load curve, characterized in that: include: The first determination module is used to simulate the power consumption of each week of the target year based on historical data and determine the power relationship between year, month and week; specifically, based on the data of the previous year and the next year of the historical year j, the first and last week of the year are supplemented with a full week starting from Monday and ending on Sunday, and the 8904-hour power load curve containing 53 full weeks of the historical year j is obtained, and the weekly power consumption E is calculated. h jw :Based on weekly electricity consumption E h jw Calculate the proportion of electricity consumption in week w to the annual electricity consumption h jw ; Based on N years of historical data, calculate the average proportion of electricity consumption in week w to annual electricity consumption and will As the proportion of electricity consumption in the wth week of the target year to the annual electricity consumption; According to the target annual electricity consumption E f Get the simulated electricity consumption for each week of the target year; The second determination module is used to select the typical weekly electricity curve of each month, determine the relationship between the electricity and load between the week and the day, and obtain the initial annual power load curve; specifically, for the mth month, determine all the historical weekly load curves corresponding to the mth month {D h jt ,t∈w|w∈m,j=1,2,…N}, for each historical weekly load curve, calculate the daily power consumption E according to Formula 5 h jwd : Among them, t∈d, d∈w represents all hours of the dth day of the wth week; The weekly power curve {E h jwd ,d=1,2,…7} takes the largest element as the reference value and converts it into {e h jwd ,d=1,2,…7}: Among them, max() is to find the maximum value; If there are n weekly electricity curves, the Euclidean distance between every two weekly electricity curves is calculated to form the Euclidean distance association matrix shown in Formula 7: Among them, d Euler is the Euclidean distance between the two curves; Calculate the Euclidean distance d between two curves based on their elements Euler ; According to Formula 8, the curve with the shortest average Euclidean distance to other curves is selected as the typical weekly power curve: The first correction module is used to establish an optimization model to optimize the maximum load of each month, determine the load relationship between the year, month and week, and correct the initial annual power load curve; specifically, it is used to calculate the proportion of each hour of the week to the weekly power consumption according to the typical weekly power consumption curve of each month. h wt , based on the simulated electricity consumption E in each week of the target year f w As well as the proportion of electricity consumption at each hour of each week to the weekly electricity consumption, the target year 8760-hour power load curve {D f(1) t |D f (1) t =e h wt *E f w , t∈w, t=1,2,…8760} and the maximum load of each month {D f(1) m , m=1,2,…12}: With the constraints that the seasonal imbalance coefficient is in a reasonable range, the maximum loads of each month satisfy a certain relationship, and the maximum load is consistent with the given boundary conditions, and the goal is to minimize the sum of the squares of the maximum load adjustments of each month, a monthly maximum load optimization model is established, and the new monthly maximum load {D f(2) m , m=1,2,…12}, before and after optimization, the magnification factor of the maximum load of each month is determined according to formula 10 {α m , m=1,2,…12}: According to the magnification factor, the load curves of each month are magnified in the same proportion, and the new 8760-hour power load curve {D f(2) t , t=1,2,…8760}: The second correction module is used to correct the daily power load curve week by week and day by day according to the power consumption difference, determine the load and power relationship between the day and the hour, and obtain the simulated final annual power load curve.

6. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for simulating the annual power load curve as claimed in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by a processor, the steps of the annual power load curve simulation method according to any one of claims 1 to 4 are implemented.

Images