Electricity consumption scheduling method and system based on user side load data

Abstract

The application discloses a power consumption scheduling method and system based on user-side load data. The method comprises the following steps: fitting processing based on user-side load data to obtain a first optimal parameter combination for indicating timing characteristics, wherein the first optimal parameter combination comprises a first periodic term parameter, a first trend term parameter and a first residual term parameter; generating a first power consumption curve matched with the user-side load data based on the first optimal parameter combination, wherein the first power consumption curve represents active power of the user-side in a preset time period; determining photovoltaic power of the user-side in the preset time period based on photovoltaic load data of the user-side; adjusting the first power consumption curve based on the photovoltaic power to obtain a second power consumption curve; and determining an optimized power consumption scheme according to the second power consumption curve of the user-side, so as to schedule the user-side according to the optimized power consumption scheme, which can realize the collaborative cooperation of the user-side in power consumption, thereby reducing the impact on the stability of the power grid.

Description

Technical Field

[0001] The present application relates to the technical field of power dispatching, and in particular to a power dispatching method and system based on user-side load data. Background Art

[0002] In existing technologies, electricity users typically operate independently. Consequently, their disorganized electricity consumption can easily cause abrupt fluctuations in grid load during certain time periods. For example, load fluctuations can be significant during periods of peak and trough electricity prices, significantly impacting the grid. Furthermore, an increasing number of electricity users are installing photovoltaic power generation systems to power themselves or even feed back into the grid. The instability of these systems can also impact grid stability. Summary of the Invention

[0003] In order to solve the above technical problems, the embodiments of the present application propose a power scheduling method and system based on user-side load data, which can achieve coordinated cooperation in power consumption at the user end, thereby reducing the impact on power grid stability.

[0004] In a first aspect, an embodiment of the present application provides a method for power scheduling based on user-side load data, comprising:

[0005] Performing fitting processing based on load data of the user terminal to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter;

[0006] Based on the first optimal parameter combination, generating a first power consumption curve that matches the load data of the user terminal, wherein the first power consumption curve represents the active power of the user terminal within a preset time period;

[0007] Determining the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal;

[0008] Adjusting the first power consumption curve based on the photovoltaic power to obtain a second power consumption curve;

[0009] An optimized power consumption plan is determined according to the second power consumption curve of the user terminal, so as to perform power consumption scheduling for the user terminal according to the optimized power consumption plan.

[0010] Optionally, performing fitting processing based on the load data of the user terminal to obtain a first optimal parameter combination for indicating the timing characteristics includes:

[0011] respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude;

[0012] The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination;

[0013] According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

[0014] Optionally, optimizing the first parameter combination to be fitted by using a grid search method according to the load data of the user terminal to fit the first optimal parameter combination includes:

[0015] Constructing a plurality of first parameter combination grids of the first parameter combination to be fitted using a grid search method;

[0016] Performing model training using each first parameter combination grid to obtain an evaluation value corresponding to each first parameter combination grid according to the model training result;

[0017] Based on the evaluation value corresponding to each first parameter combination grid and the load data of the user terminal, a least square method is used to determine the difference value corresponding to each first parameter combination grid;

[0018] Based on the difference values ​​corresponding to the plurality of first parameter combination grids, the first optimal parameter combination is determined in the plurality of first parameter combination grids.

[0019] Optionally, the first extended logistic regression function Determined by the following formula:

[0020]

[0021]

[0022]

[0023]

[0024] Where t represents the time, Represents the maximum asymptotic value function of the curve, a1 is the first asymptotic value parameter, b1 is the second asymptotic value parameter, c1 is the third asymptotic value parameter, d1 is the fourth asymptotic value parameter, Represents the curve growth rate function, a2 is the first growth rate parameter, b2 is the second growth rate parameter, c2 is the third growth rate parameter, d2 is the fourth growth rate parameter, represents the midpoint function of the curve, a3 is the first midpoint parameter, b3 is the second midpoint parameter, c3 is the third midpoint parameter, d3 is the fourth midpoint parameter, and e is the natural base;

[0025] The first trend item parameters to be fitted include a1, a2, a3, b1, b2, b3, c1, c2, c3, d1, d2 and d3.

[0026] Optionally, determining the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal includes:

[0027] respectively determining a second to-be-fitted period term parameter in the second Fourier series for indicating a period length, a second to-be-fitted trend term parameter in the second extended logistic regression function for indicating a function curve shape, and a second to-be-fitted residual term parameter in the second normal distribution function for indicating an amplitude;

[0028] The second to-be-fitted period item parameter, the second to-be-fitted trend item parameter and the second to-be-fitted residual item parameter form a second to-be-fitted parameter combination;

[0029] Constructing a plurality of second parameter combination grids of the second parameter combination to be fitted using a grid search method;

[0030] Performing model training using each second parameter combination grid to obtain an evaluation value corresponding to each second parameter combination grid according to the model training result;

[0031] Based on the evaluation value corresponding to each second parameter combination grid and the photovoltaic load data of the user end, a least squares method is used to determine the difference value corresponding to each second parameter combination grid;

[0032] Determining a second optimal parameter combination in the plurality of second parameter combination grids based on the difference values ​​corresponding to each of the plurality of second parameter combination grids;

[0033] Based on the second optimal parameter combination, a second power consumption curve matching the photovoltaic load data of the user terminal is generated, thereby determining the photovoltaic power of the user terminal within the preset time period indicated by the second power consumption curve.

[0034] Optionally, a plurality of second power consumption curves correspond one-to-one to a plurality of user terminals, and the second power consumption curves are used to represent power consumption demand information of the corresponding user terminals within the preset time period. Determining the optimized power consumption plan based on the second power consumption curves of the user terminals includes:

[0035] Determining, using a cyclic mobile plant operating condition algorithm, aggregated power demand information corresponding to each operating condition based on each of the second power consumption curves, wherein the aggregated power demand information represents the total power demand of each of the user terminals within the preset time period under the corresponding operating condition;

[0036] Based on the aggregated electricity demand information corresponding to each operating condition, a target operating condition is determined, so as to determine the optimized electricity consumption plan according to the target operating condition.

[0037] Optionally, one operating condition corresponds to one user terminal, the preset time period is a current cycle among a plurality of consecutive cycles, and the method of using the cyclic moving plant operating condition algorithm to determine the aggregated power demand information corresponding to each operating condition according to each second power consumption curve includes:

[0038] For each working condition:

[0039] Fixing the second power consumption curve corresponding to the operating condition in time sequence, and shifting other second power consumption curves in time sequence, wherein the other second power consumption curves are all second power consumption curves except the second power consumption curve corresponding to the operating condition;

[0040] Filling the gaps in the current cycle of the other second power consumption curves after translation according to the parts of the other second power consumption curves after translation that enter the next cycle or the previous cycle;

[0041] The fixed second power consumption curve and all other filled second power consumption curves are added together in time sequence to obtain a third power consumption curve;

[0042] Based on the third power consumption curve, aggregated power demand information corresponding to the operating condition is determined.

[0043] Optionally, determining the target operating condition based on the aggregated power demand information corresponding to each operating condition includes:

[0044] Calculate the total electricity cost for each operating condition based on the aggregated electricity demand information for each operating condition and the preset peak and valley electricity price table;

[0045] The operating condition with the minimum corresponding total electricity cost is taken as the target operating condition.

[0046] In a second aspect, an embodiment of the present application provides a power dispatching system based on user-side load data, including:

[0047] a first optimal parameter combination determining module, configured to perform fitting processing based on user-side load data to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter;

[0048] a first power usage curve generating module, configured to generate a first power usage curve matching the load data of the user terminal based on the first optimal parameter combination, wherein the first power usage curve represents the active power of the user terminal within a preset time period;

[0049] A photovoltaic power determination module, configured to determine the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal;

[0050] a second power usage curve generating module, configured to adjust the first power usage curve based on the photovoltaic power to obtain a second power usage curve;

[0051] The optimized power consumption plan determining module is used to determine the optimized power consumption plan according to the second power consumption curve of the user terminal, so as to perform power consumption scheduling for the user terminal according to the optimized power consumption plan.

[0052] Optionally, performing fitting processing based on the load data of the user terminal to obtain a first optimal parameter combination for indicating the timing characteristics includes:

[0053] respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude;

[0054] The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination;

[0055] According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

[0056] In summary, the embodiments of the present application have at least the following beneficial effects:

[0057] According to an embodiment of the present application, a first optimal parameter combination for indicating timing characteristics is obtained by performing fitting processing based on the load data of the user end, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter and a first residual item parameter; based on the first optimal parameter combination, a first power consumption curve matching the load data of the user end is generated, wherein the first power consumption curve represents the active power of the user end within a preset time period; based on the photovoltaic load data of the user end, the photovoltaic power of the user end within the preset time period is determined; based on the photovoltaic power, the first power consumption curve is adjusted to obtain a second power consumption curve; an optimized power consumption plan is determined according to the second power consumption curve of the user end, so as to schedule power consumption of the user end according to the optimized power consumption plan, so as to achieve coordinated cooperation of the user ends in power consumption, thereby reducing the impact on the stability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of a method for power dispatching based on user-side load data provided in an embodiment of the present application;

[0059] Figure 2 is a schematic diagram of a cyclic mobile plant operating condition algorithm provided in an embodiment of the present application;

[0060] Figure 3 This is a schematic diagram of the power corresponding to each user terminal provided in an embodiment of the present application;

[0061] Figure 4 This is a schematic diagram of the power corresponding to each user terminal provided in an embodiment of the present application;

[0062] Figure 5 This is a structural diagram of the power dispatching system based on user-side load data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0064] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of this application, unless otherwise specified, "multiple" means two or more. In the description of this application, the term "including" and its variations are open inclusions, i.e., "including but not limited to". The term "based on" means "at least partially based on". The term "according to" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments".

[0065] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0066] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meanings as those commonly understood by those skilled in the art. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit this application. Those of ordinary skill in the art will understand the specific meanings of the above terms in this application in specific circumstances.

[0067] First, see Figure 1 , shows a flow chart of a method for power scheduling based on user-side load data provided by an embodiment of the present application, the method including steps S101-S105, specifically as follows:

[0068] S101, performing fitting processing based on load data of a user terminal to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter;

[0069] It should be noted that the first period item parameter in this embodiment can be used to indicate the periodic characteristics of the time series, the first trend item parameter can be used to indicate the changing trend characteristics of the time series, and the first remainder item parameter can be used to indicate the residual of the time series.

[0070] In one example, fitting the load data of the user end to obtain the first optimal parameter combination for indicating the timing characteristics can include: using a decomposition method in time series analysis to decompose original time series data corresponding to the load data into several components, including a periodic term (characterized by a first periodic term parameter), a trend term (characterized by a first trend term parameter), and a residual term (characterized by a first residual term parameter), i.e., Yt=St+Tt+Rt, where Yt represents the time series, St represents the periodic term, Tt represents the trend term, and Rt represents the residual term.

[0071] S102, generating, based on the first optimal parameter combination, a first power consumption curve matching the load data of the user end, where the first power consumption curve characterizes the active power of the user end in a preset time period;

[0072] In one example, generating, based on the first optimal parameter combination, the first power consumption curve matching the load data of the user end can include: applying the first optimal parameter combination as a model parameter to a preset model (e.g., a time series model containing a trend, periodicity, and residual term, or a complex state space model), and inputting the load data of the user end into the preset model, so that the preset model can generate the first power consumption curve matching the load data of the user end.

[0073] S103, determining, based on the photovoltaic load data of the user end, a photovoltaic power of the user end in the preset time period;

[0074] It can be understood that the user end in the embodiment is pre-installed with a photovoltaic power generation device.

[0075] S104, adjusting the first power consumption curve based on the photovoltaic power to obtain a second power consumption curve;

[0076] In one example, the photovoltaic power in the first power consumption curve can be subtracted to obtain the second power consumption curve, so that the second power consumption curve can be used to indicate the amount of electricity required to be obtained from the power grid by the corresponding user end.

[0077] S105, determining an optimized power consumption scheme according to the second power consumption curve of the user end, and scheduling power consumption of the user end according to the optimized power consumption scheme.

[0078] In an optional embodiment, the fitting of the load data of the user end to obtain the first optimal parameter combination for indicating the timing characteristics includes:

[0079] respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude;

[0080] The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination;

[0081] According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

[0082] In some specific examples, the first Fourier series can be determined by the following formula:

[0083]

[0084] Where t represents the time, represents a periodic function, N is the first periodic term parameter to be fitted, represents the first Fourier coefficient, represents the second Fourier coefficient, n is the order of the harmonic, and P is the period.

[0085] The first normal distribution function can be determined by the following formula: ,in, is the first remaining parameter to be fitted.

[0086] In an optional embodiment, optimizing the first parameter combination to be fitted using a grid search method based on the load data of the user terminal to fit the first optimal parameter combination includes:

[0087] Constructing a plurality of first parameter combination grids of the first parameter combination to be fitted using a grid search method;

[0088] Performing model training using each first parameter combination grid to obtain an evaluation value corresponding to each first parameter combination grid according to the model training result;

[0089] Based on the evaluation value corresponding to each first parameter combination grid and the load data of the user terminal, a least square method is used to determine the difference value corresponding to each first parameter combination grid;

[0090] Based on the difference values ​​corresponding to the plurality of first parameter combination grids, the first optimal parameter combination is determined in the plurality of first parameter combination grids.

[0091] In one example, the plurality of first parameter combination grids may be sorted according to the corresponding difference values, so as to screen out the first optimal parameter combination from the plurality of first parameter combination grids according to the sorting result.

[0092] In an optional embodiment, the first extended logistic regression function Determined by the following formula:

[0093]

[0094]

[0095]

[0096]

[0097] Where t represents the time, Represents the maximum asymptotic value function of the curve, a1 is the first asymptotic value parameter, b1 is the second asymptotic value parameter, c1 is the third asymptotic value parameter, d1 is the fourth asymptotic value parameter, Represents the curve growth rate function, a2 is the first growth rate parameter, b2 is the second growth rate parameter, c2 is the third growth rate parameter, d2 is the fourth growth rate parameter, represents the midpoint function of the curve, a3 is the first midpoint parameter, b3 is the second midpoint parameter, c3 is the third midpoint parameter, d3 is the fourth midpoint parameter, and e is the natural base;

[0098] The first trend item parameters to be fitted include a1, a2, a3, b1, b2, b3, c1, c2, c3, d1, d2 and d3.

[0099] It should be noted that the first extended logistic regression function is given below The derivation process.

[0100] General logistic regression function for:

[0101]

[0102] Then, its derivative is:

[0103]

[0104] And there are and .

[0105] At this time, the logistic regression function is expanded, the maximum asymptotic value C of the curve, the growth rate k of the curve, and the midpoint m of the curve are added, and the logistic regression formula is rewritten as follows:

[0106]

[0107] In this way, C, k, and m are defined as time-dependent functions.

[0108] In an optional implementation manner, determining the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal includes:

[0109] respectively determining a second to-be-fitted period term parameter in the second Fourier series for indicating a period length, a second to-be-fitted trend term parameter in the second extended logistic regression function for indicating a function curve shape, and a second to-be-fitted residual term parameter in the second normal distribution function for indicating an amplitude;

[0110] The second to-be-fitted period item parameter, the second to-be-fitted trend item parameter and the second to-be-fitted residual item parameter form a second to-be-fitted parameter combination;

[0111] Constructing a plurality of second parameter combination grids of the second parameter combination to be fitted using a grid search method;

[0112] Performing model training using each second parameter combination grid to obtain an evaluation value corresponding to each second parameter combination grid according to the model training result;

[0113] Based on the evaluation value corresponding to each second parameter combination grid and the photovoltaic load data of the user end, a least squares method is used to determine the difference value corresponding to each second parameter combination grid;

[0114] Determining a second optimal parameter combination in the plurality of second parameter combination grids based on the difference values ​​corresponding to each of the plurality of second parameter combination grids;

[0115] Based on the second optimal parameter combination, a second power consumption curve matching the photovoltaic load data of the user terminal is generated, thereby determining the photovoltaic power of the user terminal within the preset time period indicated by the second power consumption curve.

[0116] In an optional embodiment, a plurality of second power usage curves correspond one-to-one to a plurality of user terminals, and the second power usage curves are used to represent power demand information of the corresponding user terminals within the preset time period. Determining the optimized power usage plan based on the second power usage curves of the user terminals includes:

[0117] Determining, using a cyclic mobile plant operating condition algorithm, aggregated power demand information corresponding to each operating condition based on each of the second power consumption curves, wherein the aggregated power demand information represents the total power demand of each of the user terminals within the preset time period under the corresponding operating condition;

[0118] Based on the aggregated electricity demand information corresponding to each operating condition, a target operating condition is determined, so as to determine the optimized electricity consumption plan according to the target operating condition.

[0119] In one example, the total power demand power indicated can be sorted according to its peak value, and the operating condition with the smallest peak value of the total power demand power can be used as the target operating condition. In this way, the impact of power consumption at each user end on the power grid can be effectively avoided, thereby improving the stability of the power grid.

[0120] In an optional embodiment, one operating condition corresponds to one user terminal, the preset time period is a current cycle among a plurality of consecutive cycles, and the determining of the aggregated power demand information corresponding to each operating condition based on each second power consumption curve using the cyclic moving plant operating condition algorithm includes:

[0121] For each working condition:

[0122] Fixing the second power consumption curve corresponding to the operating condition in time sequence, and shifting other second power consumption curves in time sequence, wherein the other second power consumption curves are all second power consumption curves except the second power consumption curve corresponding to the operating condition;

[0123] Filling the gaps in the current cycle of the other second power consumption curves after translation according to the parts of the other second power consumption curves after translation that enter the next cycle or the previous cycle;

[0124] The fixed second power consumption curve and all other filled second power consumption curves are added together in time sequence to obtain a third power consumption curve;

[0125] Based on the third power consumption curve, aggregated power demand information corresponding to the operating condition is determined.

[0126] In some cases, see Figure 2 , Figure 2 The curves on the left side of the figure are all second power consumption curves. You can select any one of them and fix it, and shift the other second power consumption curves in time sequence (i.e., shift left and right in the figure). It is not difficult to understand that due to periodicity, the overflow part can be directly used to fill the gap. For example, for the curve shifted to the right, its overflow part is shifted from the leftmost side of the figure to the right to fill the gap. For further information, please refer to Figure 3 and Figure 4 ,in Figure 3 is the power corresponding to each user terminal before applying the cyclic mobile plant operating condition algorithm (the total power is the total power demand), Figure 4 is the power corresponding to each user end after applying the cyclic mobile plant operating condition algorithm. It can be seen that the total power peak is significantly suppressed.

[0127] In an optional embodiment, determining the target operating condition based on the aggregated power demand information corresponding to each operating condition includes:

[0128] Calculate the total electricity cost for each operating condition based on the aggregated electricity demand information for each operating condition and the preset peak and valley electricity price table;

[0129] The operating condition with the minimum corresponding total electricity cost is taken as the target operating condition.

[0130] Correspondingly, an embodiment of the present application also provides an electricity scheduling system based on user-side load data, which can implement all processes of the electricity scheduling method based on user-side load data provided in the above embodiment.

[0131] See also Figure 5 , shows a schematic structural diagram of a power dispatching system based on user-side load data provided by an embodiment of the present application, the system comprising:

[0132] A first optimal parameter combination determining module 501 is configured to perform fitting processing based on user-side load data to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter;

[0133] A first power usage curve generating module 502 is configured to generate a first power usage curve matching the load data of the user terminal based on the first optimal parameter combination, wherein the first power usage curve represents the active power of the user terminal within a preset time period;

[0134] The photovoltaic power determination module 503 is configured to determine the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal;

[0135] A second power usage curve generating module 504 is configured to adjust the first power usage curve based on the photovoltaic power to obtain a second power usage curve;

[0136] The optimized power consumption plan determining module 505 is configured to determine an optimized power consumption plan according to the second power consumption curve of the user terminal, so as to perform power consumption scheduling for the user terminal according to the optimized power consumption plan.

[0137] In an optional implementation, performing fitting processing based on the load data of the user terminal to obtain a first optimal parameter combination for indicating the timing characteristics includes:

[0138] respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude;

[0139] The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination;

[0140] According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

[0141] In an optional embodiment, optimizing the first parameter combination to be fitted using a grid search method based on the load data of the user terminal to fit the first optimal parameter combination includes:

[0142] Constructing a plurality of first parameter combination grids of the first parameter combination to be fitted using a grid search method;

[0143] Performing model training using each first parameter combination grid to obtain an evaluation value corresponding to each first parameter combination grid according to the model training result;

[0144] Based on the evaluation value corresponding to each first parameter combination grid and the load data of the user terminal, a least square method is used to determine the difference value corresponding to each first parameter combination grid;

[0145] Based on the difference values ​​corresponding to the plurality of first parameter combination grids, the first optimal parameter combination is determined in the plurality of first parameter combination grids.

[0146] In an optional embodiment, the first extended logistic regression function Determined by the following formula:

[0147]

[0148]

[0149]

[0150]

[0151] Where t represents the time, Represents the maximum asymptotic value function of the curve, a1 is the first asymptotic value parameter, b1 is the second asymptotic value parameter, c1 is the third asymptotic value parameter, d1 is the fourth asymptotic value parameter, Represents the curve growth rate function, a2 is the first growth rate parameter, b2 is the second growth rate parameter, c2 is the third growth rate parameter, d2 is the fourth growth rate parameter, represents the midpoint function of the curve, a3 is the first midpoint parameter, b3 is the second midpoint parameter, c3 is the third midpoint parameter, d3 is the fourth midpoint parameter, and e is the natural base;

[0152] The first trend item parameters to be fitted include a1, a2, a3, b1, b2, b3, c1, c2, c3, d1, d2 and d3.

[0153] In an optional implementation manner, determining the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal includes:

[0154] respectively determining a second to-be-fitted period term parameter in the second Fourier series for indicating a period length, a second to-be-fitted trend term parameter in the second extended logistic regression function for indicating a function curve shape, and a second to-be-fitted residual term parameter in the second normal distribution function for indicating an amplitude;

[0155] The second to-be-fitted period item parameter, the second to-be-fitted trend item parameter and the second to-be-fitted residual item parameter form a second to-be-fitted parameter combination;

[0156] Constructing a plurality of second parameter combination grids of the second parameter combination to be fitted using a grid search method;

[0157] Performing model training using each second parameter combination grid to obtain an evaluation value corresponding to each second parameter combination grid according to the model training result;

[0158] Based on the evaluation value corresponding to each second parameter combination grid and the photovoltaic load data of the user end, a least squares method is used to determine the difference value corresponding to each second parameter combination grid;

[0159] Determining a second optimal parameter combination in the plurality of second parameter combination grids based on the difference values ​​corresponding to each of the plurality of second parameter combination grids;

[0160] Based on the second optimal parameter combination, a second power consumption curve matching the photovoltaic load data of the user terminal is generated, thereby determining the photovoltaic power of the user terminal within the preset time period indicated by the second power consumption curve.

[0161] In an optional embodiment, a plurality of second power usage curves correspond one-to-one to a plurality of user terminals, and the second power usage curves are used to represent power demand information of the corresponding user terminals within the preset time period. Determining the optimized power usage plan based on the second power usage curves of the user terminals includes:

[0162] Determining, using a cyclic mobile plant operating condition algorithm, aggregated power demand information corresponding to each operating condition based on each of the second power consumption curves, wherein the aggregated power demand information represents the total power demand of each of the user terminals within the preset time period under the corresponding operating condition;

[0163] Based on the aggregated electricity demand information corresponding to each operating condition, a target operating condition is determined, so as to determine the optimized electricity consumption plan according to the target operating condition.

[0164] In an optional embodiment, one operating condition corresponds to one user terminal, the preset time period is a current cycle among a plurality of consecutive cycles, and the determining of the aggregated power demand information corresponding to each operating condition based on each second power consumption curve using the cyclic moving plant operating condition algorithm includes:

[0165] For each working condition:

[0166] Fixing the second power consumption curve corresponding to the operating condition in time sequence, and shifting other second power consumption curves in time sequence, wherein the other second power consumption curves are all second power consumption curves except the second power consumption curve corresponding to the operating condition;

[0167] Filling the gaps in the current cycle of the other second power consumption curves after translation according to the parts of the other second power consumption curves after translation that enter the next cycle or the previous cycle;

[0168] The fixed second power consumption curve and all other filled second power consumption curves are added together in time sequence to obtain a third power consumption curve;

[0169] Based on the third power consumption curve, aggregated power demand information corresponding to the operating condition is determined.

[0170] In an optional embodiment, determining the target operating condition based on the aggregated power demand information corresponding to each operating condition includes:

[0171] Calculate the total electricity cost for each operating condition based on the aggregated electricity demand information for each operating condition and the preset peak and valley electricity price table;

[0172] The operating condition with the minimum corresponding total electricity cost is taken as the target operating condition.

[0173] In summary, the embodiments of the present application have at least the following beneficial effects:

[0174] According to an embodiment of the present application, a first optimal parameter combination for indicating timing characteristics is obtained by performing fitting processing based on the load data of the user end, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter and a first residual item parameter; based on the first optimal parameter combination, a first power consumption curve matching the load data of the user end is generated, wherein the first power consumption curve represents the active power of the user end within a preset time period; based on the photovoltaic load data of the user end, the photovoltaic power of the user end within the preset time period is determined; based on the photovoltaic power, the first power consumption curve is adjusted to obtain a second power consumption curve; an optimized power consumption plan is determined according to the second power consumption curve of the user end, so as to schedule power consumption of the user end according to the optimized power consumption plan, so as to achieve coordinated cooperation of the user ends in power consumption, thereby reducing the impact on the stability of the power grid.

[0175] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus the necessary hardware platform, and of course, it can also be implemented entirely through hardware. Based on this understanding, all or part of the contribution of the technical solution of the present application to the background technology can be embodied in the form of a software product. The computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present application.

[0176] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A power dispatching method based on user-side load data, characterized in that: include: Performing fitting processing based on load data of a user terminal pre-installed with a photovoltaic power generation device to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter; Based on the first optimal parameter combination, generating a first power consumption curve that matches the load data of the user terminal, wherein the first power consumption curve represents the active power of the user terminal within a preset time period; Determining the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal; Adjusting the first power consumption curve based on the photovoltaic power to obtain a second power consumption curve, wherein the second power consumption curve is used to indicate the amount of power that the user terminal needs to obtain from the power grid; determining an optimized power consumption plan according to the second power consumption curve of the user terminal, so as to perform power consumption scheduling for the user terminal according to the optimized power consumption plan; The plurality of second power consumption curves correspond one-to-one to the plurality of user terminals, and the second power consumption curves are used to represent the power demand information of the corresponding user terminals within the preset time period. The determining of the optimized power consumption plan based on the second power consumption curves of the user terminals includes: Determining, using a cyclic mobile plant operating condition algorithm, aggregated power demand information corresponding to each operating condition based on each of the second power consumption curves, wherein the aggregated power demand information represents the total power demand of each of the user terminals within the preset time period under the corresponding operating condition; Determining a target operating condition based on the aggregated power demand information corresponding to each operating condition, and determining the optimized power consumption plan according to the target operating condition; Wherein, one operating condition corresponds to one user terminal, the preset time period is a current cycle among a plurality of consecutive cycles, and the method of using the cyclic moving plant operating condition algorithm to determine the aggregated power demand information corresponding to each operating condition according to each second power consumption curve includes: For each working condition: Fixing the second power consumption curve corresponding to the operating condition in time sequence, and shifting other second power consumption curves in time sequence, wherein the other second power consumption curves are all second power consumption curves except the second power consumption curve corresponding to the operating condition; Filling the gaps in the current cycle of the other second power consumption curves after translation according to the parts of the other second power consumption curves after translation that enter the next cycle or the previous cycle; The fixed second power consumption curve and all other filled second power consumption curves are added together in time sequence to obtain a third power consumption curve; Determining aggregated electricity demand information corresponding to the operating condition based on the third electricity consumption curve; The determining of the target operating condition based on the aggregated power demand information corresponding to each operating condition includes: Calculate the total electricity cost for each operating condition based on the aggregated electricity demand information for each operating condition and the preset peak and valley electricity price table; The operating condition with the minimum corresponding total electricity cost is taken as the target operating condition.

2. The power dispatching method based on user-side load data according to claim 1, characterized in that: The fitting process is performed based on the load data of the user terminal pre-installed with the photovoltaic power generation device to obtain a first optimal parameter combination for indicating the timing characteristics, including: respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude; The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination; According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

3. The power dispatching method based on user-side load data according to claim 2, characterized in that: The step of optimizing the first parameter combination to be fitted by using a grid search method according to the load data of the user terminal to fit the first optimal parameter combination includes: Constructing a plurality of first parameter combination grids of the first parameter combination to be fitted using a grid search method; Performing model training using each first parameter combination grid to obtain an evaluation value corresponding to each first parameter combination grid according to the model training result; Based on the evaluation value corresponding to each first parameter combination grid and the load data of the user terminal, a least square method is used to determine the difference value corresponding to each first parameter combination grid; Based on the difference values ​​corresponding to the plurality of first parameter combination grids, the first optimal parameter combination is determined in the plurality of first parameter combination grids.

4. The power dispatching method based on user-side load data according to claim 1, characterized in that: The determining, based on the photovoltaic load data of the user terminal, the photovoltaic power of the user terminal within the preset time period includes: respectively determining a second to-be-fitted period term parameter in the second Fourier series for indicating a period length, a second to-be-fitted trend term parameter in the second extended logistic regression function for indicating a function curve shape, and a second to-be-fitted residual term parameter in the second normal distribution function for indicating an amplitude; The second to-be-fitted period item parameter, the second to-be-fitted trend item parameter and the second to-be-fitted residual item parameter form a second to-be-fitted parameter combination; Constructing a plurality of second parameter combination grids of the second parameter combination to be fitted using a grid search method; Performing model training using each second parameter combination grid to obtain an evaluation value corresponding to each second parameter combination grid according to the model training result; Based on the evaluation value corresponding to each second parameter combination grid and the photovoltaic load data of the user end, a least squares method is used to determine the difference value corresponding to each second parameter combination grid; Determining a second optimal parameter combination in the plurality of second parameter combination grids based on the difference values ​​corresponding to each of the plurality of second parameter combination grids; Based on the second optimal parameter combination, a second power consumption curve matching the photovoltaic load data of the user terminal is generated, thereby determining the photovoltaic power of the user terminal within the preset time period indicated by the second power consumption curve.

5. The power dispatching method based on user-side load data according to claim 2, characterized in that: The first extended logistic regression function Determined by the following formula: Where t represents the time, Represents the maximum asymptotic value function of the curve, a1 is the first asymptotic value parameter, b1 is the second asymptotic value parameter, c1 is the third asymptotic value parameter, d1 is the fourth asymptotic value parameter, Represents the curve growth rate function, a2 is the first growth rate parameter, b2 is the second growth rate parameter, c2 is the third growth rate parameter, d2 is the fourth growth rate parameter, represents the midpoint function of the curve, a3 is the first midpoint parameter, b3 is the second midpoint parameter, c3 is the third midpoint parameter, d3 is the fourth midpoint parameter, and e is the natural base; The first trend item parameters to be fitted include a1, a2, a3, b1, b2, b3, c1, c2, c3, d1, d2 and d3.

6. A power dispatching system based on user-side load data, characterized in that: include: A first optimal parameter combination determination module is configured to perform fitting processing based on load data of a user terminal pre-installed with a photovoltaic power generation device to obtain a first optimal parameter combination for indicating a time series characteristic, wherein the first optimal parameter combination includes a first period item parameter, a first trend item parameter, and a first residual item parameter; a first power usage curve generating module, configured to generate a first power usage curve matching the load data of the user terminal based on the first optimal parameter combination, wherein the first power usage curve represents the active power of the user terminal within a preset time period; A photovoltaic power determination module, configured to determine the photovoltaic power of the user terminal within the preset time period based on the photovoltaic load data of the user terminal; a second power consumption curve generating module, configured to adjust the first power consumption curve based on the photovoltaic power to obtain a second power consumption curve, wherein the second power consumption curve is used to indicate the amount of power that the user terminal needs to obtain from the power grid; an optimized power consumption plan determining module, configured to determine an optimized power consumption plan according to the second power consumption curve of the user terminal, so as to perform power consumption scheduling for the user terminal according to the optimized power consumption plan; The plurality of second power consumption curves correspond one-to-one to the plurality of user terminals, and the second power consumption curves are used to represent the power demand information of the corresponding user terminals within the preset time period. The determining of the optimized power consumption plan based on the second power consumption curves of the user terminals includes: Determining, using a cyclic mobile plant operating condition algorithm, aggregated power demand information corresponding to each operating condition based on each of the second power consumption curves, wherein the aggregated power demand information represents the total power demand of each of the user terminals within the preset time period under the corresponding operating condition; Determining a target operating condition based on the aggregated power demand information corresponding to each operating condition, and determining the optimized power consumption plan according to the target operating condition; Wherein, one operating condition corresponds to one user terminal, the preset time period is a current cycle among a plurality of consecutive cycles, and the method of using the cyclic moving plant operating condition algorithm to determine the aggregated power demand information corresponding to each operating condition according to each second power consumption curve includes: For each working condition: Fixing the second power consumption curve corresponding to the operating condition in time sequence, and shifting other second power consumption curves in time sequence, wherein the other second power consumption curves are all second power consumption curves except the second power consumption curve corresponding to the operating condition; Filling the gaps in the current cycle of the other second power consumption curves after translation according to the parts of the other second power consumption curves after translation that enter the next cycle or the previous cycle; The fixed second power consumption curve and all other filled second power consumption curves are added together in time sequence to obtain a third power consumption curve; Determining aggregated electricity demand information corresponding to the operating condition based on the third electricity consumption curve; The determining of the target operating condition based on the aggregated power demand information corresponding to each operating condition includes: Calculate the total electricity cost for each operating condition based on the aggregated electricity demand information for each operating condition and the preset peak and valley electricity price table; The operating condition with the minimum corresponding total electricity cost is taken as the target operating condition.

7. The power dispatching system based on user-side load data according to claim 6, characterized in that: The fitting process is performed based on the load data of the user terminal pre-installed with the photovoltaic power generation device to obtain a first optimal parameter combination for indicating the timing characteristics, including: respectively determining a first to-be-fitted period term parameter in the first Fourier series for indicating a period length, a first to-be-fitted trend term parameter in the first extended logistic regression function for indicating a function curve shape, and a first to-be-fitted residual term parameter in the first normal distribution function for indicating an amplitude; The first to-be-fitted period item parameter, the first to-be-fitted trend item parameter and the first to-be-fitted residual item parameter form a first to-be-fitted parameter combination; According to the load data of the user terminal, the first parameter combination to be fitted is optimized using a grid search method to fit the first optimal parameter combination.

8. The power dispatching system based on user-side load data according to claim 7, characterized in that: The step of optimizing the first parameter combination to be fitted by using a grid search method according to the load data of the user terminal to fit the first optimal parameter combination includes: Constructing a plurality of first parameter combination grids of the first parameter combination to be fitted using a grid search method; Performing model training using each first parameter combination grid to obtain an evaluation value corresponding to each first parameter combination grid according to the model training result; Based on the evaluation value corresponding to each first parameter combination grid and the load data of the user terminal, a least square method is used to determine the difference value corresponding to each first parameter combination grid; Based on the difference values ​​corresponding to the plurality of first parameter combination grids, the first optimal parameter combination is determined in the plurality of first parameter combination grids.

9. The power dispatching system based on user-side load data according to claim 6, characterized in that: The determining, based on the photovoltaic load data of the user terminal, the photovoltaic power of the user terminal within the preset time period includes: respectively determining a second to-be-fitted period term parameter in the second Fourier series for indicating a period length, a second to-be-fitted trend term parameter in the second extended logistic regression function for indicating a function curve shape, and a second to-be-fitted residual term parameter in the second normal distribution function for indicating an amplitude; The second to-be-fitted period item parameter, the second to-be-fitted trend item parameter and the second to-be-fitted residual item parameter form a second to-be-fitted parameter combination; Constructing a plurality of second parameter combination grids of the second parameter combination to be fitted using a grid search method; Performing model training using each second parameter combination grid to obtain an evaluation value corresponding to each second parameter combination grid according to the model training result; Based on the evaluation value corresponding to each second parameter combination grid and the photovoltaic load data of the user end, a least squares method is used to determine the difference value corresponding to each second parameter combination grid; Determining a second optimal parameter combination in the plurality of second parameter combination grids based on the difference values ​​corresponding to each of the plurality of second parameter combination grids; Based on the second optimal parameter combination, a second power consumption curve matching the photovoltaic load data of the user terminal is generated, thereby determining the photovoltaic power of the user terminal within the preset time period indicated by the second power consumption curve.

Images