Energy data aggregation method and device

Abstract

The present invention discloses a method and device for aggregating energy data, the method comprising: obtaining renewable energy data and energy data to be tested, establishing individual physical models and aggregate cluster models of renewable energy based on the renewable energy data, dividing the renewable energy data into training set data and test set data, training and verifying the individual physical model and the aggregate cluster model based on the K-means clustering algorithm, the training set data and the test set data, obtaining target individual physical models and target aggregate cluster models, inputting the energy data to be tested into the target individual physical model and the target aggregate cluster model, and obtaining predicted aggregate index data of the energy data to be tested. The method is conducive to solving the technical problem that it is difficult to aggregate socially idle distributed renewable resources, and unifies the access standards for renewable resources.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed energy in power systems, and in particular to a method and device for aggregating energy data. Background Art

[0002] As society's demand for energy increases, the proposal of the "dual carbon target" promotes the transformation of energy structure, and distributed renewable energy has also ushered in massive growth. Massive distributed renewable resources of various types and structures, including distributed photovoltaics, distributed wind turbines, distributed energy storage, electric vehicles, and flexible loads, provide economical clean energy for the power grid, and can give full play to the unique advantages of wind and solar resources in different regions according to local conditions, and can also ensure the safe and economic operation of the power grid through peak shaving, frequency modulation, and demand response. At present, the proportion of massive distributed renewable energy in energy is rapidly increasing, with huge quantities and a wide variety. However, the geographical distribution is scattered, the affiliation is not unified, and its power generation characteristics are uncertain due to climate and human factors. If it is directly connected to the power grid in an unordered manner, on the one hand, it will not only cause management difficulties and bring shocks to the power grid, but also cause waste of idle social resources on the other hand, and cannot guarantee the full absorption of renewable resources. In view of this, combined with the characteristics of small size, scattered distribution, and different affiliations of massive distributed renewable resources, they can be aggregated through some intermediate institutions for unified access, thereby enhancing the flexibility of the power system.

[0003] For the aggregation and access of massive distributed renewable resources, the current mainstream practice at home and abroad is mainly to access through aggregators, virtual power plants, etc. After aggregation, they are regarded as "quasi-conventional power sources" to participate in the overall response and dispatch instructions of the day-ahead plan. For the effective use of various types of distributed renewable energy, it is necessary to establish a standardized model, which not only meets the requirements of shielding the underlying physical characteristics of various resources to ensure the external controllability of open attributes, but also meets the heterogeneous characteristics of various resources themselves.

[0004] At present, most of the distributed renewable energy sources connected to the power grid are large-scale wind farms and photovoltaic units, and there is a lack of effective means to utilize the huge amount of idle distributed resources in society.

[0005] Therefore, in order to unify the access standards of renewable resources and solve the current technical problem that it is difficult to aggregate idle and distributed renewable resources in society, it is urgent to build a method for aggregating energy data. Summary of the invention

[0006] The present invention provides a method and device for aggregating energy data, which solves the current technical problem that it is difficult to aggregate the idle distributed renewable resources in society.

[0007] In a first aspect, the present invention provides a method for aggregating energy data, comprising:

[0008] Obtain renewable energy data and energy data to be measured;

[0009] Based on the renewable energy data, establishing individual physical models and aggregate cluster models of renewable energy;

[0010] Dividing the renewable energy data into training set data and test set data;

[0011] Based on the K-means clustering algorithm, the training set data and the test set data, the individual physical model and the aggregate cluster model are trained and verified to obtain a target individual physical model and a target aggregate cluster model;

[0012] The energy data to be measured is input into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be measured.

[0013] Optionally, obtaining renewable energy data and energy data to be measured includes:

[0014] Obtaining initial data on renewable energy;

[0015] The renewable energy initial data is cleaned and repaired to obtain the renewable energy data and the energy data to be tested.

[0016] Optionally, based on a K-means clustering algorithm, the training set data and the test set data, the individual physical model and the aggregate cluster model are trained and verified to obtain a target individual physical model and a target aggregate cluster model, including:

[0017] Using the K-means clustering algorithm, combined with the training set data, the individual physical model and the aggregate cluster model are trained to obtain a trained aggregate cluster model and a trained aggregate cluster model;

[0018] Based on the test set data, the trained aggregate cluster model and the trained aggregate cluster model are verified to obtain the target individual physical model and the target aggregate cluster model.

[0019] Optionally, the K-means clustering algorithm is used in combination with the training set data to train the individual physical model and the aggregate cluster model to obtain a trained aggregate cluster model and a trained aggregate cluster model, including:

[0020] Inputting the training set data into the individual physical model and the aggregate cluster model to obtain the predicted aggregate value of the corresponding renewable energy data;

[0021] Determining a training error according to the data labels corresponding to the training set data and the predicted aggregate value;

[0022] Based on the training error, the individual physical model and the aggregate cluster model are adjusted through the K-means clustering algorithm to obtain optimal parameters, and the individual physical model and the aggregate cluster model are optimized using the optimal parameters to obtain the trained aggregate cluster model and the trained aggregate cluster model.

[0023] Optionally, before inputting the training set data into the individual physical model and the aggregate cluster model to obtain the corresponding predicted aggregate value of renewable energy data, the method further includes:

[0024] Initialize the parameters of the individual physical model and the parameters of the aggregate cluster model.

[0025] In a second aspect, the present invention provides an energy data aggregation device, comprising:

[0026] An acquisition module, used for acquiring renewable energy data and energy data to be tested;

[0027] An establishment module is used to establish an individual physical model and an aggregate cluster model of renewable energy based on the renewable energy data;

[0028] A partitioning module, used for partitioning the renewable energy data into training set data and test set data;

[0029] A training module, used for training and verifying the individual physical model and the aggregate cluster model based on a K-means clustering algorithm, the training set data and the test set data, to obtain a target individual physical model and a target aggregate cluster model;

[0030] The indicator module is used to input the energy data to be measured into the target individual physical model and the target aggregate cluster model to obtain predicted aggregate indicator data of the energy data to be measured.

[0031] Optionally, the acquisition module includes:

[0032] The acquisition submodule is used to obtain the initial data of renewable energy;

[0033] The repair submodule is used to clean and repair the renewable energy initial data to obtain the renewable energy data and the energy data to be tested.

[0034] Optionally, the training module includes:

[0035] A training submodule, for training the individual physical model and the aggregate cluster model using the K-means clustering algorithm in combination with the training set data, to obtain a trained aggregate cluster model and a trained aggregate cluster model;

[0036] The verification submodule is used to verify the trained aggregation cluster model and the trained aggregation cluster model based on the test set data to obtain the target individual physical model and the target aggregation cluster model.

[0037] Optionally, the training submodule includes:

[0038] A prediction unit, used to input the training set data into the individual physical model and the aggregate cluster model to obtain a predicted aggregate value of the corresponding renewable energy data;

[0039] An error unit, used to determine a training error according to a data label corresponding to the training set data and the predicted aggregate value;

[0040] An optimization unit is used to adjust the individual physical model and the aggregate cluster model based on the training error through the K-means clustering algorithm to obtain optimal parameters, and use the optimal parameters to optimize the individual physical model and the aggregate cluster model to obtain the trained aggregate cluster model and the trained aggregate cluster model.

[0041] Optionally, the training submodule further includes:

[0042] A parameter unit is used to initialize the parameters of the individual physical model and the parameters of the aggregate cluster model.

[0043] It can be seen from the above technical scheme that the present invention has the following advantages: the present invention provides a method for aggregating energy data, by acquiring renewable energy data and energy data to be tested, establishing an individual physical model and an aggregate cluster model of renewable energy based on the renewable energy data, dividing the renewable energy data into training set data and test set data, training and verifying the individual physical model and the aggregate cluster model based on the K-means clustering algorithm, the training set data and the test set data, obtaining a target individual physical model and a target aggregate cluster model, inputting the energy data to be tested into the target individual physical model and the target aggregate cluster model, obtaining predicted aggregate index data of the energy data to be tested, and solving the technical problem that it is difficult to aggregate the currently existing idle distributed renewable resources in society through a method for aggregating energy data, and unifying the access standards for renewable resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0045] Figure 1 A flowchart of a first embodiment of a method for aggregating energy data according to the present invention;

[0046] Figure 2 This is a flow chart of a second embodiment of a method for aggregating energy data according to the present invention;

[0047] Figure 3 Update the flow chart for a standardized model of renewable energy of the present invention;

[0048] Figure 4 This is a structural block diagram of an embodiment of an energy data aggregation device of the present invention. DETAILED DESCRIPTION

[0049] The embodiments of the present invention provide a method and device for aggregating energy data, which are used to solve the current technical problem that it is difficult to aggregate the idle distributed renewable resources in society.

[0050] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] For example, see Figure 1 , Figure 1 The process step diagram of Embodiment 1 of an energy data aggregation method of the present invention includes:

[0052] Step S101, obtaining renewable energy data and energy data to be tested;

[0053] It should be noted that energy data includes individual photovoltaic array data, wind turbine data, distributed energy storage data, electric vehicle data, flexible load data, as well as intra-cluster characteristic characterization data and extra-cluster characteristic characterization data.

[0054] The internal characteristic characterization data of the cluster include capacity support capability (including baseline value and upper and lower boundaries), power support capability (including baseline value and upper and lower boundaries), dynamic response capability and effective convergence time ratio.

[0055] The external cluster characterization data include stability factor, reliability factor, network loss factor, unit regulation power and dynamic sensitivity factor.

[0056] In the embodiment of the present invention, the renewable energy initial data is acquired, and the renewable energy initial data is cleaned and repaired to obtain the renewable energy data and the energy data to be tested.

[0057] Step S102, establishing an individual physical model and an aggregate cluster model of renewable energy based on the renewable energy data;

[0058] It should be noted that the individual physical models of renewable energy include individual photovoltaic array models, wind turbine models, distributed energy storage models, electric vehicle models and flexibility load models.

[0059] In the embodiment of the present invention, an individual physical model and an aggregate cluster model of renewable energy are established according to the renewable energy data.

[0060] Step S103, dividing the renewable energy data into training set data and test set data;

[0061] Step S104, based on the K-means clustering algorithm, the training set data and the test set data, the individual physical model and the aggregate cluster model are trained and verified to obtain a target individual physical model and a target aggregate cluster model;

[0062] In an embodiment of the present invention, the K-means clustering algorithm is used in combination with the training set data to train the individual physical model and the aggregate cluster model to obtain the trained aggregate cluster model and the trained aggregate cluster model. Based on the test set data, the trained aggregate cluster model and the trained aggregate cluster model are verified to obtain the target individual physical model and the target aggregate cluster model.

[0063] Step S105 , inputting the energy data to be measured into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be measured.

[0064] In an energy data aggregation method provided in an embodiment of the present invention, by acquiring renewable energy data and energy data to be tested, an individual physical model and an aggregation cluster model of renewable energy are established based on the renewable energy data, the renewable energy data are divided into training set data and test set data, and the individual physical model and the aggregation cluster model are trained and verified based on a K-means clustering algorithm, the training set data and the test set data to obtain a target individual physical model and a target aggregation cluster model, and the energy data to be tested are input into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be tested. Through an energy data aggregation method, the technical problem that it is difficult to aggregate the currently existing idle distributed renewable resources in society is solved, and the access standards for renewable resources are unified.

[0065] For example 2, please refer to Figure 2 , Figure 2 A flow chart of a method for aggregating energy data of the present invention includes:

[0066] Step S201, obtaining initial data of renewable energy;

[0067] In an embodiment of the present invention, initial data of renewable energy is obtained, including individual photovoltaic array data, wind turbine data, distributed energy storage data, electric vehicle data, flexibility load data, and intra-cluster characteristic characterization data and extra-cluster characteristic characterization data.

[0068] The internal characteristic characterization data of the cluster include capacity support capability (including baseline value and upper and lower boundaries), power support capability (including baseline value and upper and lower boundaries), dynamic response capability and effective convergence time ratio.

[0069] The external cluster characterization data include stability factor, reliability factor, network loss factor, unit regulation power and dynamic sensitivity factor.

[0070] Step S202, cleaning and repairing the renewable energy initial data to obtain renewable energy data and energy data to be tested;

[0071] Step S203, establishing an individual physical model and an aggregate cluster model of renewable energy based on the renewable energy data;

[0072] In an embodiment of the present invention, based on the renewable energy data, an individual physical model and an aggregated cluster model of renewable energy are established. The individual physical model of renewable energy includes an individual photovoltaic array model, a wind turbine model, a distributed energy storage model, an electric vehicle model and a flexibility load model.

[0073] In the specific implementation, see Figure 3 , Figure 3 It is a standardized model update flow chart of renewable energy of the present invention, wherein t is the time, p is the power generation, e is the accumulated power generation, r is the ramp speed, and D is the external characteristic factor;

[0074] Distributed renewable resources consider individual photovoltaic arrays, wind turbines, distributed energy storage, electric vehicles, and flexible loads. The physical modeling of these resources is mainly about their electrical quantities and constraints, as follows:

[0075] The generation characteristics of a general flexible resource in time period t can be expressed as a six-tuple describe. They are respectively the upper and lower bounds of the power generation, cumulative power generation and ramp rate of the j-th type and k-numbered flexible resource under aggregator node i in the t-th time period.

[0076] For photovoltaic units, the maximum power point tracking (MPPT) technology can ensure that the parameters of photovoltaic cells and loads are optimally matched, and the output power is always maintained at the maximum power point P. m Based on this, the P m The predicted value is regarded as the maximum output prediction Through power electronic devices, it is possible to Continuously adjust the unit output, the upper and lower limits of PV power and cumulative power generation are:

[0077]

[0078]

[0079]

[0080] The control characteristics of wind turbines are similar to those of photovoltaic units. Continuously adjust the unit output, the upper and lower limits of wind turbine unit power and cumulative power generation are:

[0081]

[0082]

[0083]

[0084] Compared with photovoltaic power generation, the output fluctuation of generator sets is larger and irregular, making it more difficult to To make a prediction, the hourly average wind power output change rate is used as the measure:

[0085]

[0086] in, is the hourly output change rate of wind power, is the hourly output fluctuation of wind power, is the rated installed capacity of wind power.

[0087] The energy storage operation characteristics can be described based on the charge (State of Charge, SOC) of the energy storage device at time t:

[0088]

[0089] Where SOC(t) is the state of charge of the energy storage at the end of the tth period; η c , η d are the charging and discharging efficiency of energy storage respectively; E r is the rated capacity of energy storage; σ is the self-discharge rate of energy storage; P c , P d are the charging and discharging powers of the current period respectively.

[0090] Energy storage device operation constraints:

[0091] The first is the charge and discharge power limit of the energy storage device, which is that the energy storage charge and discharge per unit time shall not exceed 20% of its rated capacity, that is:

[0092]

[0093] In order to prevent over-charging and discharging of the energy storage device and to have sufficient margin for emergency dispatch, the minimum charge of the energy storage is set to SOC min , the maximum charge is SOC max ,Right now:

[0094] SOC min ≤SOC ess (t)≤SOC max ;

[0095] Usually take SOC min is 0.2, SOC max is 0.9.

[0096] The charge and discharge power allowed per unit time of the energy storage device changes with SOC, that is:

[0097]

[0098] The essence of electric vehicles is still energy storage, but as a mobile energy storage with stronger user attributes, it brings greater uncertainty. Its user characteristics reflect its social attributes. Electric vehicles are described from the perspective of charging piles, and their social attributes are portrayed by the travel patterns of electric vehicles.

[0099] The probability distribution of the departure time and arrival time of electric vehicles obeys the generalized extreme value distribution, and the probability distribution of the travel distance obeys the Weibull distribution. The Monte Carlo random simulation method can be used to perform random simulation of electric vehicle travel. The inverse function of the departure time, arrival time, and travel distance distribution function of a single electric vehicle is:

[0100]

[0101]

[0102]

[0103] Among them, x out For electric car travel time, f PEV,out is the inverse probability density function of electric vehicle travel time, y out is the probability of electric vehicle travel time, μ PEV,out , σ PEV,out ,ξ PEV,out are the location parameter, scale parameter and shape parameter of the probability distribution of electric vehicle travel time; x in , f PEV,in ,y in , μ PEV,in , σ PEV,in ,ξ PEV,in are the corresponding electric vehicle arrival time, the inverse probability density function of the arrival time, the probability of the arrival time, and the location parameter, scale parameter and shape parameter of the probability distribution of the electric vehicle arrival time; x dis , f dis ,y dis , k dis ,λ dis are the travel distance of electric vehicles, the inverse probability density function of travel distance, the probability of travel distance, and the shape and scale parameters of the travel distance distribution function.

[0104] The flexibility load is characterized from the perspective of interruptibility and translation:

[0105] Mathematical model of interruptible load resources:

[0106] Constraints on the maximum number of resource calls:

[0107]

[0108]

[0109] Resource availability period constraints:

[0110]

[0111] Minimum and maximum resource usage time constraints:

[0112]

[0113] The relationship between the actual operating power of a resource and whether a response is initiated:

[0114]

[0115] Response cost of interruptible load resources:

[0116]

[0117] Among them, S DR is the set of interruptible load resources; i,t is the usage status of interruptible load resources; i,t is the startup variable of the resource, and the above two groups of variables are 0-1 variables; N is the maximum response number of the resource; Ψ use is the set of available moments; T min and T max are the minimum and maximum response time of the resource respectively; p DR,i,t is the actual power of resource i; p i,t,0 and P DR,i,t are the initial power and response power of the resource respectively; c i is the unit power response cost of the interruptible load resource.

[0118] Mathematical model of shiftable load resources:

[0119] Constraints on the maximum number of resource calls:

[0120]

[0121]

[0122] Resource availability period constraints:

[0123]

[0124]

[0125] The relationship between the outflow and inflow of transferable load resources:

[0126]

[0127] Resource move-in time constraints:

[0128]

[0129] Actual operating power of resources:

[0130]

[0131] Relationship between input variables and response variables:

[0132]

[0133] Response cost of load-shiftable resources:

[0134]

[0135] Among them, S TR is a set of load resources that can be translated; i,OUT,t is the usage status of the transferable load resource i; z i,OUT,t is the startup variable of the resource, y i,IN,t Indicates whether the load is moved in during this period. The above three groups of variables are all 0-1 variables; Ψ in is the set of moments that can be moved into; p TR,i,t is the actual power of resource i; p i,t,0 and P TR,i are the initial power and response power of the resource respectively.

[0136] The cluster model has two-layer attributes. Its attributes facing individual aggregation downward are similar to those of the individual model. However, the cluster layer no longer distinguishes between source, storage and load, so the lower model is a unification of the three types of source, storage and load models. The cluster attributes facing the dispatch or market at the upper layer, the physical part, should be characterized by its internal characteristics, that is, the ability to adjust the power up and down and the power baseline of the aggregated virtual.

[0137] use Indicates the power generated by the flexible resource of type j and number k in the access cluster node i in the tth time period. If the power value is positive, it means that the resource sends active power to the grid, and if it is negative, it means that the resource absorbs active power from the grid. It represents the cumulative power generation of the flexible resource of type j and number k in the access cluster node i from the 1st to the tth time period. The specific performance is shown in the following table:

[0138]

[0139] The parameters of this six-tuple represent the active power boundary, cumulative power generation boundary and ramp rate boundary of the flexible resource in period t and from period t to period t+1, namely:

[0140]

[0141]

[0142]

[0143] in, They are respectively the upper and lower bounds of the power generation, cumulative power generation and ramp rate of the j-th type and k-numbered flexible resource under cluster node i in the t-th time period.

[0144] The baseline power is the aggregate power when the cluster does not take any regulatory measures. It is used to confirm the regulatory effect of the cluster and needs to be calculated in a rolling cycle of 5 to 30 minutes. The upward and downward adjustment boundaries are calculated using the individual model boundary calculation method. At the same time, the cluster aggregation evaluation and the social attributes of the cluster are considered to form external characteristics, and finally unified as:

[0145] A={C,P,D,T,Sta,Se,L,K,Sen,…}

[0146] Sta=f 1 (C,D,P,T)

[0147] Se=f 2 (C,D,P,T)

[0148] L = f 3 (C,D,P,T,Site(Z))

[0149] K=f 4 (C,D,P,T,Site(Z));

[0150] Among them, C, P, D, and T are the internal characteristics of the cluster, which are capacity support capability (including reference value and upper and lower boundaries), power support capability (including reference value and upper and lower boundaries), dynamic response capability, and effective convergence time ratio. St, Se, L, K, and Sen are the external characteristics of the cluster, which are stability factor, reliability factor, network loss factor, unit regulation power, and dynamic sensitivity factor.

[0151] Step S204, dividing the renewable energy data into training set data and test set data;

[0152] Step S205, using the K-means clustering algorithm, combined with the training set data, to train the individual physical model and the aggregate cluster model to obtain a trained aggregate cluster model and a trained aggregate cluster model;

[0153] In an optional embodiment, the K-means clustering algorithm is used in combination with the training set data to train the individual physical model and the aggregate cluster model to obtain a trained aggregate cluster model and a trained aggregate cluster model, including:

[0154] Initializing parameters of the individual physical model and parameters of the aggregate cluster model;

[0155] Inputting the training set data into the individual physical model and the aggregate cluster model to obtain the predicted aggregate value of the corresponding renewable energy data;

[0156] Determining a training error according to the data labels corresponding to the training set data and the predicted aggregate value;

[0157] Based on the training error, the individual physical model and the aggregate cluster model are adjusted through the K-means clustering algorithm to obtain optimal parameters, and the individual physical model and the aggregate cluster model are optimized using the optimal parameters to obtain the trained aggregate cluster model and the trained aggregate cluster model.

[0158] In an embodiment of the present invention, the parameters of the individual physical model and the parameters of the aggregate cluster model are initialized, the training set data is input into the individual physical model and the aggregate cluster model to obtain the corresponding predicted aggregate value of the renewable energy data, the training error is determined according to the data label corresponding to the training set data and the predicted aggregate value, based on the training error, the individual physical model and the aggregate cluster model are adjusted by the K-means clustering algorithm to obtain the optimal parameters, and the optimal parameters are used to optimize the individual physical model and the aggregate cluster model to obtain the trained aggregate cluster model and the trained aggregate cluster model.

[0159] Step S206, based on the test set data, verify the trained aggregation cluster model and the trained aggregation cluster model to obtain the target individual physical model and the target aggregation cluster model;

[0160] Step S207, inputting the energy data to be measured into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be measured;

[0161] In an embodiment of the present invention, the energy data to be tested is input into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be tested, and determine the aggregation degree of the energy corresponding to the energy data to be tested.

[0162] In the specific implementation, the aggregation characteristics of the cluster depend on the characteristics of the various distributed resources involved in the combination, as well as its internal resource control strategy. The cluster server needs to adjust the output or working status of the interactive resources within the allowable range to make the overall power of the system close to the target curve. Since the interactive resources respond to the control instructions of the cluster and have certain power adjustment characteristics, after aggregation, the cluster power adjustment capability will produce a scale effect, and the power curve will be smoother.

[0163] The spatial area to which users belong is divided based on the grid connection point to obtain the spatial aggregation characteristics of the cluster:

[0164]

[0165] After aggregating the response characteristics distributed in different geographical locations, we can obtain the temporal aggregation characteristics of the cluster:

[0166]

[0167] Where V i,j,k is a 0-1 variable. When V i,j,k When it is 1, it means participating in the response, ΔP Group,t represents the response value provided by the cluster in period t, P Group,t Indicates the actual aggregate load value of the cluster in period t after the response.

[0168] In an energy data aggregation method provided in an embodiment of the present invention, by acquiring renewable energy data and energy data to be tested, an individual physical model and an aggregation cluster model of renewable energy are established based on the renewable energy data, the renewable energy data are divided into training set data and test set data, and the individual physical model and the aggregation cluster model are trained and verified based on a K-means clustering algorithm, the training set data and the test set data to obtain a target individual physical model and a target aggregation cluster model, and the energy data to be tested are input into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be tested. Through an energy data aggregation method, the technical problem that it is difficult to aggregate the currently existing idle distributed renewable resources in society is solved, and the access standards for renewable resources are unified.

[0169] See also Figure 4 , Figure 4 The structure block diagram of an embodiment of an energy data aggregation device of the present invention includes:

[0170] An acquisition module 401 is used to acquire renewable energy data and energy data to be tested;

[0171] Establishing module 402, for establishing individual physical models and aggregate cluster models of renewable energy based on the renewable energy data;

[0172] A division module 403, used to divide the renewable energy data into training set data and test set data;

[0173] A training module 404 is used to train and verify the individual physical model and the aggregate cluster model based on the K-means clustering algorithm, the training set data and the test set data to obtain a target individual physical model and a target aggregate cluster model;

[0174] The indicator module 405 is used to input the energy data to be measured into the target individual physical model and the target aggregate cluster model to obtain predicted aggregate indicator data of the energy data to be measured.

[0175] In an optional embodiment, the acquisition module 401 includes:

[0176] The acquisition submodule is used to obtain the initial data of renewable energy;

[0177] The repair submodule is used to clean and repair the renewable energy initial data to obtain the renewable energy data and the energy data to be tested.

[0178] In an optional embodiment, the training module 404 includes:

[0179] A training submodule, for training the individual physical model and the aggregate cluster model using the K-means clustering algorithm in combination with the training set data, to obtain a trained aggregate cluster model and a trained aggregate cluster model;

[0180] The verification submodule is used to verify the trained aggregation cluster model and the trained aggregation cluster model based on the test set data to obtain the target individual physical model and the target aggregation cluster model.

[0181] In an optional embodiment, the training submodule includes:

[0182] A prediction unit, used to input the training set data into the individual physical model and the aggregate cluster model to obtain a predicted aggregate value of the corresponding renewable energy data;

[0183] An error unit, used to determine a training error according to a data label corresponding to the training set data and the predicted aggregate value;

[0184] An optimization unit is used to adjust the individual physical model and the aggregate cluster model based on the training error through the K-means clustering algorithm to obtain optimal parameters, and use the optimal parameters to optimize the individual physical model and the aggregate cluster model to obtain the trained aggregate cluster model and the trained aggregate cluster model.

[0185] In an optional embodiment, the training submodule further includes:

[0186] A parameter unit is used to initialize the parameters of the individual physical model and the parameters of the aggregate cluster model.

[0187] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0188] In the several embodiments provided in the present application, it should be understood that the methods and devices disclosed in the present invention can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0189] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0190] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0191] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a readable storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0192] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for aggregating energy data, It is characterized in that include: Obtain renewable energy data and energy data to be measured; Based on the renewable energy data, an individual physical model and an aggregated cluster model of renewable energy are established; the individual physical model includes an individual photovoltaic array model, a wind turbine model, a distributed energy storage model, an electric vehicle model and a flexible load model; the aggregated cluster model has a double-layer attribute, the lower layer of the aggregated cluster model faces individual aggregate attributes similar to the individual physical model, and the upper layer faces cluster attributes of scheduling or the market; Dividing the renewable energy data into training set data and test set data; Based on the K-means clustering algorithm, the training set data and the test set data, the individual physical model and the aggregate cluster model are trained and verified to obtain a target individual physical model and a target aggregate cluster model; The energy data to be measured is input into the target individual physical model and the target aggregation cluster model to obtain predicted aggregation index data of the energy data to be measured.

2. The method for aggregating energy data according to claim 1, It is characterized in that Obtain renewable energy data and energy data to be measured, including: Obtaining initial data on renewable energy; The renewable energy initial data is cleaned and repaired to obtain the renewable energy data and the energy data to be tested.

3. The method for aggregating energy data according to claim 1, It is characterized in that Based on the K-means clustering algorithm, the training set data and the test set data, the individual physical model and the aggregate cluster model are trained and verified to obtain a target individual physical model and a target aggregate cluster model, including: Using the K-means clustering algorithm, combined with the training set data, to train the individual physical model and the aggregate cluster model, to obtain a trained individual physical model and a trained aggregate cluster model; Based on the test set data, the trained individual physical model and the trained aggregate cluster model are verified to obtain the target individual physical model and the target aggregate cluster model.

4. The method for aggregating energy data according to claim 3, It is characterized in that Using the K-means clustering algorithm, combined with the training set data, the individual physical model and the aggregate cluster model are trained to obtain the trained individual physical model and the trained aggregate cluster model, including: Inputting the training set data into the individual physical model and the aggregate cluster model to obtain the predicted aggregate value of the corresponding renewable energy data; Determining a training error according to the data labels corresponding to the training set data and the predicted aggregate value; Based on the training error, the individual physical model and the aggregate cluster model are adjusted through the K-means clustering algorithm to obtain optimal parameters, and the individual physical model and the aggregate cluster model are optimized using the optimal parameters to obtain the trained individual physical model and the trained aggregate cluster model.

5. The method for aggregating energy data according to claim 4, It is characterized in that Before inputting the training set data into the individual physical model and the aggregate cluster model to obtain the corresponding predicted aggregate value of renewable energy data, the method further includes: Initialize the parameters of the individual physical model and the parameters of the aggregate cluster model.

6. An energy data aggregation device, It is characterized in that include: An acquisition module, used for acquiring renewable energy data and energy data to be tested; Establishing a module for establishing an individual physical model and an aggregated cluster model of renewable energy based on the renewable energy data; the individual physical model includes an individual photovoltaic array model, a wind turbine model, a distributed energy storage model, an electric vehicle model, and a flexibility load model; the aggregated cluster model has a double-layer attribute, the lower layer of the aggregated cluster model faces individual aggregate attributes similar to the individual physical model, and the upper layer faces cluster attributes of scheduling or the market; A partitioning module, used for partitioning the renewable energy data into training set data and test set data; A training module, used for training and verifying the individual physical model and the aggregate cluster model based on a K-means clustering algorithm, the training set data and the test set data, to obtain a target individual physical model and a target aggregate cluster model; The indicator module is used to input the energy data to be measured into the target individual physical model and the target aggregate cluster model to obtain predicted aggregate indicator data of the energy data to be measured.

7. The energy data aggregation device according to claim 6, It is characterized in that The acquisition module comprises: The acquisition submodule is used to obtain the initial data of renewable energy; The repair submodule is used to clean and repair the renewable energy initial data to obtain the renewable energy data and the energy data to be tested.

8. The energy data aggregation device according to claim 6, It is characterized in that The training module includes: A training submodule, for training the individual physical model and the aggregate cluster model by using the K-means clustering algorithm in combination with the training set data, to obtain a trained individual physical model and a trained aggregate cluster model; The verification submodule is used to verify the trained individual physical model and the trained aggregate cluster model based on the test set data to obtain the target individual physical model and the target aggregate cluster model.

9. The energy data aggregation device according to claim 8, It is characterized in that The training submodule includes: A prediction unit, used to input the training set data into the individual physical model and the aggregate cluster model to obtain a predicted aggregate value of the corresponding renewable energy data; An error unit, used to determine a training error according to a data label corresponding to the training set data and the predicted aggregate value; An optimization unit is used to adjust the individual physical model and the aggregate cluster model based on the training error through the K-means clustering algorithm to obtain optimal parameters, and use the optimal parameters to optimize the individual physical model and the aggregate cluster model to obtain the trained individual physical model and the trained aggregate cluster model.

10. The energy data aggregation device according to claim 9, It is characterized in that The training submodule also includes: A parameter unit is used to initialize the parameters of the individual physical model and the parameters of the aggregate cluster model.

Images