Load resource dynamic aggregation and optimal scheduling method and system and storage medium

Abstract

The invention discloses a load resource dynamic aggregation and optimization scheduling method and system and a storage medium, and belongs to the field of energy intelligent management, and the method comprises the following steps: obtaining historical data of a photovoltaic power generation panel at a source end, building a neural network model based on the historical data, and predicting the day-ahead and day-in output power of the source end; the method comprises the following steps: acquiring data of load side resources on a time scale, constructing a demand power model of a plurality of load side resources, and acquiring predicted demand power of a plurality of loads before and in the day according to the demand power model; and finally, according to the supply and demand balance equation and the load side resource optimization dynamic aggregation equation, carrying out multi-target optimization by taking the load side demand as a main part, regulating and controlling the lithium battery and photovoltaic power generation power of the source side, and achieving supply and demand balance of the source side and the load side by taking economic optimization as a target. The demand side interactive response capability is exerted, the power supply and demand balance is promoted, and the energy utilization rate is improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent energy management, and specifically relates to a method, system and storage medium for dynamic aggregation and optimal scheduling of load resources. Background Art

[0002] With the construction of new power systems, a large number of new power load components have emerged, and the importance of the level of interaction between power supply and demand and the load regulation capability on the demand side has become increasingly prominent. Compared with traditional power loads, new power loads use power electronics as grid-connected interface devices, which makes modeling more difficult. The current modeling mainly focuses on re-evaluating existing load models, which can be divided into statistical synthesis methods, overall measurement and identification methods, etc.

[0003] Among the existing load modeling methods, the statistical synthesis method adopts a bottom-up approach, which can clearly obtain the load components and their characteristics, load model categories, etc., but does not consider the detailed situation of the distribution network. The overall identification method takes the load as a whole and has a good simulation effect on the external characteristics of the load, but it relies too much on the richness of data samples and must be combined with the actual power grid architecture and the physical models of each component.

[0004] In the existing research on dynamic aggregation and regulation of load-side resources, the changes in the mathematical nature of the multi-energy flow equation itself and the corresponding optimization model as the time scale of load-side resources changes are not taken into consideration, resulting in insufficient demand-side load regulation capabilities and reduced economic efficiency. Summary of the invention

[0005] In response to one or more of the above defects or improvement needs of the prior art, the present invention provides a method, system and storage medium for dynamic aggregation and optimal scheduling of load resources. The power consumption characteristics of the load side are obtained by integrating the statistical synthesis method with the overall identification method. On this basis, a unified load modeling method is adopted to build models of distributed photovoltaic, air-conditioning load, energy storage, electric vehicles, 5G base stations, etc., and then a three-stage internal optimization model of load-side resources is constructed, including day-ahead resource capacity prediction-intraday rolling aggregation optimization-real-time tracking and correction. A digital multi-load resource adjustment potential and dynamic aggregation simulation platform is constructed to achieve flexible regulation of load-side resources.

[0006] To achieve the above object, the present invention provides a method for dynamic aggregation and optimal scheduling of load resources, which comprises the following steps:

[0007] S100: Acquire historical data of photovoltaic power generation panels at the source end, and build a neural network model based on the historical data to predict the output power of the source end before and during the day;

[0008] S200: Obtain data on load-side resources on a time scale, construct demand power models for multiple load-side resources, and obtain predicted demand power for multiple loads before and during the day based on the demand power models; meanwhile, collect actual load demand power in real time;

[0009] S300: Finally, based on the supply and demand balance equation and the dynamic aggregation equation for load-side resource optimization, multi-objective optimization is performed with the load-side demand as the main focus, and the lithium battery and photovoltaic power generation power at the source end are regulated to achieve supply and demand balance between the source and load ends with economic optimization as the goal.

[0010] As a further improvement of the present invention, in step S200, data of load-side resources on a time scale is obtained, and the data includes operating characteristics of load-side resources, a power grid dispatch instruction cycle, and a prediction error.

[0011] As a further improvement of the present invention, the multiple load-side resources include electric vehicles, 5G base stations and air conditioners.

[0012] As a further improvement of the present invention, the construction of the air conditioning part in the multiple load-side resources takes the outdoor stable and set temperatures as the model output, and the output is the power required for the air conditioning operation;

[0013] The model of the 5G base station part in the multiple load-side resources is constructed, and the possible number of people and the number of mobile phone activations in the region are estimated by the regional population density and regional area, and the probability distribution of residents using the 5G network in a day is simulated with reference to the actual situation to obtain the required power of the 5G base station part in the load-side resources;

[0014] The model of the electric vehicle part in the multiple load-side resources is constructed, and the simulation is performed by using the Monte Carlo simulation method, and the date type, the starting charging time of the electric vehicle and the daily mileage are randomly selected as the output of the model, and the load state of the electric vehicle is calculated, and finally the charging load states of each electric vehicle are superimposed to obtain the total charging load of the electric vehicle;

[0015] As a further improvement of the present invention, the model construction of the load side resources of the electric vehicle, 5G base station and air conditioner specifically includes the following steps:

[0016] A unified load modeling method is adopted, and load-side resources are divided into time-shiftable loads and power-shiftable loads according to power demand;

[0017] As a further improvement of the present invention, in step S100, the following steps are specifically included:

[0018] Calculate the expected supply power based on direct solar irradiance and outdoor temperature;

[0019] When the bus voltage is constant, the energy is connected to the load end through the micro energy grid;

[0020] The power of photovoltaic panels and energy storage lithium batteries is constrained, and the constraint formula is as follows:

[0021]

[0022] Where: P PV_rated , P B_in and P B_out They are the rated powers of photovoltaic panels, lithium battery charging and lithium battery discharging respectively.

[0023] As a further improvement of the present invention, in step S300, the supply-demand balance equation is as follows:

[0024] P air (t)+P car (t)+P 5G (t) = P PV (t)+P B (t)

[0025] Where P PV , P B , P B , P air , P car and P 5G They are the power supplied by photovoltaic panels and lithium batteries at the source end and the power required by air conditioners, electric vehicles and 5G base stations at the load end.

[0026] As a further improvement of the present invention, in step S300, the load-side resource optimization dynamic aggregation equation is as follows:

[0027]

[0028] In the formula, They are the power of each load at the day before, during the day and in real time;

[0029] C1 is the electricity purchase cost of the main grid the day before;

[0030] C2 is the intraday regulation cost, which includes the additional cost of purchasing electricity from the main grid due to insufficient photovoltaic output (cost of purchasing lithium batteries) and the electricity sales income generated by excessive photovoltaic output;

[0031] C3 is the real-time adjustment fee, which includes the additional cost of purchasing electricity from the main grid due to insufficient photovoltaic power generation, the electricity sales revenue generated by excessive photovoltaic output, and the compensation fee for demand response;

[0032] μ1, μ2, μ3, and μ4 are the proportional coefficients of intraday and real-time power regulation, respectively. When there is surplus electricity to be sold, 0<μ4<μ2<1, and when additional electricity needs to be purchased, μ3>μ1>1;

[0033] c(t) is the day-ahead electricity price; g(s) is the probability of each scenario occurring; b is the demand response compensation unit price. The intraday and real-time electricity purchase and sales prices are the day-ahead electricity prices multiplied by the corresponding proportional coefficients.

[0034] On this basis, the present invention also provides a load resource dynamic aggregation and optimal scheduling system, which is used to implement the above-mentioned load resource dynamic aggregation and optimal scheduling method.

[0035] On this basis, the present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned load resource dynamic aggregation and optimization scheduling method is executed.

[0036] The above-mentioned improved technical features can be combined with each other as long as they do not conflict with each other.

[0037] In general, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0038] (1) The method, system and storage medium for dynamic aggregation and optimal scheduling of load resources of the present invention obtain the load-side power consumption characteristics by integrating the statistical synthesis method with the overall identification method, and on this basis, a unified load modeling method is used to build models of distributed photovoltaic, air-conditioning load, energy storage, electric vehicles, 5G base stations, etc., and then a three-stage internal optimization model of load-side resources is constructed, namely, day-ahead resource capacity prediction - intraday rolling aggregation optimization - real-time tracking and correction, to build a digital multi-load resource regulation potential and dynamic aggregation simulation platform to achieve flexible regulation of load-side resources.

[0039] (2) The load resource dynamic aggregation and optimization scheduling method, system and storage medium of the present invention incorporate new load elements into the micro-energy network scheduling system, optimize the multi-element load resource model including the new load elements, make it more consistent with the actual source and load conditions, and innovatively propose to study the station-level aggregation strategy adapted to different application scenarios from the day-ahead, intra-day and real-time dimensions, give full play to the interactive response capability of the demand side, promote the balance of power supply and demand, and improve energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the flow chart structure of the method for dynamic aggregation and optimal scheduling of load resources in an embodiment of the present invention;

[0041] Figure 2It is a curve diagram of 5G base station usage probability-time in the load resource dynamic aggregation and optimization scheduling method in an embodiment of the present invention;

[0042] Figure 3 is a curve diagram of electric vehicle power usage-time in the load resource dynamic aggregation and optimal scheduling method in an embodiment of the present invention;

[0043] Figure 4 It is a source-end photovoltaic power curve diagram in the load resource dynamic aggregation and optimal scheduling method in an embodiment of the present invention.

[0044] Figure 5 It is an output and input composition diagram of the load side resource model constructed in the dynamic aggregation and optimal scheduling of load resources in the embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0046] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0047] See also Figures 1 to 5 The load resource dynamic aggregation and optimization scheduling method in the preferred embodiment of the present invention obtains the load side power consumption characteristics by integrating the statistical synthesis method and the overall identification method. On this basis, a unified load modeling method is used to build models of distributed photovoltaic, air-conditioning load, energy storage, electric vehicles, 5G base stations, etc., and then a three-stage internal optimization model of load side resource day-ahead resource capacity prediction-intraday rolling aggregation optimization-real-time tracking and correction is constructed, and a digital multi-load resource regulation potential and dynamic aggregation simulation platform is constructed to realize flexible regulation of load side resources.

[0048] Specifically, the scheduling method in the preferred embodiment of the present invention comprises the following steps:

[0049] S100: Acquire historical data of photovoltaic power generation panels at the source end, and build a neural network model based on the historical data to predict the output power of the source end before and during the day;

[0050] Further, in step S100, the following steps are specifically included: according to the direct solar irradiance and the outdoor temperature, the expected supply power is calculated; when the bus voltage is constant, the energy is connected to the load end through the micro energy grid; and the power of the photovoltaic panel and the energy storage lithium battery is constrained, and the constraint formula is as follows:

[0051]

[0052] Where: P PV_rated , P B_in and P B_out They are the rated powers of photovoltaic panels, lithium battery charging and lithium battery discharging respectively.

[0053] S200: Obtain data on load-side resources on a time scale, construct demand power models for multiple load-side resources, and obtain predicted demand power for multiple loads before and during the day based on the demand power models; at the same time, collect actual load demand power in real time.

[0054] In more detail, in step S200, data of load-side resources on a time scale is obtained, the data including operating characteristics of the load-side resources, a grid dispatch instruction cycle, and a prediction error. The multiple load-side resources include electric vehicles, 5G base stations, and air conditioners.

[0055] Furthermore, the construction of the air-conditioning part among multiple load-side resources takes the outdoor stable and set temperature as the model output, and the output is the power required for the air-conditioning operation; the construction of the model of the 5G base station part among multiple load-side resources estimates the possible number of people and the number of mobile phone activations in the region through the regional population density and regional area, and simulates the probability distribution of residents using the 5G network in a day with reference to the actual situation to obtain the required power of the 5G base station part in the load-side resources; the construction of the model of the electric vehicle part among multiple load-side resources is simulated through the Monte Carlo simulation method, and the date type, electric vehicle starting charging time and daily mileage are randomly selected as the output of the model to calculate the load state of the electric vehicle, and finally the charging load states of each electric vehicle are superimposed to obtain the total electric vehicle charging load.

[0056] Furthermore, the model construction of the load-side resources of electric vehicles, 5G base stations, and air conditioners specifically includes the following steps: using a unified load modeling method, and dividing the load-side resources into time-shiftable loads and power-shiftable loads according to electricity demand;

[0057] S300: Finally, based on the supply and demand balance equation and the dynamic aggregation equation for load-side resource optimization, multi-objective optimization is performed with the load-side demand as the main focus, and the lithium battery and photovoltaic power generation power at the source end are regulated to achieve supply and demand balance between the source and load ends with economic optimization as the goal.

[0058] Furthermore, in step S300, the supply-demand balance equation is as follows:

[0059] P air (t)+P car (t)+P 5G (t) = P PV (t)+P B (t)

[0060] Where P PV , P B , P B , P air , P car and P 5G They are the power supplied by photovoltaic panels and lithium batteries at the source end and the power required by air conditioners, electric vehicles and 5G base stations at the load end.

[0061] Furthermore, in step S300, the load-side resource optimization dynamic aggregation equation is as follows:

[0062]

[0063] In the formula, They are the power of each load at the day before, during the day and in real time;

[0064] C1 is the day-ahead main grid electricity purchase fee; C2 is the intraday regulation fee, which includes the additional main grid electricity purchase fee (lithium battery purchase fee) caused by insufficient photovoltaic output and the electricity sales revenue generated by excessive photovoltaic output; C3 is the real-time regulation fee, which includes the additional main grid electricity purchase fee caused by insufficient photovoltaic power generation, the electricity sales revenue generated by excessive photovoltaic output and the compensation fee for demand response;

[0065] μ1, μ2, μ3, and μ4 are the proportional coefficients of intraday and real-time power regulation, respectively. When there is surplus electricity to be sold, 0<μ4<μ2<1, and when additional electricity needs to be purchased, μ3>μ1>1;

[0066] c(t) is the day-ahead electricity price; g(s) is the probability of each scenario occurring; b is the demand response compensation unit price. The intraday and real-time electricity purchase and sales prices are the day-ahead electricity prices multiplied by the corresponding proportional coefficients.

[0067] On this basis, the present invention also provides a load resource dynamic aggregation and optimal scheduling system, which is used to implement the above-mentioned load resource dynamic aggregation and optimal scheduling method.

[0068] On this basis, the present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned load resource dynamic aggregation and optimization scheduling method is executed.

[0069] (2) The method, system and storage medium for dynamic aggregation and optimal scheduling of load resources of the present invention obtain the load-side power consumption characteristics by integrating the statistical synthesis method with the overall identification method, and on this basis, a unified load modeling method is used to build models of distributed photovoltaic, air-conditioning load, energy storage, electric vehicles, 5G base stations, etc., and then a three-stage internal optimization model of load-side resources is constructed, namely, day-ahead resource capacity prediction - intraday rolling aggregation optimization - real-time tracking and correction, to build a digital multi-load resource regulation potential and dynamic aggregation simulation platform to achieve flexible regulation of load-side resources.

[0070] (2) The load resource dynamic aggregation and optimization scheduling method, system and storage medium of the present invention incorporate new load elements into the micro-energy network scheduling system, optimize the multi-element load resource model including the new load elements, make it more consistent with the actual source and load conditions, and innovatively propose to study the station-level aggregation strategy adapted to different application scenarios from the day-ahead, intra-day and real-time dimensions, give full play to the interactive response capability of the demand side, promote the balance of power supply and demand, and improve energy utilization.

[0071] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for dynamic aggregation and optimal scheduling of load resources, characterized in that: The method comprises the following steps: S100: Acquire historical data of photovoltaic power generation panels at the source end, and build a neural network model based on the historical data to predict the output power of the source end before and during the day; S200: Obtain data on load-side resources on a time scale, construct demand power models for multiple load-side resources, and obtain predicted demand power for multiple loads before and during the day based on the demand power models; meanwhile, collect actual load demand power in real time; S300: Finally, based on the supply and demand balance equation and the dynamic aggregation equation for load-side resource optimization, multi-objective optimization is performed with the load-side demand as the main focus, and the lithium battery and photovoltaic power generation power at the source end are regulated to achieve supply and demand balance between the source and load ends with economic optimization as the goal.

2. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: In step S200, data of load-side resources on a time scale are obtained, and the data include operating characteristics of load-side resources, a power grid dispatch instruction cycle, and a prediction error.

3. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: The multiple load-side resources include electric vehicles, 5G base stations and air conditioners.

4. The method for dynamic aggregation and optimal scheduling of load resources according to claim 3 is characterized in that: The construction of the air conditioning part of the multiple load-side resources takes the outdoor stable and set temperature as the model output, and the output is the power required for the air conditioning operation; The model of the 5G base station part in the multiple load-side resources is constructed, and the possible number of people and the number of mobile phone activations in the region are estimated by the regional population density and regional area, and the probability distribution of residents using the 5G network in a day is simulated with reference to the actual situation to obtain the required power of the 5G base station part in the load-side resources; The model of the electric vehicle part in the multiple load-side resources is constructed and simulated by the Monte Carlo simulation method. The date type, the starting charging time of the electric vehicle and the daily mileage are randomly selected as the output of the model, and the load state of the electric vehicle is calculated. Finally, the charging load states of each electric vehicle are superimposed to obtain the total charging load of the electric vehicle.

5. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: The model construction of the load side resources of the electric vehicle, 5G base station and air conditioner specifically includes the following steps: A unified load modeling method is adopted, and the load-side resources are divided into time-shiftable loads and power-shiftable loads according to electricity demand.

6. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: In step S100, the following steps are specifically included: Calculate the expected supply power based on direct solar irradiance and outdoor temperature; When the bus voltage is constant, the energy is connected to the load end through the micro energy grid; The power of photovoltaic panels and energy storage lithium batteries is constrained, and the constraint formula is as follows: Where: P PV_rated , P B_in and P B_out They are the rated powers of photovoltaic panels, lithium battery charging and lithium battery discharging respectively.

7. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: In step S300, the supply-demand balance equation is as follows: P air (t)+P car (t)+P 5G (t)=P PV (t)+P B (t) Where P PV , P B , P B , P air , P car and P 5G They are the power supplied by photovoltaic panels and lithium batteries at the source end and the power required by air conditioners, electric vehicles and 5G base stations at the load end.

8. The method for dynamic aggregation and optimal scheduling of load resources according to claim 1, characterized in that: In step S300, the load-side resource optimization dynamic aggregation equation is as follows: In the formula, They are the power of each load at the day before, during the day and in real time; C1 is the electricity purchase cost of the main grid the day before; C2 is the intraday regulation cost, which includes the additional cost of purchasing electricity from the main grid due to insufficient photovoltaic output (cost of purchasing lithium batteries) and the electricity sales income generated by excessive photovoltaic output; C3 is the real-time adjustment fee, which includes the additional cost of purchasing electricity from the main grid due to insufficient photovoltaic power generation, the electricity sales revenue generated by excessive photovoltaic output, and the compensation fee for demand response; μ1, μ2, μ3, and μ4 are the proportional coefficients of intraday and real-time power regulation, respectively. When there is surplus electricity to be sold, 0<μ4<μ2<1, and when additional electricity needs to be purchased, μ3>μ1>1; c(t) is the day-ahead electricity price; g(s) is the probability of each scenario occurring; b is the demand response compensation unit price; the intraday and real-time electricity purchase and sales unit prices are the day-ahead electricity prices multiplied by the corresponding proportional coefficients.

9. A load resource dynamic aggregation and optimization scheduling system, characterized in that: Used to implement the load resource dynamic aggregation and optimal scheduling method described in any one of claims 1 to 8.

10. A storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for dynamic aggregation and optimal scheduling of load resources as claimed in any one of claims 1 to 8 is executed.

Images