A source-load cluster based control method and system for a honeycomb power distribution network structure

Abstract

The application discloses a kind of source and load cluster based on the regulation and control method and terminal of honeycomb power distribution network structure, the power optimization deployment of each source and load cluster energy management center in the source and load cluster of belonging to internal, and through the energy interaction base station of honeycomb power distribution network, the surplus / deficiency power of issue is carried out;The power distribution network control center of the power distribution network according to the surplus / deficiency power issued by each energy management center, and the interconnection between each source and load cluster, carries out the secondary flow distribution of power, only when the power between each source and load cluster cannot be complementary, by the power distribution network, further support is carried out to power;In the power distribution network of honeycomb, adjacent source and load cluster is interconnected by the energy interaction base station, and the energy interaction base station monitors the energy and information interaction between each source and load cluster;The present application realizes the maximization of use of distributed power supply on the basis of ensuring the quality of distributed load electric energy.

Description

Technical Field

[0001] This invention relates to the field of power system operation and planning technology, and in particular to a source-load cluster control method and system based on a honeycomb distribution network structure. Background Technology

[0002] In the early stages of distributed resource development, due to the immaturity of distributed resource equipment and technologies, their scale in the distribution network was small, allowing for direct control by the distribution network operator. Their randomness and intermittency had a negligible impact on distribution network operation. At this stage, the integration of distributed sources and loads did not significantly affect the safety and stability of the distribution network. However, with the vigorous development of new energy sources, on the one hand, the scale of large-scale centralized photovoltaic and wind power installations has significantly increased; on the other hand, massive amounts of distributed power sources will be integrated into the distribution network, both contributing to the cleanliness and greening of electricity. Furthermore, with the development of flexible power devices such as power electronics, distributed loads such as electric vehicles and user-side energy storage are also continuously growing. The distribution network must possess a higher level of load-side management, no longer simply acting as a power distributor. Therefore, with the integration of massive distributed power sources and loads, the uncertainties on both the source and load sides of the distribution network are becoming increasingly prominent. Simultaneously, the distribution network is no longer a unidirectional power flow network but a bidirectional power flow system, leading to increasingly prominent power quality issues such as overvoltage, posing challenges to the control and management of the distribution network.

[0003] In traditional power distribution networks, centralized control is used for power dispatch. However, with the integration of large-scale distributed resources, the number of monitoring and control points has surged, making centralized control insufficient to meet the energy and information interaction needs of massive nodes. Therefore, with the explosive growth and widespread integration of distributed power sources and diverse flexible loads, the aggregation of power sources and loads to participate in grid interaction will become an important feature of the development of distributed smart grids. Thus, power source clusters, as a whole, participate in power system control, enabling large-scale, multi-type renewable energy to be consumed locally and used plug-and-play. These clusters possess internal energy management platforms that can operate autonomously to a certain extent, dispatching their internal distributed power sources and loads. This avoids the need for the distribution network to dispatch large-scale distributed resource equipment, making power source clusters a crucial link between distributed resources and the upper-level distribution network.

[0004] When there are many source-load clusters in a distribution network, the existing distribution network structure limits its intelligence and autonomy. However, with the development of power electronics technologies such as flexible soft switching, the proposal of honeycomb-shaped distribution network structure provides a new approach to the interconnection between source-load clusters. How to combine source-load clusters with honeycomb-shaped distribution networks to provide a better control method is a technical problem currently under research. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a control method and system for source-load clusters based on a honeycomb distribution network structure, and to provide a better control scheme by combining source-load clusters and honeycomb distribution networks.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for regulating source-load clusters based on a honeycomb distribution network structure includes the following steps:

[0008] S1. The energy management center of each source-load cluster performs power optimization and allocation within its respective source-load cluster, and publishes surplus / deficient power through the energy interaction base station of the honeycomb distribution network.

[0009] S2. The distribution network control center of the distribution network performs secondary power flow allocation based on the surplus / deficit power released by each of the energy management centers and the interconnection status between each of the source-load clusters. Only when the power between each of the source-load clusters cannot be complementary will the distribution network provide further power support.

[0010] In the honeycomb-shaped power distribution network, adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information interaction between each source-load cluster.

[0011] To solve the above-mentioned technical problems, another technical solution adopted by the present invention is as follows:

[0012] A source-load cluster control system based on a honeycomb distribution network structure includes a source-load cluster, an energy interaction base station, and a distribution network control center.

[0013] The source-load cluster includes distributed loads, distributed power sources, an energy management center, and a distributed source-load aggregation controller.

[0014] The energy management center controls and communicates with multiple source-load aggregation controllers, and each source-load aggregation controller controls and communicates with all distributed loads / distributed power sources belonging to a certain type.

[0015] The source-load clusters form a honeycomb-like power distribution network structure through the energy interaction base station. Adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information exchange between each source-load cluster.

[0016] The energy management center communicates with the power distribution network control center through the energy interaction base station;

[0017] The source-load aggregation controller, the energy management center, and the distribution network control center jointly implement the steps in the above-described method for controlling a source-load cluster based on a honeycomb distribution network structure.

[0018] The beneficial effects of this invention are as follows: The method and system for regulating source-load clusters based on a honeycomb distribution network structure establishes a hierarchical and clustered operation mechanism. The distribution network only needs to directly regulate the source-load clusters, avoiding direct regulation of dispersed distributed source-load devices, thus providing a regulation approach for massive distributed source-load access to the distribution network. Each source-load cluster is connected through a honeycomb distribution network, and energy and information exchange is achieved through the base stations of the honeycomb distribution network, resulting in strong scalability. Energy complementarity exists between different source-load clusters, while the source-load clusters receive support from the upper-level power grid, thereby maximizing the utilization of distributed power sources while ensuring the power quality of distributed loads. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a source-load cluster control method based on a honeycomb distribution network structure according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the hierarchical control architecture of the source-load cluster according to an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram illustrating the distributed source-load aggregation concept of an embodiment of the present invention;

[0022] Figure 4 This is an example diagram of the operating state of a variable frequency air conditioner according to an embodiment of the present invention;

[0023] Figure 5 This is a schematic diagram of a honeycomb power distribution network topology according to an embodiment of the present invention;

[0024] Figure 6 This is a flowchart illustrating the optimization solution process of a source-load cluster control method based on a honeycomb distribution network structure according to an embodiment of the present invention.

[0025] Figure 7 This is a schematic diagram of the power self-scheduling results of a source-load cluster in a high-tech zone of a city, according to an embodiment of the present invention.

[0026] Figure 8 This is a schematic diagram of the power self-dispatch results of a source-load cluster in an industrial area of ​​a city according to an embodiment of the present invention.

[0027] Figure 9 This is a schematic diagram of the power self-dispatch results of a source-load cluster in a residential area of ​​a city according to an embodiment of the present invention.

[0028] Figure 10This is an example diagram illustrating the interaction power between various source-load clusters and between source-load clusters and the distribution network in an embodiment of the present invention. Detailed Implementation

[0029] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0030] Please refer to Figure 1 A source-load cluster control method based on a honeycomb distribution network structure includes the following steps:

[0031] S1. The energy management center of each source-load cluster performs power optimization and allocation within its respective source-load cluster, and publishes surplus / deficient power through the energy interaction base station of the honeycomb distribution network.

[0032] S2. The distribution network control center of the distribution network performs secondary power flow allocation based on the surplus / deficit power released by each of the energy management centers and the interconnection status between each of the source-load clusters. Only when the power between each of the source-load clusters cannot be complementary will the distribution network provide further power support.

[0033] In the honeycomb-shaped power distribution network, adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information interaction between each source-load cluster.

[0034] As can be seen from the above description, the beneficial effects of the present invention are as follows: The source-load cluster control method and system based on a honeycomb distribution network structure of the present invention establishes a hierarchical and clustered operation mechanism. The distribution network only needs to directly control the source-load cluster, avoiding direct control of dispersed distributed source-load devices, and providing a control approach for massive distributed source-load access to the distribution network; the various source-load clusters are connected through the honeycomb distribution network, and energy and information interaction is realized through the base station of the honeycomb distribution network, which has strong scalability; the energy complementarity between different source-load clusters, and the source-load clusters receive support from the upper-level power grid, thereby maximizing the utilization of distributed power sources while ensuring the power quality of distributed loads.

[0035] Furthermore, step S1 includes the following steps:

[0036] S11. Through the distributed source load aggregation controllers of the source load cluster, the distributed source loads of the same type are aggregated to obtain the overall source load information and uploaded to the energy management center.

[0037] S12. The energy management center optimizes and allocates power within the source-load cluster based on the overall source-load information, and obtains information on surplus / deficient power.

[0038] As described above, the source-load aggregation controller directly controls the distributed source loads, packaging various source loads into large-scale response resources. The energy management center of the source-load cluster directly regulates each source-load aggregation controller, enabling energy complementarity between the controllers and maximizing internal balance within the cluster. The entire cluster possesses a certain degree of autonomy and can perform its own energy scheduling.

[0039] Furthermore, the types of distributed power sources include electric vehicle load clusters, air conditioning load clusters, user energy storage clusters, photovoltaic clusters, and wind power clusters.

[0040] As can be seen from the above description, distributed source loads include the above types as a specific embodiment of the present invention.

[0041] Furthermore, the power aggregation of the electric vehicle load in step S11 includes the following steps:

[0042] When electric vehicle i is charged at a constant power at time t, its state of charge can be expressed as:

[0043]

[0044] In the formula, P represents the state of charge of the electric vehicle at time t. i EV The charging power of the electric vehicle; the charging state of the electric vehicle at time t, 1 for charging and 0 for not charging; For electric vehicle battery capacity; Electric vehicle charging efficiency is the percentage of electrical energy consumed during charging relative to the energy stored in the battery.

[0045] Relationship between SOC (State of Charge) of electric vehicles, initial battery level, and driving range:

[0046]

[0047] In the formula, SOC (State of Charge) at the moment of departure after the electric vehicle has been fully charged during the charging period; The initial SOC during the charging period of the electric vehicle; d is the driving range of the electric vehicle outside the charging period; D is the driving range of the electric vehicle.

[0048] For a cluster of electric vehicles, the probability density of the charging start time is:

[0049]

[0050] In the formula, t st The charging start time for electric vehicles; μ is the mean of the charging start time; σ is the variance of the charging start time.

[0051] Range constraints of electric vehicle SOC:

[0052]

[0053] In the formula, This is the preset minimum state of charge for electric vehicles; This is the preset maximum state of charge value for electric vehicles;

[0054] The power of N electric vehicle charging groups is:

[0055]

[0056] The Monte Carlo algorithm is used to sample the initial charging time and daily mileage based on probabilistic characteristics, and then the charging power of each electric vehicle load is obtained:

[0057] a. Input the number of electric vehicles N EV Battery capacity C EV Charging power P EV And the number of Monte Carlo simulations, M;

[0058] b. Initialize N according to the normal distribution EV Electric vehicle charging time t st and initial state of charge Calculate the charging time for NEV electric vehicles:

[0059]

[0060] Where i = 1, 2, 3, ..., N EV ;

[0061] c. Calculate the minute-level load curve for the i-th electric vehicle:

[0062]

[0063] During charging The remaining time is 0;

[0064] d. Repeat steps b and c to calculate N. EV Electric vehicle load curve x i (i = 1, 2, 3, ..., N) EV ), and obtain the j-th Monte Carlo experiment N EV Total load added to electric vehicles;

[0065] e. Repeat steps b to d M times, and average the results of the M Monte Carlo experiments to obtain the daily load curve of the electric vehicle aggregation:

[0066]

[0067] As described above, the power of the electric vehicle load group is determined by the charging power of each electric vehicle and the charging time period. The charging time period of the electric vehicle is determined by factors such as the time of access to the charging station, the amount of electricity charged at the beginning of the charge, and the mileage during non-charging time. The Monte Carlo algorithm is used to sample the starting charging time and daily mileage based on probability characteristics to further obtain the charging power of each electric vehicle load.

[0068] Furthermore, the power aggregation of the air conditioning load in step S11 includes the following steps:

[0069] The basic model of a variable frequency air conditioner can be divided into a steady-state model and a dynamic model. In steady state, the indoor temperature T in,t It remains unchanged and equal to the air conditioner temperature setpoint T. set :

[0070] T in,t =T set ;

[0071] The steady-state power of the variable frequency air conditioner is:

[0072]

[0073] In the formula, T a Where A is the ambient temperature and A is the thermal conductivity. Energy efficiency ratio (EER) of inverter air conditioners;

[0074] During the dynamic transition period, the dynamic power and indoor temperature changes are as follows:

[0075]

[0076] In the formula, The dynamic power during the transition period; k is the power change coefficient, which is proportional to the change in the set temperature. When the set temperature rises, k < 0, and vice versa; R is the equivalent thermal resistance; C is the equivalent heat capacity.

[0077] Assuming that inverter air conditioners mostly operate in a steady-state state, the aggregate power of an inverter air conditioner can be approximated by the steady-state power of a single inverter air conditioner. Therefore, the aggregate power of an inverter air conditioner is:

[0078]

[0079] Using the Monte Carlo method, parameters for a batch of inverter air conditioners are generated within a reasonable numerical range based on a uniform distribution law. This simulates the steady-state power aggregation of a large number of inverter air conditioners. The aggregated power of the inverter air conditioners is then estimated using the expected values ​​of each calculated parameter. Therefore, the aggregated power of the inverter air conditioners is:

[0080]

[0081] As can be seen from the above description, air conditioning equipment, due to its strong adjustability, is also included in the source-load cluster, and the aggregation of air conditioning load is achieved through the above steps.

[0082] Furthermore, the power aggregation of the user energy storage in step S11 includes the following steps:

[0083] User energy storage charging and discharging efficiency It can be uniformly represented as:

[0084]

[0085] In the formula, This indicates the charging and discharging power of the energy storage system; a positive value indicates discharging, and a negative value indicates charging. Indicates charging efficiency; Indicates discharge efficiency;

[0086] Energy storage charging and discharging power constraints:

[0087]

[0088] Energy storage SOC constraints:

[0089]

[0090] In the formula, This represents the upper limit of the discharge power of a single energy storage unit; This refers to the upper limit of the charging power of a single energy storage unit; C ES This refers to the capacity of a single energy storage unit; The state of charge of a single energy storage unit; These represent the minimum and maximum states of charge of a single energy storage unit, respectively.

[0091] N energy storage charging and discharging power for:

[0092]

[0093] As described above, the power of the energy storage load group is determined by the charging and discharging power of each individual energy storage device.

[0094] Furthermore, the power aggregation of photovoltaics in step S11 includes the following steps:

[0095] N within the source load cluster pv The cluster power of photovoltaic equipment can be expressed as:

[0096]

[0097] in, This indicates the photovoltaic power of a single unit within the source-load cluster.

[0098] As described above, since the source-load cluster is composed of various source-load clusters that are geographically close, it can be assumed that the solar irradiance received by each distributed photovoltaic in the cluster is approximately the same. The power of each individual photovoltaic power station can be superimposed to realize the aggregation of photovoltaic power.

[0099] Furthermore, step S12 specifically includes:

[0100] The energy management center establishes an optimization model based on the overall source-load information and with the goal of optimizing the source-load matching degree. This model optimizes the distributed wind and solar power output and load behavior within the region and obtains information on surplus / deficient power.

[0101] As described above, an optimization model is established with the goal of optimizing the source-load matching degree to optimize the distributed wind and solar power output and load behavior within the region, which is a specific embodiment of the present invention.

[0102] Further, step S2 includes the following steps:

[0103] Electricity sold from the distribution network to all source-load clusters To establish an objective function that minimizes the supported power fluctuation rate, power allocation among the source-load clusters is performed:

[0104]

[0105] in:

[0106]

[0107]

[0108] In the formula, The electricity sold by the distribution network to source-load cluster i; The energy exchanged between source load clusters i and j; j∈i represents all source load clusters j connected to source load cluster i through a base station; The change in energy supplied by the distribution network to the source-load cluster i; The energy supplied by the distribution network to the source-load cluster i.

[0109] As described above, the company sells electricity to all source-load clusters. The objective function is established to minimize the power fluctuation rate of the source-load clusters, thereby achieving power allocation among the source-load clusters. At the same time, in order to further reduce the fluctuation impact of the source-load cluster access on the distribution network, the fluctuation of the interaction between the source-load clusters and the distribution network is considered in the objective function, in order to reduce the regulation burden of the power grid to a certain extent.

[0110] Please refer to Figure 2 A source-load cluster control system based on a honeycomb distribution network structure includes a source-load cluster, an energy interaction base station, and a distribution network control center.

[0111] The source-load cluster includes distributed loads, distributed power sources, an energy management center, and a distributed source-load aggregation controller.

[0112] The energy management center controls and communicates with multiple source-load aggregation controllers, and each source-load aggregation controller controls and communicates with all distributed loads / distributed power sources belonging to a certain type.

[0113] The source-load clusters form a honeycomb-like power distribution network structure through the energy interaction base station. Adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information exchange between each source-load cluster.

[0114] The energy management center communicates with the power distribution network control center through the energy interaction base station;

[0115] The source-load aggregation controller, the energy management center, and the distribution network control center jointly implement the steps in the above-described method for controlling a source-load cluster based on a honeycomb distribution network structure.

[0116] The present invention provides a source-load cluster control method and terminal based on a honeycomb distribution network structure, which is applicable to the scenario of optimized control of distribution networks, especially in the scenario where there are a large number of distributed power sources and loads connected to the distribution network.

[0117] Please refer to Figures 1 to 10 Embodiment 1 of the present invention is as follows:

[0118] A method for regulating source-load clusters based on a honeycomb distribution network structure includes the following steps:

[0119] In the honeycomb-shaped power distribution network, adjacent source-load clusters are interconnected through the energy interaction base station (hereinafter referred to as the base station), which monitors the energy and information interaction between each source-load cluster.

[0120] For reference Figure 2 Distributed resources in a source-load cluster can be divided into two categories: (1) wind power, photovoltaic power, and other relatively rigid and difficult-to-control power sources; and (2) air conditioning, user-side energy storage, and other relatively flexible and adjustable loads. These two types of resources, geographically close together, are combined to form a source-load cluster. Figure 1As can be seen, the architecture is mainly divided into four layers from top to bottom. Within each source-load cluster, there is an energy management platform. This platform governs the distributed source-load cluster controllers. Each distributed source-load cluster controller aggregates its own similar loads and packages them into large-scale response resources, giving them a certain degree of flexible adjustment capability and allowing them to be controlled by the energy management platform. The distribution network operator issues power commands through base stations, disseminating power flow information to each source-load cluster, as well as power transaction information with the upper-level power grid.

[0121] S1. The energy management center of each source-load cluster performs power optimization and allocation within its respective source-load cluster, and publishes surplus / deficient power through the energy interaction base station of the honeycomb distribution network.

[0122] Step S1 includes the following steps:

[0123] S11. Through the distributed source-load aggregation controllers of the source-load cluster, the distributed source-loads of the same type are aggregated to obtain the overall source-load information and uploaded to the energy management center.

[0124] The load aggregator within the source-load cluster first packages and aggregates similar distributed power sources and distributed loads. This invention selects electric vehicles, air conditioners, user-side energy storage, and distributed photovoltaics as examples to describe the aggregation process. The common aggregation approach for various distributed source-load types is described in [link to relevant documentation]. Figure 3 .

[0125] 1. Calculation of aggregated power of electric vehicle load groups

[0126] In this embodiment, electric vehicle charging facilities are considered an important distributed load in the source-load cluster. The charging characteristics of the electric vehicle load cluster within the microgrid are first analyzed to obtain the energy consumption characteristics of the corresponding charging facilities. Electric vehicles interact with the grid through the charging and discharging of their onboard batteries. The most critical calculation parameter for electric vehicle charging is the battery's state of charge (SOC, percentage of battery charge and capacity at time t).

[0127] (1) SOC model of a single electric vehicle

[0128] 1) When electric vehicle i is charged at a constant power at time t, its state of charge can be expressed as:

[0129]

[0130] In the formula, P represents the state of charge of the electric vehicle at time t. i EV The charging power of the electric vehicle; the charging state of the electric vehicle at time t, 1 for charging and 0 for not charging; For electric vehicle battery capacity; Electric vehicle charging efficiency is the percentage of electrical energy consumed during charging relative to the energy stored in the battery.

[0131] 2) Relationship between SOC and driving range of electric vehicles

[0132] To calculate the charging duration of an electric vehicle (EV), it is necessary to analyze the State of Charge (SOC) of the EV when it is connected to a charging station. The SOC of the EV battery is mainly affected by the daily mileage. In this embodiment, the battery charge consumption is directly proportional to the mileage. Therefore, the current SOC of the EV can be calculated from the initial charge level and the mileage. The relationship between the EV SOC and the initial charge level and mileage is as follows:

[0133]

[0134] In the formula, SOC (State of Charge) at the moment of departure after the electric vehicle has been fully charged during the charging period; The initial SOC for the electric vehicle during the charging period; d is the driving range of the electric vehicle outside the charging period; D is the driving range of the electric vehicle.

[0135] 3) Electric vehicle starting charging time

[0136] For a large-scale electric vehicle cluster, assuming that the charging start time follows a normal distribution, the probability density of the charging start time is:

[0137]

[0138] In the formula, t st The charging start time for electric vehicles; μ is the mean of the charging start time; σ is the variance of the charging start time.

[0139] 4) The State of Charge (SOC) of an electric vehicle is within a certain range; the range constraints of the State of Charge (SOC) of an electric vehicle are as follows:

[0140]

[0141] In the formula, This is the preset minimum state of charge for electric vehicles; This is the preset maximum state of charge value for electric vehicles;

[0142] 5) The total power of the N electric vehicle charging groups is:

[0143]

[0144] As shown in equations (1) to (5), the power of the electric vehicle load group is determined by the charging power of each electric vehicle and the charging time period. The charging time period is determined by factors such as the time of access to the charging station, the initial charging amount, and the mileage during non-charging time. In this embodiment, it is assumed that the electric vehicle load group contains N electric vehicles, the initial charging time of the electric vehicles follows a normal distribution, and the daily mileage follows a log-normal distribution. The Monte Carlo algorithm is used to sample the initial charging time and daily mileage based on probability characteristics to further obtain the charging power of each electric vehicle load.

[0145] Assuming the charging power of each electric vehicle is known and constant, the Monte Carlo algorithm is used to sample the starting charging time and daily mileage based on probabilistic characteristics to further obtain the charging power of each electric vehicle's load:

[0146] a. Input the number of electric vehicles N EV Battery capacity C EV Charging power P EV And the number of Monte Carlo simulations, M;

[0147] b. Initialize N according to the normal distribution EV Electric vehicle charging time t st and initial state of charge Calculate the charging time for NEV electric vehicles:

[0148]

[0149] Where i = 1, 2, 3, ..., N EV ;

[0150] c. Calculate the minute-level load curve for the i-th electric vehicle:

[0151]

[0152] During charging The remaining time is 0;

[0153] d. Repeat steps b and c to calculate N. EV Electric vehicle load curve x i (i = 1, 2, 3, ..., N) EV ), and obtain the j-th Monte Carlo experiment N EV Total load added to electric vehicles;

[0154] e. Repeat steps b to d M times, and average the results of the M Monte Carlo experiments to obtain the daily load curve of the electric vehicle aggregation:

[0155]

[0156] 2. Air Conditioning Load Cluster Modeling

[0157] Air conditioning equipment, due to its highly adjustable characteristics, is also included in the source-load cluster. The basic model of a variable frequency air conditioner can be divided into a steady-state model and a dynamic model. In steady state, the indoor temperature T... in,t It remains unchanged and equal to the air conditioner temperature setpoint T. set :

[0158] T in,t =T set ;

[0159] The steady-state power of the variable frequency air conditioner is:

[0160]

[0161] In the formula, T a Where A is the ambient temperature and A is the thermal conductivity. Energy efficiency ratio (EER) of inverter air conditioners;

[0162] During the dynamic transition period, the dynamic power and indoor temperature changes are as follows:

[0163]

[0164] In the formula, R is the dynamic power during the transition period; k is the power change coefficient, which is proportional to the change in the set temperature. When the set temperature rises, k < 0, and vice versa; R is the equivalent thermal resistance; C is the equivalent heat capacity.

[0165] In this embodiment, for the sake of model simplification, k is set to a constant of -0.15 when adjusting the temperature setpoint.

[0166] The inverter air conditioner with parameters taken from the average of Table 1 was operated for 20 minutes. The temperature setpoint was then increased from 26℃ to 27℃. The operating status can be obtained from the steady-state and dynamic load models of the inverter air conditioner as follows: Figure 4 As shown.

[0167] Table 1

[0168]

[0169]

[0170] Considering that users typically set the temperature of their inverter air conditioner to a comfortable range and rarely adjust it repeatedly, and even if some users do adjust the temperature setting, the adjustment range is unlikely to be large, it can be approximated that inverter air conditioners mostly operate in a steady-state state. Therefore, the aggregate power of an inverter air conditioner can be approximated by the steady-state power of a single inverter air conditioner. Thus, the aggregate power of an inverter air conditioner is:

[0171]

[0172] Considering the excessive workload involved in collecting the actual air conditioning parameters, the Monte Carlo method was used to generate batch inverter air conditioner parameters within a reasonable numerical range based on a uniform distribution law. This simulated the steady-state power aggregation of a large number of inverter air conditioners. Furthermore, to simplify the calculation, the expected values ​​of each parameter of the calculated inverter air conditioners were used to estimate the aggregated power of the inverter air conditioners. Therefore, the aggregated power of the inverter air conditioners is:

[0173]

[0174] 3. User Energy Storage Cluster Model

[0175] During periods of unstable power supply or peak grid load, energy storage devices can rapidly release and supply energy, ensuring power quality and stable grid operation. User-side energy storage solutions can also reduce grid construction and maintenance costs. Therefore, user-side energy storage loads interact with the grid through charging and discharging, and the key parameter for charging and discharging energy storage systems is also the State of Charge (SOC).

[0176] (1) SOC calculation model of single energy storage system

[0177] User energy storage charging and discharging efficiency It can be uniformly represented as:

[0178]

[0179] In the formula, This indicates the charging and discharging power of the energy storage system; a positive value indicates discharging, and a negative value indicates charging. Indicates charging efficiency; Indicates discharge efficiency;

[0180] 1) Energy storage charging and discharging power constraints:

[0181]

[0182] 2) Energy storage SOC constraints:

[0183]

[0184] In the formula, This represents the upper limit of the discharge power of a single energy storage unit; This refers to the upper limit of the charging power of a single energy storage unit; C ES This refers to the capacity of a single energy storage unit; The state of charge of a single energy storage unit; These represent the minimum and maximum states of charge of a single energy storage unit, respectively.

[0185] (2) Charge and discharge power of N energy storage units for:

[0186]

[0187] As can be seen from equations (13) to (15), the power of the energy storage load group is determined by the charging and discharging power of each individual energy storage device.

[0188] 4. Photovoltaic Cluster Model

[0189] In this embodiment, a photovoltaic cluster model is established using distributed photovoltaic as an example. Since the source-load cluster is composed of various source-load clusters that are geographically close, it can be assumed that the solar irradiance received by each distributed photovoltaic in the cluster is approximately the same, and the power of each individual photovoltaic power station can be superimposed.

[0190] Therefore, N within the source load cluster pv The cluster power of photovoltaic equipment can be expressed as:

[0191]

[0192] in, This indicates the photovoltaic power of a single unit within the source-load cluster.

[0193] The modeling approach is the same for distributed wind power as for other distributed photovoltaic power.

[0194] S12. The energy management center optimizes and allocates power within the source-load cluster based on the overall source-load information, and obtains information on surplus / deficient power.

[0195] Step S12 is as follows:

[0196] The energy management center establishes an optimization model based on the overall source-load information and with the goal of optimizing the source-load matching degree. This model optimizes the distributed wind and solar power output and load behavior within the region and obtains information on surplus / deficient power.

[0197] The interconnection topology between the various source payload clusters is shown in the figure. Figure 5 The solution process for the source-load cluster hierarchical regulation model can be found in [link to solution]. Figure 6 In the steps of this invention, the solution process is mainly divided into two steps: priority complementarity of power among source-load aggregation controllers within the source-load cluster, and allocation of surplus / deficient power among source-load clusters.

[0198] (1) Power priority complementarity among source-load aggregation controllers within the source-load cluster.

[0199] Within each source-load cluster, self-regulation is prioritized for the day-ahead phase based on the principle of optimal economic efficiency. Each distributed source-load aggregation controller is responsible for aggregating a massive number of dispersed distributed source-loads of the same type and uploading the aggregated source-load information, i.e., the overall source-load information (predicted power curves, control margins, etc.), to the source-load cluster energy management center. The energy management center obtains the power plan for each distributed source-load aggregation group through scheduling strategies. From the above, the power plan within the source-load cluster, as well as the surplus / deficit power of the entire source-load cluster, can be obtained.

[0200] Within the source-load cluster, it is necessary to realize the consumption of internal new energy sources.

[0201] In this embodiment, an optimization model is established with the goal of optimizing the source-load matching degree to optimize the distributed wind and solar power output and load behavior within the region.

[0202] In this embodiment, the distributed power source in source-load cluster i is taken as wind power as an example, and its output sequence is set as follows. First, standardize it:

[0203]

[0204] In the formula, The total electrical energy of distributed wind power. This represents the total electrical energy generated by distributed wind power (per unit). This represents the maximum limit of the total energy of distributed wind power.

[0205] Let the output sequence of the distributed load (the total power of air conditioning, electric vehicles, and user-side energy storage) be as follows: Standardize it:

[0206]

[0207] In the formula, The total electrical energy of distributed loads. This represents the total electrical energy (per unit) of the distributed load. This represents the maximum limit of total electrical energy for distributed loads.

[0208] The degree of matching between the source and the load can be expressed as:

[0209]

[0210] Furthermore, the overall energy fluctuation of the source-load cluster should not be too large, which can be expressed as:

[0211]

[0212] In the formula, This represents the overall surplus / deficit energy change (per unit value) of the source-load cluster. The total surplus / deficit energy of the source load cluster (per unit value).

[0213] Therefore, we can consider setting an objective function:

[0214]

[0215] The optimized energy characteristics of the source-load cluster are obtained, and its overall demand characteristics are obtained.

[0216] In this embodiment, the constraints of the above model include the power balance constraint (22) within the source-load cluster, and the energy consumption characteristics of each load aggregation group in step two, including equations (1)-(16).

[0217]

[0218] This represents the total power of the source load cluster interacting with the outside world.

[0219] S2. The distribution network control center of the distribution network performs secondary power flow allocation based on the surplus / deficit power released by each of the energy management centers and the interconnection status between each of the source-load clusters. Only when the power between each of the source-load clusters cannot be complementary will the distribution network provide further power support.

[0220] In this embodiment, after optimization in step S1, each source-load cluster obtains surplus / deficient power and reports it to the energy interaction base station of the distribution network. The upper-level distribution network control center performs optimized allocation of the aforementioned power, that is, based on the topology of the honeycomb distribution network and the interconnection between each source-load cluster, it performs secondary power flow allocation, prioritizing power allocation between each source-load cluster. When the power between different source-load clusters cannot be complementary, the upper-level distribution network provides further power support. Since each source-load cluster is connected to the distribution network as a whole, it is necessary to reduce the regulation burden on the grid to a certain extent. Therefore, the fluctuation rate of the power supported by the upper-level grid should also be considered as an optimization target.

[0221] Step S2 includes the following steps:

[0222] Let the power surplus / deficit of a certain source-load cluster be... (A positive value indicates surplus power, and a negative value indicates deficit power), representing the electricity sold by the distribution network to all source-load clusters. To establish an objective function that minimizes the supported power fluctuation rate, power allocation among the source-load clusters is performed:

[0223]

[0224] in:

[0225]

[0226]

[0227] In the formula, The electricity sold by the distribution network to source-load cluster i; The energy exchanged between source load clusters i and j; j∈i represents all source load clusters j connected to source load cluster i through a base station; The change in energy supplied by the distribution network to the source-load cluster i; The energy supplied by the distribution network to the source-load cluster i. Equations (23) and (24) above represent the relationship between the support received from the upper-level grid and the surplus / deficit power for a certain source-load cluster.

[0228] Meanwhile, in order to further reduce the volatility impact of the source-load cluster access on the distribution network, the volatility of the interaction between the source-load cluster and the distribution network is considered in the objective function. Here, the volatility index of the power interaction between the distribution network and each source-load cluster is defined, which can be expressed as the above formula (25).

[0229] Searchable Figure 7-9 This represents the power self-scheduling results of a power source-load cluster in a high-tech zone, industrial zone, and residential area within a certain city. Figure 10 The diagram illustrates the power interactions between various source-load clusters and between these clusters and the distribution network. As can be seen from the figure, this invention connects the source-load clusters to the distribution network as a whole, resulting in smaller power curve fluctuations and significantly reducing the control burden on the distribution network.

[0230] Please refer to Figure 2 and Figure 4 Embodiment two of the present invention is as follows:

[0231] A source-load cluster control system based on a honeycomb distribution network structure includes a source-load cluster, an energy interaction base station, and a distribution network control center.

[0232] The source-load cluster includes distributed loads, distributed power sources, an energy management center, and a distributed source-load aggregation controller.

[0233] The energy management center controls and communicates with multiple source-load aggregation controllers, and each source-load aggregation controller controls and communicates with all distributed loads / distributed power sources belonging to a certain type.

[0234] The source-load clusters form a honeycomb-like power distribution network structure through the energy interaction base station. Adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information exchange between each source-load cluster.

[0235] The energy management center communicates with the power distribution network control center through the energy interaction base station;

[0236] The source-load aggregation controller, the energy management center, and the distribution network control center jointly implement the steps in the control method of source-load cluster based on honeycomb distribution network structure described in Embodiment 1 above.

[0237] In this embodiment, the source-load cluster is mainly an organic collection of various distributed power sources and distributed loads, with relatively few actively regulated power sources. As the direct regulating entity of distributed power sources and loads, the source-load cluster has the following structural characteristics and functions:

[0238] (1) The control architecture of the source-load cluster mainly consists of the following four layers: a large number of distributed loads and power sources, distributed source-load aggregation controllers, source-load cluster energy management center, and distribution network operation center.

[0239] (2) Structurally, distributed power sources include distributed photovoltaic and distributed wind power, and distributed loads include distributed load facilities with regulation potential such as air conditioning and electric vehicle charging facilities; the source-load aggregation controller aggregates similar distributed sources and loads, and the source-load cluster energy management center directly controls each source-load aggregation controller.

[0240] (3) In terms of operation, the source load aggregation controller directly controls the distributed source loads and packages various source loads into large-scale response resources; the source load cluster control center directly regulates each source load aggregation controller, so that the energy of each source load aggregation controller is complementary and the internal balance of the source load cluster is achieved as much as possible.

[0241] (4) The distribution network operation center directly regulates the energy interaction between various source load clusters and between the source load clusters and the distribution network, thereby avoiding direct regulation of massive distributed source loads.

[0242] From the above, we can derive a hierarchical and clustered control process: regulating individual source-load clusters, regulating individual source-load aggregation controllers, and regulating massive distributed source-loads. The entire source-load cluster possesses a certain degree of autonomy and is capable of self-energy scheduling.

[0243] In this embodiment, the honeycomb distribution network serves as a structure connecting the source-load cluster to the external network. The energy interaction base station is the core equipment of the honeycomb distribution network. The honeycomb distribution network adapted to the hierarchical control architecture of the source-load cluster in step one has the following characteristics:

[0244] (1) Different source load clusters are interconnected through base stations in the distribution network. The energy interaction base station is designed based on flexible soft switch.

[0245] (2) The base station is responsible for controlling the energy interaction, and can monitor the energy and information interaction between each source load cluster in real time, and receive the surplus / deficit power reported by each source load cluster and report it to the distribution network operation center.

[0246] (3) The power distribution network operation center allocates the power declared by each source-load cluster and sends the relevant energy / information through the base station to realize the joint scheduling between each source-load cluster.

[0247] (4) The base station also serves as a connection point between the source-load cluster and the upper-level power grid, meeting the energy needs of the source-load cluster from the upper-level distribution network.

[0248] The source-load aggregation controller directly aggregates and regulates a large number of similar distributed source loads to form large-scale resources, which are then controlled by the source-load cluster center. The various source-load clusters can be interconnected through a honeycomb distribution network. Since the capacity and characteristics of the distributed source loads contained in different source-load clusters vary greatly, the operators of different source-load clusters can trade energy, thereby achieving mutual support and complementarity of energy, and maximizing the utilization of the resources configured in each source-load cluster.

[0249] The hierarchical and clustered control operation between source and load clusters mainly includes the following two steps:

[0250] (1) The energy management center of the source-load cluster performs economic optimization within the source-load cluster. Each distributed source-load aggregation controller first aggregates the dispersed distributed sources and loads to receive instructions from the energy management center of the source-load cluster in order to respond to resources on a large scale; each energy management center of the source-load cluster optimizes its own source-load cluster with the goal of achieving the most economical operation and realizes autonomous energy scheduling as much as possible; each source-load cluster publishes surplus / deficit power curves through the base station.

[0251] (2) The distribution network operation center performs power allocation among source-load clusters. Since different source-load clusters have different temporal characteristics, they are interconnected through a cellular distribution network structure. Simultaneously, the source-load clusters are mainly clusters of distributed power sources and distributed loads. Distributed power source output fluctuates significantly, leading to energy shortages in the source-load clusters. This necessitates power supply from the upper-level grid. The source-load clusters are interconnected with the upper-level grid through the base stations of the cellular distribution network to obtain energy from the upper-level grid. The above process is optimized by the distribution network operation center through the base stations described in step two, aiming for the most reasonable power allocation, thereby achieving reasonable adjustment of surplus / deficit power among various source-load clusters.

[0252] This invention first establishes a source-load cluster structure containing distributed power sources and loads, elucidating the aggregation characteristics of typical distributed power sources and loads such as photovoltaics and air conditioners. Various source-load aggregation controllers aggregate distributed sources and loads of similar scale to obtain relevant source-load characteristics. The source-load cluster operates autonomously to a certain extent, directly regulated by various source-load aggregation controllers through an energy management system. Simultaneously, the distribution network operator allocates surplus / deficient power to each source-load cluster. Each source-load cluster communicates with other source-load clusters through distribution network base stations, achieving mutual power support and backup, ensuring maximum utilization of allocated resources. Furthermore, the source-load cluster connects to the upper-level distribution network, receiving support during power shortages. This reduces the control burden on distribution operators: shifting from regulating massive distributed resources to directly controlling source-load clusters. Therefore, this invention, based on the topology of a cellular distribution network, establishes a hierarchical and grouped control model for source-load clusters.

[0253] In summary, the present invention provides a source-load cluster control method and system based on a honeycomb distribution network structure. It establishes a hierarchical and clustered operation mechanism, allowing the distribution network to directly control the source-load clusters, avoiding direct control of dispersed distributed source-load devices. This provides a control approach for massive distributed source-load access to the distribution network. The various source-load clusters are connected through the honeycomb distribution network, and energy and information exchange is achieved through the base stations of the honeycomb distribution network, resulting in strong scalability. Energy complementarity exists between different source-load clusters, while the source-load clusters receive support from the upper-level power grid, thereby maximizing the utilization of distributed power sources while ensuring the power quality of distributed loads.

[0254] In other words, compared with the prior art, the present invention has the following advantages:

[0255] (1) Facilitates unified control: A source-load cluster composed of distributed power sources and distributed loads is established, and each distributed load and distributed power source cluster aggregates its subordinate dispersed distributed source loads.

[0256] (2) It alleviates the regulation burden of distribution operators: By establishing a hierarchical and clustered operation mechanism, the distribution network only needs to directly regulate the source load cluster, avoiding direct regulation of the dispersed distributed source load equipment, and providing a regulation approach for the massive distributed source load to access the distribution network.

[0257] (3) The various source-load clusters are connected through a honeycomb distribution network, and energy and information are exchanged through the base stations of the honeycomb distribution network. It also has strong scalability.

[0258] (4) Improved power quality of the new power system: The energy characteristics of different source load clusters are complementary, and the source load clusters are supported by the upper-level power grid, thereby maximizing the utilization of distributed power sources while ensuring the power quality of distributed loads.

[0259] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for regulating source-load clusters based on a honeycomb distribution network structure, characterized in that, Including the following steps: S1. The energy management center of each source-load cluster performs power optimization and allocation within its respective source-load cluster, and publishes surplus / deficient power through the energy interaction base station of the honeycomb distribution network. S2. The distribution network control center of the distribution network performs secondary power flow allocation based on the surplus / deficit power released by each of the energy management centers and the interconnection status between each of the source-load clusters. Only when the power between each of the source-load clusters cannot be complementary will the distribution network provide further power support. In the honeycomb-shaped power distribution network, adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information interaction between each source-load cluster. Step S1 includes the following steps: S11. Through the distributed source load aggregation controllers of the source load cluster, the distributed source loads of the same type are aggregated to obtain the overall source load information and uploaded to the energy management center. S12. The energy management center optimizes and allocates power within the source-load cluster based on the overall source-load information, and obtains information on surplus / deficient power. The types of distributed power sources include electric vehicle load clusters, air conditioning load clusters, user energy storage clusters, photovoltaic clusters, and wind power clusters; Step S2 includes the following steps: Electricity sold from the distribution network to all source-load clusters To establish an objective function that minimizes the supported power fluctuation rate, power allocation among the source-load clusters is performed: ； in: ； ； In the formula, For distribution network to source-load cluster i Electricity sales volume; For source load cluster i, j Energy exchange between them; Represents all clusters related to the source load. i Source-payload clusters connected via base stations j ; For distribution network to source-load cluster i Changes in energy supply; For distribution network to source-load cluster i The energy supplied.

2. The control method for source-load clusters based on a honeycomb distribution network structure according to claim 1, characterized in that, Step S11, power aggregation of electric vehicle load, includes the following steps: electric vehicles i When charged at a constant power at time t, its state of charge can be expressed as: ； In the formula, For electric vehicles t State of charge at any given moment; The charging power for electric vehicles; For electric vehicles at all times t The charging status is 1 for charging and 0 for not charging. For electric vehicle battery capacity; Electric vehicle charging efficiency is the percentage of electrical energy consumed during charging relative to the energy stored in the battery. Relationship between SOC (State of Charge) of electric vehicles, initial battery level, and driving range: ； In the formula, The time to leave after the electric vehicle is fully charged. SOC ; Initial period of charging electric vehicles SOC ; d The driving range of an electric vehicle outside of charging time; D For the driving range of electric vehicles; For a cluster of electric vehicles, the probability density of the charging start time is: ； In the formula, t st Start time for charging electric vehicles; μ This represents the average value at the start of charging. σ Variance of the charging start time; Range constraints of electric vehicle SOC: ； In the formula, This is the preset minimum state of charge for electric vehicles; This is the preset maximum state of charge value for electric vehicles; N The power capacity of the electric vehicle charging group in Taiwan is: ； The Monte Carlo algorithm is used to sample the initial charging time and daily mileage based on probabilistic characteristics, and then the charging power of each electric vehicle load is obtained: a. Input the number of electric vehicles N EV Battery capacity C EV Charging power P EV and the number of Monte Carlo simulations M ; b. Initialize according to the normal distribution N EV Electric vehicle charging time and initial state of charge Calculate the charging time for NEV electric vehicles: ； in, i =1, 2, 3, ... N EV ; c. Calculate the first i Minute-level load curves for electric vehicles: ； During charging , The remaining time is 0; d. Repeat steps b and c to calculate. N EV Electric vehicle load curve x i ( i =1, 2, 3, ... N EV ), and obtain the first j Monte Carlo experiment N EV Total load added to electric vehicles; e. Repeat steps b to d M times, and then... M The daily load curve of electric vehicles was obtained by averaging the results of the Monte Carlo experiments. 。 3. The control method for source-load clusters based on a honeycomb distribution network structure according to claim 1, characterized in that, Step S11, power aggregation of the air conditioning load, includes the following steps: The basic model of a variable frequency air conditioner can be divided into a steady-state model and a dynamic model. In steady state, the indoor temperature... T in,t Keeps constant and equal to the air conditioner temperature setting. T set : T in,t =T set ； The steady-state power of the variable frequency air conditioner is: ； In the formula, For the outside temperature, Thermal conductivity; Energy efficiency ratio (EER) of inverter air conditioners; During the dynamic transition period, the dynamic power and indoor temperature changes are as follows: ； In the formula, Dynamic power during the transition period; k This is the power change coefficient, proportional to the change in temperature. When the set temperature rises, k <0, otherwise k >0; R Equivalent thermal resistance; C Equivalent heat capacity; Assuming that inverter air conditioners mostly operate in a steady-state state, the aggregate power of an inverter air conditioner can be approximated by the steady-state power of a single inverter air conditioner. Therefore, the aggregate power of an inverter air conditioner is: ； Using the Monte Carlo method, parameters for a batch of inverter air conditioners are generated within a reasonable numerical range based on a uniform distribution law. This simulates the steady-state power aggregation of a large number of inverter air conditioners. The aggregated power of the inverter air conditioners is then estimated using the expected values ​​of each calculated parameter. Therefore, the aggregated power of the inverter air conditioners is: 。 4. The control method for source-load clusters based on a honeycomb distribution network structure according to claim 1, characterized in that, Step S11, power aggregation of user energy storage, includes the following steps: User energy storage charging and discharging efficiency It can be uniformly represented as: ； In the formula, This indicates the charging and discharging power of the energy storage system; a positive value indicates discharging, and a negative value indicates charging. Indicates charging efficiency; Indicates discharge efficiency; Energy storage charging and discharging power constraints: ； Energy storage SOC constraints: ； In the formula, This represents the upper limit of the discharge power of a single energy storage unit; This represents the upper limit of the charging power of a single energy storage unit. C ES This refers to the capacity of a single energy storage unit; The state of charge of a single energy storage unit; , These represent the minimum and maximum states of charge of a single energy storage unit, respectively. N Taiwan's energy storage charging and discharging power for: 。 5. The method for regulating a source-load cluster based on a honeycomb distribution network structure according to claim 1, characterized in that, Step S11, the power aggregation of photovoltaics, includes the following steps: within the source load cluster N pv The cluster power of photovoltaic equipment can be expressed as: ； in, This indicates the photovoltaic power of a single unit within the source-load cluster.

6. The control method for source-load clusters based on a honeycomb distribution network structure according to claim 1, characterized in that, Step S12 is as follows: The energy management center establishes an optimization model based on the overall source-load information and with the goal of optimizing the source-load matching degree. This model optimizes the distributed wind and solar power output and load behavior within the region and obtains information on surplus / deficient power.

7. A control system for source-load clusters based on a honeycomb distribution network structure, characterized in that, This includes source-load clusters, energy interaction base stations, and distribution network control centers; The source-load cluster includes distributed loads, distributed power sources, an energy management center, and a distributed source-load aggregation controller. The energy management center controls and communicates with multiple source-load aggregation controllers, and each source-load aggregation controller controls and communicates with all distributed loads / distributed power sources belonging to a certain type. The source-load clusters form a honeycomb-like power distribution network structure through the energy interaction base station. Adjacent source-load clusters are interconnected through the energy interaction base station, which monitors the energy and information exchange between each source-load cluster. The energy management center communicates with the power distribution network control center through the energy interaction base station; The source-load aggregation controller, the energy management center, and the distribution network control center jointly implement the steps in the control method of source-load cluster based on honeycomb distribution network structure as described in any one of claims 1-6.

Images