A virtual power plant flexible resource dynamic aggregation response capability evaluation method

Abstract

The present application relates to a kind of single cluster aggregation response potential evaluation method considering uncertain factor, comprising the following steps: constructing EVs resource state model obtains EVs resource state model parameter;TCLs resource state model is constructed, and obtains TCLs resource related key state model parameter;EVs resource priority state queue is constructed;TCLs resource priority state queue is constructed, and the dynamic aggregation response capability of VPP cluster is evaluated.The virtual power plant flexible resource dynamic aggregation response capability evaluation method provided by the present application.The method constructs the unified state control model of VPP internal adjustable resource, greatly reduces the complexity of VPP management large adjustable resource.Meanwhile, the VPP dynamic resource aggregation response capability evaluation method based on state priority queue is proposed, by dividing different response cluster, and forming response state priority queue in cluster, solve the problem that when virtual power plant participates in grid balancing regulation, it is difficult to determine the resource cluster and its response amount of participating response.

Description

Technical Field

[0001] This invention belongs to the field of smart grids, and in particular, it is a method for evaluating the dynamic aggregation response capability of flexible resources in a virtual power plant. Background Technology

[0002] The rapid development of smart grids has led to a shift from traditional centralized power plants to undispatchable distributed renewable energy generators, bringing numerous challenges to power system operation. One such challenge is maintaining power system balance under conditions of randomly distributed renewable energy. Currently, China's power system primarily relies on centralized fossil fuel power plants to provide ancillary services. However, with the increasing number of renewable energy units, system uncertainty is rising, making it difficult for traditional power plants to stabilize system frequency by providing rotational inertia. By leveraging information communication and advanced measurement technologies to achieve bidirectional interactive smart electricity consumption between sources and loads, load-side resources can provide grid balancing and regulation services such as peak shaving, frequency regulation, and reserve capacity.

[0003] Demand response refers to the behavior of electricity users proactively changing their existing electricity consumption patterns in response to market price signals or incentive mechanisms. Demand response utilizes the load side as a substitute resource for the supply side's electricity, which can not only alleviate the predicament of electricity supply and demand imbalance but also promote environmental protection and energy conservation. Generators providing ancillary services are constrained by ramp rates, resulting in slower response times and higher costs. If conventional units were to handle new ancillary service demands, it would lead to huge construction investments, reduce system operating efficiency, and make it difficult to meet the safe operation requirements of high-power-receiving grids in some parts of China. With adequate communication infrastructure, most demand response resources respond faster than generators to both dispatch system commands and price signals. However, loads are susceptible to various objective factors such as production activities and unforeseen events, and are spatially dispersed with overlapping response times, which presents certain difficulties for these resources to participate in grid balancing and regulation services.

[0004] Virtual power plants (VPPs) represent a type of multi-energy integrated management system that can fully utilize the characteristics of various resource types to mitigate the uncertainty of load-side resources participating in ancillary services. A virtual power plant (VPP) is an integrated power plant composed of an energy management system and the small and micro distributed energy resources it controls. The distributed energy resources it includes can be distributed generator sets, distributed energy storage devices, or demand response resources distributed among numerous demand-side users. Through the coordinated operation of various internal resources, a virtual power plant can address the uncertainty of its own distributed generation while also providing additional balancing and regulation services to the power grid. Therefore, research on the dynamic aggregation response potential among various resources within a virtual power plant is of great significance.

[0005] However, when virtual power plants participate in grid balancing regulation, it is difficult to determine the resource clusters participating in the response and their response quantities. When resources within a virtual power plant participate in grid balancing regulation, there is a problem of output fluctuation across a continuous time span. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for evaluating the dynamic aggregation response capability of flexible resources in a virtual power plant (VPP) for grid balance regulation. This method solves the problem of power output fluctuation on a continuous time cross-section when resources within a VPP participate in grid balance regulation. The method divides the EV and TCL resources within the VPP into clusters according to the equipment operating status, and forms a response status priority queue within the cluster based on the response capability indicators of different resources. When some loads cannot participate in the response due to special reasons, the VPP can arrange new response resources to participate in grid balance regulation according to the status priority queue, thereby enabling the VPP to participate in grid balance regulation stably.

[0007] This invention addresses the trend of virtual power plant adjustable resource clusters participating in grid demand response services by providing a method for evaluating the aggregated response potential of a single cluster, considering uncertainties. The method includes the following steps:

[0008] Step 1: Construct an adjustable load state model

[0009] Step 1.1: Construct the EVs resource state model and obtain the EVs resource state model parameters;

[0010] Step 1.2: Construct a TCLs resource state model to obtain key state model parameters related to TCLs resources;

[0011] Step 2: Construct a response status priority queue

[0012] Step 2.1: Construct the EVs resource priority state queue;

[0013] Step 2.2: Construct the TCLs resource priority state queue;

[0014] Step 3: Evaluate the dynamic aggregation response capability of the VPP cluster

[0015] Step 3.1: Construct the aggregated state model;

[0016] Step 3.2: Construct a state control model;

[0017] Obtain the running status parameters of EVs at time t, divide them into multiple clusters, calculate the RTM and SOCM indices within the clusters, form an EVs state priority queue based on the indices, and calculate the maximum responsive potential of EVs based on the state control model.

[0018] Obtain the key state variables of TCLs at time t, divide TCLs into multiple clusters based on the key state variables, calculate the RTC and RTM indices within the clusters, form a TCLs state priority queue based on the indices, and calculate the maximum responsive potential of TCLs based on the state control model.

[0019] Output the maximum response potential of each resource cluster and calculate the maximum response potential of each resource cluster at the next time step.

[0020] Furthermore, the EVs resource state model is as follows:

[0021] (2)

[0022] (3)

[0023] in, For EV i exist t Power output at any given moment; For EV i The normalized SOC value at time t varies in the range of [0, 1]. We can obtain the following from formula (4):

[0024] (4)

[0025] in, For EV i Battery capacity; and They represent EVs respectively i The charging and discharging efficiency.

[0026] Furthermore, the resource state model of the TCLs is obtained from formula (8):

[0027] (8)

[0028] in, for Predicted temperature at any given time; equal , and They represent TCL respectively l The equivalent thermal resistance and capacitance; and They represent TCL respectively l The equipment is in constant time t Indoor and outdoor temperatures; For TCL l The electrical power consumed at time t.

[0029] Furthermore, RTM refers to EVs that are in the network access state. i exist During the period from t From the moment on The duration of continuous participation in a given response method is abbreviated as: As shown in equation (11),

[0030] (11)

[0031] EVs with larger RTM values ​​are given priority when participating in the response.

[0032] Furthermore, the EV clusters are first sorted in descending order based on RTM. When RTMs are equal, EVs with larger SOCM are assigned a higher response priority. SOCM is directly related to battery SOC. When the output is adjusted downwards or upwards, the EVs... i exist SOCM value within the time period Should be respectively and Based on the standard, as shown in equation (12),

[0033] (12)

[0034] In the formula: for The adjustment demand of EVs during the period, a negative value means that the output of EVs needs to be reduced, and a positive value means that the output of EVs needs to be increased;

[0035] When EVs need to reduce their output, and The larger the difference, the greater the SOCM of the EV, which will prioritize stopping discharging and even starting charging; when EVs need to increase their output... and The larger the difference, the larger the SOCM of the EV, which will prioritize stopping charging or even starting to discharge.

[0036] Furthermore, the TCL clusters are first sorted in descending order based on RTM. If the RTMs are equal, the TCL with the larger RTC is given a higher adjustment priority. RTCM is directly related to the temperature tolerance set by the TCL. When the output is reduced or increased, the TCL... l exist RTCM values ​​within the time period Should be respectively and Based on the standard, as shown in equation (14),

[0037] (14)

[0038] In the formula: for The adjustment demand of TCLs during the time period. A negative value indicates that the load of TCLs needs to be increased, while a positive value means that the load needs to be decreased.

[0039] Furthermore, the aggregation state model is given by formulas (15)-(20):

[0040] (15)

[0041] (16)

[0042] (17)

[0043] (18)

[0044] (19)

[0045] (20)

[0046] in, , These represent the number of devices in the EVs and TCLs clusters, respectively.

[0047] The state control model for various adjustable resources within the VPP can be expressed by formula (21):

[0048] (twenty one)

[0049] Among them, It is the state matrix of the internal devices of VPP; It is the power output matrix of the internal devices of the VPP; It is the time interval matrix after the internal resources of VPP are adjusted.

[0050] Furthermore, when the power grid is t Always need When adjusting the capacity, adjustment is achieved by regulating the output of real-time power, i.e., by adding an output control matrix before the state transition matrix. Therefore, formula (22) can also be expressed as:

[0051] (twenty two)

[0052] It is a diagonal matrix. , of which elements , , , This is an adjustable resource output control variable used to increase or decrease the output of VPP aggregated resources. , Power control is achieved by controlling the discharge current; and DG j It is in charging state. This indicates that it is in a discharge state; TCL j The switch is in the "off" position. This indicates that it is in the "on" state. Adjusting power consumption by changing electricity prices Indicates DR j When operating at maximum power consumption, it is no longer possible to reduce VPP output by increasing load, but it has the maximum potential to increase VPP output at this time.

[0053] Based on the state control model in (22), the maximum potential for VPP output adjustment at any given time can be obtained further through formula (23):

[0054] (twenty three)

[0055] in, For the maximum up-adjustment control matrix, all diagonal elements are the maximum values ​​of the control variables; This is the current control matrix; The maximum or rated power of the equipment within the VPP; Let VPP be a vector matrix with all elements equal to 1. Similarly, the maximum potential for VPP output reduction can be obtained as follows:

[0056] (twenty four)

[0057] in, For the maximum down-adjustment control matrix, all diagonal elements are the minimum values ​​of the control variables.

[0058] The advantages and positive effects of this invention are:

[0059] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a method for evaluating the dynamic aggregation response capability of flexible resources in a virtual power plant. This method constructs a unified state control model for adjustable resources within the VPP, significantly reducing the complexity of managing a large number of adjustable resources in the VPP. Simultaneously, it proposes a method for evaluating the dynamic resource aggregation response capability of a VPP based on state priority queues. By dividing different response clusters and forming response state priority queues within each cluster, it solves the problem of difficulty in determining the resource clusters participating in the response and their response quantities when a virtual power plant participates in grid balance regulation. Attached Figure Description

[0060] Figure 1 A schematic diagram of the maximum controllable charge and discharge area of ​​a single EV;

[0061] Figure 2 This is a schematic diagram of the remaining adjustable time of TCL at time t;

[0062] Figure 3 A schematic diagram of the EV response mode breakdown;

[0063] Figure 4 This is a schematic diagram illustrating four different scenarios in Method I;

[0064] Figure 5 This is a schematic diagram of TCL's critical status cluster grouping;

[0065] Figure 6 This is a flowchart for assessing the downward adjustment capability of VPP output;

[0066] Figure 7 This is a schematic diagram illustrating the daily outdoor temperature variation.

[0067] Figure 8 These are the output simulation curves of various adjustable resources within the VPP over a day;

[0068] Figure 9 This is a schematic diagram showing the estimated maximum response capability of each adjustable resource cluster at various time points in VPP. Detailed Implementation

[0069] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0070] A method for evaluating the flexible resource dynamic aggregation response capability of a virtual power plant, the specific process of which is as follows: Figure 6 As shown, it includes the following steps:

[0071] Step 1: Construction of Adjustable Load State Model

[0072] Step 1.1: Construct the EVs resource state model and obtain the EVs resource state model parameters;

[0073] Step 1.2: Construct a TCLs resource state model to obtain key state model parameters related to TCLs resources.

[0074] Step 2: Construct a response status priority queue

[0075] Step 2.1: Construct the EVs resource priority state queue;

[0076] Step 2.2: Construct the TCLs resource priority state queue.

[0077] Step 3: VPP Cluster Dynamic Aggregation Response Capability Assessment

[0078] Step 3.1: Construct the aggregated state model;

[0079] Step 3.2: Construct a state control model.

[0080] Step 1.1 of the invention shown: Constructing an EVs resource state model and obtaining EVs resource state model parameters. Let EVs... i The grid connection and grid disconnection times are respectively and EVs do not have adjustment potential when not connected to the grid, therefore EVs i Only in It has adjustable potential. To prevent overcharging and discharging, the EV is designed with... i The upper and lower limits of SOC are respectively , Thus, the EV shown in Figure 1 can be obtained. i The maximum controllable charge / discharge range. EV i exist When connected to the power grid, the SOC is (point a), then along a - b - c Run, first in EV i Rated power Start charging, and then when SOC is reached Then switch to idle state; when EV i When VPP output needs to be increased after grid connection, immediately follow the... d - e - f To run, first use Discharge occurs when the SOC drops to... Discharge is stopped at a certain point, and then the SOC is ensured to reach EV level when disconnected from the grid. i Minimum battery power requirement for user travel Finally, to ensure the capacity requirements before departure, the EV i Forced charging.

[0081] At any given moment, EV i Adjustable capability and operating point within the controllable charging and discharging range Location-dependent, for example, at point P... Charging (PX) yields Figure 1 The horizontal axis range corresponding to PX represents the duration of its maximum charging capacity. To better meet travel needs, EVs... i After connecting to the power grid, the first step is to charge the device and maintain a State of Charge (SOC) of no less than [value missing]. Then, at a given time EV iThe adjustable capacity upper and lower limits are:

[0082] (1)

[0083] When evaluating the response capability of a single EV at any given time, the on-grid status and the upper and lower limits of SOC should be considered simultaneously. If the SOC is less than... Only then can it be charged, and the SOC should be maintained at no less than [a certain value] during charging. Therefore, the state model of EV can be obtained from formulas (2) and (3):

[0084] (2)

[0085] (3)

[0086] in, For EV i exist t Power output at any given moment; For EV i The normalized SOC value at time t varies in the range of [0, 1]. We can obtain the following from formula (4):

[0087] (4)

[0088] in, For EV i Battery capacity; and They represent EVs respectively i The charging and discharging efficiency.

[0089] VPP Acquisition , , , , , , , , and We need to establish a DGs operational status model based on the data.

[0090] Step 1.2 of the invention shown: Construct a TCLs resource state model to obtain key state model parameters related to TCLs resources. The operation of temperature control equipment such as air conditioners is periodic; under the influence of temperature control equipment and other heat transfer processes, the indoor temperature remains within a certain range. Fluctuations within the body. For example... Figure 2 As shown, in During this period, TCL devices are switched off, and the indoor temperature rises; During the period, TCL devices are in a powered-off state, while the indoor temperature rises. To simplify calculations, the research assumptions for TCLs are as follows. and This is a fixed value. For TCLs in the "on" state, when VPP output needs to be increased, select these TCLs to switch to the "off" state. At this time, the indoor temperature rises, and when the maximum temperature is reached... When the temperature drops to its lowest setting, the switch automatically switches to "on". For TCLs in the "off" state, when VPP output needs to be reduced, select these TCLs and switch them back to the "on" state. When the switch automatically switches to "off", the adjustable capacity upper and lower limits of TCL during any given time period can be obtained as follows:

[0091] (5)

[0092] in, For TCL l exist t The rated power consumed at all times.

[0093] exist t At that time, according to TCL l The on / off status is related to the current indoor temperature. The remaining adjustable time can be obtained from formula (6):

[0094] (6)

[0095] After normalizing the temperature state variable according to formula (7), the state model of TCL can be obtained from formula (8):

[0096] (7)

[0097] (8)

[0098] in, for Predicted temperature at any given time; equal , and They represent TCL respectively l The equivalent thermal resistance and capacitance; and They represent TCL respectively l The equipment is in constant time t Indoor and outdoor temperatures; For TCL l The electrical power consumed at time t.

[0099] VPP Acquisition , , , , , , and Based on the data, a TCLs operating status model is established.

[0100] Step 2.1 of the invention is shown: Constructing an EV resource priority state queue. This method decomposes the process of EVs participating in the response into... Figure 3 The four response modes shown are denoted as I, II, III, and IV. I and II represent VPP output upward adjustment modes, while III and IV represent VPP output downward adjustment modes. t At any given time, the EV's operating point depends on the adjustment method. This will cause movement, which in turn will affect Response capability over a given time period. Taking response method I as an example, such as... Figure 4 The study primarily examines four typical scenarios (A, B, C, D) where response capabilities change during the evaluation period, with H, I, J, and K corresponding to four different operational points. The analysis of operational status changes for other response modes is similar to that of Mode I.

[0101] 1) When At that time, EV i There are periods when the grid connection is not yet available. , Constantly connected to the power grid to participate in the discharge process, the corresponding response process is as follows: Figure 4 As shown in (a).

[0102] 2) When And there is a discharge response that makes At that time, in order to reduce battery consumption, EV i Discharge should be stopped, and the corresponding response power is as follows: Figure 4 (b) shown. EV i Time to reach the lower limit of the controllable region It can be obtained from equation (9).

[0103] (9)

[0104] 3) When Furthermore, there is a discharge response that causes the operating point to touch the forced charging boundary. ef To meet users' travel needs, it is necessary to mandate EVs. i During charging, the corresponding response power is as follows: Figure 4 (c) P - t As shown by the solid line in the graph. EV i Time to reach forced charging power It can be obtained from equation (10):

[0105] (10)

[0106] 4) When At that time, EV i exist After a certain time, it will be unable to participate in discharge due to disconnection from the power grid, and the corresponding response power is as follows: Figure 4 As shown in (d).

[0107] To account for the differences in individual response power and battery capacity within an EV cluster, a control strategy is proposed that comprehensively considers response time margin (RTM) and state of charge margin (SOCM) indicators. A state priority queue is generated by combining the RTM and SOCM indicators of each EV, selecting the best EVs to participate in the response while balancing scheduling and user travel needs. This strategy applies to EVs already in the network connection phase. i exist During the period from t From the moment on The duration of continuous participation in a given response mode is defined as its RTM under that mode, abbreviated as As shown in equation (11).

[0108] (11)

[0109] When EV i When in an offline state Setting it to 0 will give you When EVs participate in system optimization scheduling, those with longer continuous response times should be given priority to maintain stable output power regulation as much as possible and reduce EV state switching frequency. Therefore, EVs with larger RTM should be given priority when participating in response.

[0110] However, for There may be multiple EVs with the same RTM value within a certain time period (e.g., all have the same RTM). RTM alone cannot provide a clear ranking criterion, therefore SOCM is introduced. RTM is the primary reference metric, and SOCM is a secondary reference metric. First, the EV clusters are sorted in descending order based on RTM. When RTMs are equal, EVs with higher SOCM are then assigned a higher response priority. SOCM is directly related to battery SOC; when output is adjusted downwards or upwards, the EVs... i exist SOCM value within the time period Should be respectively and Based on the standard, as shown in equation (12).

[0111] (12)

[0112] In the formula: for The adjustment demand of EVs during the period. A negative value means that the output of EVs needs to be reduced, and a positive value means that the output of EVs needs to be increased.

[0113] When EVs need to reduce their output, and The larger the difference, the greater the SOCM of the EV, meaning it will be prioritized to stop discharging and even start charging; when EVs need to increase their output, and The larger the difference, the larger the SOCM of the EV, which means that it will be prioritized to stop charging or even start discharging.

[0114] Step 2.2 of the invention is shown: Constructing an EVs resource priority state queue. For multiple TCL units, their start-stop sequences are interleaved, therefore the adjustability of TCLs cannot be obtained by direct accumulation. Generally, when a TCL unit is in the on state, it can achieve load reduction, i.e., it has the potential to increase VPP output; when it is in the off state, it does not have the ability to reduce load, i.e., it has the potential to reduce VPP output. For a response time of... The scheduling requirements, based on the current time t The key state variable of 0 can be Each TCL unit is divided into 4 groups, such as Figure 5 As shown.

[0115] exist Figure 5 In summary: ① Group I is in the operating state, has the capacity to reduce load, and can meet the duration requirements. The load units within this group are suitable for participation in regulation, with the regulation power being their respective rated power. ② Group II is in the operating state, has the capacity to reduce load, but cannot meet the scheduling duration requirements, and is not suitable for participation in regulation. ③ Group III is in the off state, has no capacity to reduce load, and the duration... The internal system will not switch from a shutdown state to an on state, and will not affect the cluster's adjustment capability; ④ Group IV is in a shutdown state and has no load reduction capability. The load units within the group will remain in a shutdown state for a certain period of time. It will automatically start internally, which has a negative impact on the cluster's adjustment capabilities.

[0116] Therefore, it can be concluded that only the TCL units of groups I, II, and IV can handle durations of [duration missing]. Demand regulation plays a role, prioritizing groups I and IV when capacity is sufficient. Considering the differences in individual regulation power and remaining adjustable time within the TCL cluster, a regulation strategy is proposed that comprehensively considers the response time margin (RTM) and response temperature control margin (RTCM) indices for both upward and downward regulation.

[0117] TCL, which is currently in the network access status l exist During the period from t From the moment on The duration of continuous participation in a given regulation mode is defined as its RTM under that mode, abbreviated as As shown in equation (13).

[0118] (13)

[0119] However, for There may be multiple TCLs with the same RTM value within a given time period (e.g., all have the same RTM). RTM alone cannot provide a clear sorting criterion, so RTC is introduced. First, the TCL clusters are sorted in descending order based on RTM. If RTMs are equal, then the TCL with the larger RTC is given a higher adjustment priority. RTCM is directly related to the temperature tolerance set by the TCL; when output is reduced or increased, the TCL... l exist RTCM values ​​within the time period Should be respectively and Based on the standard, as shown in equation (14).

[0120] (14)

[0121] In the formula: for The adjustment demand of TCLs during the time period. A negative value indicates that the load of TCLs needs to be increased, while a positive value means that the load needs to be decreased.

[0122] Step 3.1 of the invention shown: Constructing a unified state model. Based on steps 1.1, 1.2, 2.1, and 2.2, a unified state model for different adjustable devices is established as formulas (15)-(20):

[0123] (15)

[0124] (16)

[0125] (17)

[0126] (18)

[0127] (19)

[0128] (20)

[0129] in, , These represent the number of devices in the EVs and TCLs clusters, respectively.

[0130] Therefore, the state control model of various adjustable resources within the VPP can be expressed by formula (21):

[0131] (twenty one)

[0132] Among them, It is the state matrix of the internal devices of VPP; It is the power output matrix of the internal devices of the VPP; It is the time interval matrix after the internal resources of VPP are adjusted.

[0133] Step 3.2 shown: Constructing a state control model. When the power grid is in... t Always need When adjusting the capacity, adjustment is achieved by regulating the output of real-time power, i.e., by adding an output control matrix before the state transition matrix. Therefore, formula (22) can also be expressed as:

[0134] (twenty two)

[0135] It is a diagonal matrix. , of which elements , , , This is an adjustable resource output control variable used to increase or decrease the output of VPP aggregated resources. , Power control is achieved by controlling the discharge current; and DG j It is in charging state. This indicates that it is in a discharge state; TCL j The switch is in the "off" position. This indicates that it is in the "on" state. Adjusting power consumption by changing electricity prices Indicates DRj When operating at maximum power consumption, it can no longer reduce the VPP output by increasing the load, but it has the maximum potential to increase the VPP output.

[0136] Based on the state control model in (22), the maximum potential for VPP output adjustment at any given time can be obtained further through formula (23):

[0137] (twenty three)

[0138] in, For the maximum up-adjustment control matrix, all diagonal elements are the maximum values ​​of the control variables; This is the current control matrix; The maximum or rated power of the equipment within the VPP; Let VPP be a vector matrix with all elements equal to 1. Similarly, the maximum potential for VPP output reduction can be obtained as follows:

[0139] (twenty four)

[0140] in, For the maximum down-adjustment control matrix, all diagonal elements are the minimum values ​​of the control variables.

[0141] Example:

[0142] Set the number of users =10000, set the number of EVs to... The EVs parameter settings are shown in Table 1. Rated power Set to fixed parameters, battery capacity and grid connection time Set to a uniform distribution that meets certain parameters. Online time travel demand It follows a normal distribution with fixed parameters, and its charge / discharge efficiency is... , and SOC upper and lower limits , Set to a fixed parameter.

[0143] Table 1: EVs Parameter Settings Table

[0144]

[0145] The number of TCLs is set to The simulation parameters are set as shown in Table 2, assuming the rated power. Equivalent thermal resistance Equivalent heat capacity and thermal power To achieve uniform distribution, set the temperature setpoint. Maximum permissible deviation All of these follow a uniform distribution within a certain range. The temperature-controlled load model is affected by changes in outdoor temperature, and uses the daily outdoor temperature variation curve as a reference, assuming that the outdoor temperature variation satisfies... Figure 7 The variation pattern is shown. The Monte Carlo method was used to randomly sample the values ​​of each simulation parameter to obtain the simulation parameters for each type of adjustable resource group.

[0146] Table 2: EVs Parameter Settings Table

[0147]

[0148] Figure 8 The simulation curves of the output of various adjustable resources within the VPP throughout the day are shown. TCLs are high-quality controllable resources with relatively small capacity within the VPP, and their power curves are relatively flat, indicating relatively stable adjustable potential. The peak charging load of EVs occurs between 16:00 and 20:00. During this period, most EVs are connected to the grid, and EVs have significant adjustable potential. The fluctuations in the total power curve within the VPP are mainly caused by the intermittent power output of the DG during the daytime; therefore, the VPP's regulation capability also exhibits some volatility during the day.

[0149] In the actual process of adjusting adjustable resources, VPP determines whether to participate in the response based on the operating status of each resource. Taking into account response priority and adjustment capacity limitations, this method is used to obtain the maximum and minimum values ​​of the response capabilities of each adjustable resource cluster at each time node of VPP. Figure 9 The upper and lower limits of the total adjustable capability of VPP are given.

[0150] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the flexible resource dynamic aggregation response capability of a virtual power plant, comprising the following steps: Step 1: Construct an adjustable load state model Step 1.1: Construct the EVs resource state model and obtain the EVs resource state model parameters; Step 1.2: Construct a TCLs resource state model to obtain key state model parameters related to TCLs resources; Step 2: Construct a response status priority queue Step 2.1: Construct the EVs resource priority state queue; Step 2.2: Construct the TCLs resource priority state queue; Step 3: Evaluate the dynamic aggregation response capability of the VPP cluster Step 3.1: Construct the aggregated state model; Step 3.2: Construct a state control model; Obtain the running status parameters of EVs at time t, divide them into multiple clusters, calculate the RTM and SOCM indices within the clusters, form an EVs state priority queue based on the indices, and calculate the maximum responsive potential of EVs based on the state control model. Obtain the key state variables of TCLs at time t, divide TCLs into multiple clusters based on the key state variables, calculate the RTC and RTM indices within the clusters, form a TCLs state priority queue based on the indices, and calculate the maximum responsive potential of TCLs based on the state control model. Output the maximum response potential of each resource cluster and calculate the maximum response potential of each resource cluster at the next time step. The state control model for various adjustable resources within the VPP can be expressed by formula (21): （21） in, in It is the state matrix of the internal devices of VPP; It is the power output matrix of the internal devices of the VPP; It is the time interval matrix after the internal resources of the VPP are adjusted; When the power grid is t Always need When adjusting the capacity, adjustment is achieved by regulating the output of real-time power, i.e., by adding an output control matrix before the state transition matrix. Therefore, formula (21) can also be expressed as: （22） It is a diagonal matrix. , of which elements , , , This is an adjustable resource output control variable used to increase or decrease the output of VPP aggregated resources; where... , Power control is achieved by controlling the discharge current; and DG j It is in charging state. This indicates that it is in a discharge state; TCL j The switch is in the "off" position. This indicates that it is in the "open" state; Adjusting power consumption by changing electricity prices Indicates DR j When operating at maximum power consumption, it is no longer possible to reduce VPP output by increasing load, but it has the maximum potential to increase VPP output at this time. Based on the state control model in (22), the maximum potential for VPP output adjustment at any given time can be obtained further through formula (23): （23） in, For the maximum up-adjustment control matrix, all diagonal elements are the maximum values ​​of the control variables; This is the current control matrix; The maximum or rated power of the equipment within the VPP; The vector matrix consists of all elements equal to 1; similarly, the maximum potential for VPP output reduction can be obtained as follows: （24） in, For the maximum down-adjustment control matrix, all diagonal elements are the minimum values ​​of the control variables.

2. The method according to claim 1, characterized in that, The EVs resource status model is as follows: （2） （3） in, For EV i exist t Power output at any given moment; For EV i The normalized SOC value at time t varies in the range of [0, 1]. We can obtain the following from formula (4): （4） in, For EV i Battery capacity; and They represent EVs respectively i The charging and discharging efficiency.

3. The method according to claim 2, characterized in that, The TCLs resource status model is obtained from formula (8): （8） in, for Predicted temperature at any given time; equal , and They represent TCL respectively l The equipment is in constant time t Indoor and outdoor temperatures; For TCL l The electrical power consumed at time t.

4. The method according to claim 3, characterized in that, RTM refers to EVs that are currently in the network access status. i exist During the period from t From the moment on The duration of continuous participation in a given response method is abbreviated as: As shown in equation (11), （11） EVs with larger RTM values ​​are given priority when participating in the response.

5. The method according to claim 4, characterized in that, First, the EV clusters are sorted in descending order based on RTM. When RTMs are equal, the EV with the larger SOCM is assigned a higher response priority. SOCM is directly related to battery SOC. When the output is adjusted downwards or upwards, the EVs... i exist SOCM value within the time period Should be respectively and Based on the standard, as shown in equation (12), （12） In the formula: for The adjustment demand of EVs during the period, a negative value means that the output of EVs needs to be reduced, and a positive value means that the output of EVs needs to be increased; When EVs need to reduce their output, and The larger the difference, the greater the SOCM of the EV, which will prioritize stopping discharging and even starting charging; when EVs need to increase their output... and The larger the difference, the larger the SOCM of the EV, which will prioritize stopping charging or even starting to discharge.

6. The method according to claim 5, characterized in that, First, the TCL clusters are sorted in descending order based on RTM. If RTMs are equal, the TCL with the larger RTC is given a higher adjustment priority. RTC is directly related to the temperature tolerance set by the TCL. When the output is adjusted downwards or upwards, the TCL... l exist RTCM values ​​within the time period Should be respectively and Based on the standard, as shown in equation (14), （14） In the formula: for The adjustment demand of TCLs during the time period. A negative value indicates that the load of TCLs needs to be increased, while a positive value means that the load needs to be decreased.

7. The method according to claim 6, characterized in that, The aggregation state model is given by formulas (15)-(20): （15） （16） （17） （18） （19） （20） in, , These represent the number of devices in the EVs and TCLs clusters, respectively.

Images