A Scheduling Instruction Decomposition Method and Energy Controller Based on Dynamic Participation Factors

Abstract

The present invention discloses a method for decomposing dispatch instructions based on dynamic participation factors and an energy controller, which belong to the field of electrical engineering. The method comprises: receiving an overall dispatch instruction issued by a power grid control center; calculating the dynamic participation factors of each electric vehicle cluster and traditional power generation source at different dispatching moments; decomposing the overall dispatch instruction into multiple first-level dispatch instructions according to the dynamic participation factors of each electric vehicle cluster and traditional power generation source at different dispatching moments, and sending the instructions to a microgrid energy controller, wherein the first-level dispatch instruction corresponds to the microgrid energy controller. The present invention adopts a method for correcting the dynamic participation factor calculation according to real-time electricity price changes, and further decomposes the dispatch instruction of a flexible load aggregate. Since the dynamic participation factor takes into account the frequency response characteristics of different frequency regulation resources, it comprehensively considers the electricity price and the controllable capacity factors of the load aggregate to make a judgment on the calling priority of different load aggregates, thereby achieving accurate frequency regulation and rationally utilizing the frequency regulation capacity of different resources.

Description

Technical Field

[0001] The present invention belongs to the field of electrical engineering technology, and more specifically, relates to a scheduling instruction decomposition method based on dynamic participation factors and an energy controller. Background Art

[0002] The low-carbon transformation of primary energy is evident, terminal energy is accelerating towards high-level electrification, and new energy sources with randomness and uncertainty are being connected to the system in large quantities. This places higher demands on the stability of the entire large-scale power grid. The difficulty of maintaining safe and stable operation by relying solely on traditional grid regulation will be greatly increased. This is mainly manifested in the following aspects:

[0003] (1) The large-scale access of renewable energy will lead to profound changes in the physical characteristics, operation mode, and functional form of the power grid. As the proportion of renewable energy generation such as wind power and photovoltaic power gradually increases, the randomness and intermittent nature of their power generation will have an impact on the entire power system and reduce the quality of electricity. In addition, the anti-peaking characteristics of wind power output will gradually increase the peak-to-valley difference and increase the pressure of wind and solar power abandonment, posing a severe challenge to the dynamic balance of supply and demand in the power grid. At present, the power grid uses deep peaking of thermal power and local demand-side response to carry out dynamic balance regulation of the power grid, which will be difficult to meet the requirements of further growth in the scale and proportion of renewable energy generation in the future. Carrying out user-side load dispatching and control is the development direction of power grid regulation and operation. Giving full play to the flexible load control capabilities of electric vehicles, user-side energy storage, etc. on the user side to participate in power grid dispatching and control and dynamic balance will undoubtedly enhance the power grid's ability to accept new and renewable energy power generation and promote my country's energy production reform.

[0004] (2) Currently, new types of electricity loads are developing rapidly, placing enormous pressure on the safe, stable, and efficient operation of the power grid. The large-scale centralized connection of electric vehicles to the grid will create a "peak upon peak" situation, severely impacting the balance of the power grid system and local power grids, and forcing the grid to increase construction investment with low investment returns. Utilizing the controllable characteristics of flexible loads, establishing a market-based mechanism and using appropriate technical means to guide them, so that they can participate in power grid optimization under market incentives such as price guidance, will effectively promote the safe, economical and efficient operation of the power grid.

[0005] (3) Flexible and adjustable loads on the user side are decentralized and connected, and their control capabilities are not effectively utilized. Electric vehicles, energy storage, electric heating and other loads have relatively small controllable power units and are dispersed in the grid access. In addition, the power consumption characteristics, control methods, and response times of different loads vary greatly, making it difficult for the grid control center to directly utilize their control capacity resources. Therefore, it is necessary to study the aggregation methods of different types of flexible loads. With the help of the load aggregation operation and control capabilities of load aggregators, the decentralized and diverse controllable load resources can be aggregated, their characteristics regularized, and dispatched and managed. Only then can the controllable load potential be converted into controllable resources and the required operation control and dynamic balance services be provided to the grid.

[0006] To sum up, giving full play to the role of flexible load resources on the user side in promoting the supply and demand and regulation balance of the power grid will effectively alleviate the prominent problems currently facing the power grid. It is of great significance to promote the safe, stable and efficient operation of the power grid, improve the power grid's ability to accept new and renewable energy, and promote the transformation of my country's energy system.

[0007] ″Complete Provision of MPC based LFC by Electric Vehicles with Inertial and Droop Support from DFIG Wind Farm″ proposed the concept of dynamic participation factor, based on which the dispatching instructions issued by the grid control center are decomposed and sent to the flexible load cluster and traditional power generation sources. This technical solution has the following deficiencies and shortcomings: (1) When calculating the dynamic participation factor, only the real-time adjustable capacity of flexible loads and traditional power generation sources is considered, and the real-time changes in the grid frequency regulation auxiliary service electricity price are not considered. Therefore, it is impossible to objectively judge the optimal calling priority of different frequency regulation resources; (2) The dispatching instructions issued by the grid control center are only decomposed at the first level and sent to the flexible load cluster (first-level decomposition). The dispatching instructions of the flexible load cluster are not further decomposed, and the individual loads cannot be accurately regulated. Summary of the Invention

[0008] In view of the defects and improvement needs of the existing technology, the present invention provides a scheduling instruction decomposition method and energy controller based on dynamic participation factors, the purpose of which is to

[0009] To achieve the above objectives, according to a first aspect of the present invention, a method for calculating a dynamic participation factor taking into account real-time electricity prices and real-time adjustable capacity factors is provided, the method comprising:

[0010] Case 1: η1(t)≤η2(t)≤η3(t)

[0011] If ∑ΔP L (t)≤P 1，EV1(t), then λ 1i =1,λ 2i =λ 3i =0;

[0012] If P 1，EV1 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t), then λ 1i (t)+λ 2i (t) = 1,

[0013]

[0014] If P 1，EV1 (t)+P 2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), then

[0015]

[0016] Case 2: η1(t)≤η3(t)≤η2(t)

[0017] If ∑ΔP L (t)≤P 1，EV1 (t), then λ 1i =1,λ 2i =λ 3i =0;

[0018] If P 1，EV1 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 3，CS (t), then λ 1i (t)+λ 3i (t) = 1,

[0019]

[0020] If P 1，EV1 (t)+P 3，CS (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), then

[0021]

[0022] Case 3: η2(t)≤η1(t)≤η3(t)

[0023] If ∑ΔP L (t)≤P 2，EV2 (t), then λ 2i =1,λ 1i =λ 3i =0;

[0024] If P 2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t), then λ 1i (t)+λ 2i (t) = 1,

[0025]

[0026] If P 1，EV1 (t)+P 2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), then

[0027]

[0028] Among them, η1(t), η2(t), and η3(t) are the P 1，EV1 (t), P 2，EV2 (t), P 3，CS (t) Compensation price for participating in grid ancillary services, ΔP L (t) is the capacity demand of grid auxiliary services at time t, P 1，EV1 (t), P 2，EV2 (t) are the adjustable capacities of electric vehicle load subgroups 1 and 2 at time t, P 3，CS (t) is the adjustable capacity of the traditional power generation source at time t, λ 1i (t), λ 2i (t), λ 3i (t) are the dynamic participation factors of electric vehicle load subgroups 1 and 2 and traditional power generation sources at time t, respectively. The subscript i identifies the dynamic participation factor.

[0029] Preferably, the following processing is performed on the dynamic factors between different scheduling periods to obtain a smooth transition dynamic factor:

[0030] If the jth scheduling resource is available in the T-1th time window but not in the Tth time window, then

[0031] If the jth scheduling resource is available in both the T-1th and Tth time windows, then

[0032] in, is the value of the dynamic participation factor of the jth scheduling resource at time t in the T-1th time window, is the adjustable capacity of the jth scheduling resource in the T-1th time window, T is the Tth computing time window, and j = 1, 2, 3.

[0033] Beneficial Effects: Because electric vehicle control resources offer lower costs, lower carbon emissions, and are not subject to the ramp rate limitations of traditional generator sets, electric vehicle loads should be prioritized when participating in grid control. Electric vehicle loads have varying controllable capacities at different times, and may even be controllable at one moment but uncontrollable the next. Therefore, this approach ensures a smooth transition.

[0034] To achieve the above-mentioned object, according to a second aspect of the present invention, a method for decomposing flexible load aggregate scheduling instructions based on dynamic participation factors is provided. The method is applied to a regional energy controller and comprises:

[0035] T1. Receive overall dispatch instructions from the power grid control center;

[0036] T2. Using the method as in the first aspect, calculate the dynamic participation factor of each electric vehicle cluster and traditional power generation power supply at different scheduling times;

[0037] T3. Decompose the overall dispatch instruction into multiple first-level dispatch instructions based on the dynamic participation factors of each electric vehicle cluster and traditional power generation source at different dispatch times, and send them to the microgrid energy controller. The first-level dispatch instructions correspond one-to-one to the microgrid energy controller.

[0038] To achieve the above objectives, according to a third aspect of the present invention, a method for decomposing flexible load unit scheduling instructions based on state queue sorting is provided. The method is applied to a microgrid energy controller and includes:

[0039] S1. Receive a level 1 dispatch instruction issued by a regional energy controller, wherein the regional energy controller adopts the flexible load aggregate dispatch instruction decomposition method based on the dynamic participation factor as described in the second aspect;

[0040] S2. Based on the state queue sorting algorithm, the first-level scheduling instruction is further decomposed into multiple second-level scheduling instructions, and sent to the flexible load cells. The second-level scheduling instructions correspond to the flexible load cells one by one.

[0041] Preferably, step S2 includes:

[0042] S21. Indicators for evaluating the adjustability of electric vehicles:

[0043] Reverse power supply capability

[0044] Charging time margin

[0045] S22. Perform dimensionless processing on the attribute values of each indicator;

[0046]

[0047]

[0048] S23. Based on the obtained comprehensive weight vector and the values of various evaluation indicators, a comprehensive evaluation is conducted on the dispatchability of electric vehicles connected to the grid. The dispatchability of vehicle i is

[0049] S24. Preliminary determination of the scheduling priority of each responding entity based on the dispatchable capacity evaluation value of all responding entities entering the network

[0050] S25. Based on the system power compensation requirements within each sampling period, formulate the generalized power allocation criteria for the responding entities:

[0051]

[0052] Among them, P c，i and η c，i are the charging power and charging efficiency of the responding subject i, t d，i and t in，i are the starting time and off-grid time of responding subject i, respectively. s，i is the capacity of the electric vehicle power battery, S d，i and S in，i are the charge states of the responding subject i at the start and end of the scheduling, The charging time margin for electric vehicles to participate in downstream power regulation, The charging time margin for electric vehicles to participate in upstream power regulation, The shortest time to fully charge an electric vehicle; t in , t e are the time when the electric vehicle is connected to the grid and when it leaves the grid; t off is the time when the power shortage occurs; Δt agc The period during which electric vehicles participate in frequency regulation; q ij is the jth evaluation index of flexible load individual i, j = 1, 2, 3, is the average value of the jth evaluation index, ξ j is the absolute value of the difference between the jth evaluation index and the average value, w jis the weighted coefficient of the jth evaluation index, n is the total number of evaluation indicators that affect the individual controllable ability of electric vehicle loads, P com is the compensation power required by the system, P i is the power required to respond allocated to responding entity i.

[0053] Beneficial effect: The present invention dimensionlessly transforms various factors affecting the controllable capability of flexible load cells. By comprehensively considering various factors affecting the controllable capability and performing weighted operations and sorting them, the present invention can prioritize the use of load cells with high controllable capability during grid frequency modulation, thereby achieving accurate execution of load control instructions and improving the control effect.

[0054] To achieve the above-mentioned object, according to a fourth aspect of the present invention, there is provided an energy controller comprising: a computer-readable storage medium and a processor;

[0055] The computer-readable storage medium is used to store executable instructions;

[0056] The processor is used to read the executable instructions stored in the computer-readable storage medium, execute the flexible load aggregate scheduling instruction decomposition method based on dynamic participation factors described in the second aspect, or execute the flexible load single unit scheduling instruction decomposition method based on state queue sorting described in the third aspect.

[0057] To achieve the above-mentioned purpose, according to the fifth aspect of the present invention, a computer-readable storage medium is provided, comprising a stored computer program; when the computer program is executed by a processor, the device where the computer-readable storage medium is located is controlled to execute the dynamic participation factor calculation method considering real-time electricity price factors and real-time adjustable capacity as described in the first aspect, or to execute the flexible load aggregate scheduling instruction decomposition method based on dynamic participation factors as described in the second aspect, or to execute the flexible load single unit scheduling instruction decomposition method based on state queue sorting as described in the third aspect.

[0058] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects:

[0059] (1) The existing dynamic participation factor calculation method sets the demand response electricity price of different frequency regulation resources as a fixed value, without considering the problem that the electricity price changes over time. The present invention adopts a method and result of correcting the dynamic participation factor calculation based on the real-time electricity price change. Since the real-time electricity price factor is taken into account, the correction of the dynamic participation factor and the accurate generation of the first-level dispatch instruction are realized.

[0060] (2) The present invention further decomposes the flexible load aggregate scheduling instructions through the above-mentioned dynamic participation factor calculation method. Since the dynamic participation factor takes into account the frequency response characteristics of different frequency regulation resources, it comprehensively considers the electricity price and the controllable capacity of the load aggregate to make a judgment on the calling priority of different load aggregates, thereby achieving accurate frequency regulation and rationally utilizing the frequency regulation capacity of different resources.

[0061] (3) Based on a state queue sorting algorithm, this invention further decomposes flexible load aggregate scheduling instructions into secondary scheduling instructions corresponding to flexible load units. By accurately assigning scheduling instructions to each flexible load unit, it achieves precise execution of load control instructions and accurate frequency regulation. This method optimizes the decomposition and precise regulation of aggregated loads based on the control center's scheduling plan or instructions for the load aggregates, ensuring high-quality transaction execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a schematic diagram of the flexible load control system.

[0063] Figure 2 It is a single-area system load frequency control model.

[0064] Figure 3 It is a single-area system load frequency control model taking into account the charging and discharging of electric vehicles.

[0065] Figure 4 It is the flow chart of dynamic participation factor calculation.

[0066] Figure 5 This is a schematic diagram of the electric vehicle grid-connected period.

[0067] Figure 6 It is a flow chart for predicting the real-time dispatchable capacity of electric vehicles.

[0068] Figure 7 It is the real-time dispatchable capacity of the concentrated electric vehicle group in one day under the simulation scenario. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0070] The present invention divides the flexible load control system into an application layer, an aggregation layer, and a load layer. The application layer serves the transmission network control center and responds to the power system's auxiliary service needs such as peak shaving, standby, and frequency regulation. The aggregation layer, with the energy controller as the main body, aggregates various types of flexible loads in layers, decomposes the load control instructions layer by layer, and evaluates and predicts the overall load adjustability. The load layer is composed of a large number of flexible loads, smart meters, and control terminals, which collect flexible load operation information and specifically control the flexible load operation status. The microgrid energy controller of the aggregation layer first evaluates the adjustable capacity of electric vehicle loads participating in peak shaving, frequency regulation, and other services based on the load operation information and control parameters obtained by the smart meters of the load layer, and reports the evaluation results to the power grid control center of the application layer. The control center issues an overall dispatching instruction to the regional energy controller based on the reported adjustable capacity of the flexible aggregated load. The regional energy controller then optimizes and decomposes it using the instruction decomposition method, and obtains the power consumption plan of each load unit under the premise of accurately executing the system dispatching instructions.

[0071] Flexible load control system such as Figure 1 As shown, it includes two aspects: technical architecture and market architecture. The technical architecture takes "layered control" as the overall framework and includes three levels: application layer, aggregation layer, and load layer.

[0072] The application layer serves the transmission network control center, addressing the power system's ancillary service needs, such as off-peak load shaving, peak reserve, and real-time frequency regulation. The transmission network control center must submit ancillary service requests to the grid ancillary services market trading platform, purchase the corresponding ancillary services, and conduct security verification of market clearing results. During the trading period, the transmission network control center issues load control instructions to the distribution network control center based on market transaction results.

[0073] The aggregation layer, with energy controllers as the core, aggregates various types of flexible loads in a hierarchical manner. The loads managed by energy controllers may be distributed over a large area, such as a city. To reduce communication and computational burdens, energy controllers are divided into two tiers: the upper tier is the regional energy controller (REC), and the lower tier is the microgrid energy controller (MGC). The loads connected to RECs are generally at the medium- and high-voltage (HV / MV) voltage levels, while the loads connected to MGs are generally at the medium- and low-voltage (MV / LV) voltage levels. RECs receive load control commands forwarded by the distribution network control center, coordinate the overall control plan for flexible loads, perform preliminary decomposition of load dispatch commands among the MGCs, assess the overall controllability of flexible loads, and aggregate and store massive amounts of load-side operational data. The MGCs assess the controllability of subordinate flexible loads and optimize the decomposition of load dispatch commands. Energy controllers utilize a parallel control mode, with energy controllers on the same tier acting as backup for each other. If a single energy controller fails and exits, the same tier's energy controllers can take over and issue control commands to its subordinate flexible loads.

[0074] The load layer consists of numerous flexible loads, smart meters, and control terminals, which are interconnected one-to-one. Smart meters are used to obtain operating information from flexible loads, including control parameters and environmental parameters. They provide real-time assessment of the adjustability of flexible loads and provide edge computing support for the flexible load control system. Control terminals, working in conjunction with smart meters, are used to modify non-adjustable power sources / loads, receive specific load control commands, and control the operating parameters of flexible loads.

[0075] Based on the above-mentioned flexible load control system, the present invention proposes a method for decomposing flexible aggregated load scheduling instructions. The microgrid energy controller first evaluates the adjustable capacity of electric vehicle loads for participating in peak-shaving, frequency regulation and other services based on the load operation information and control parameters obtained by the smart meter, and reports the evaluation results to the power grid control center. The control center issues an overall scheduling instruction to the load based on the reported adjustable capacity of the flexible aggregated load to the regional energy controller. The energy controller then uses the instruction decomposition method to optimize and decompose it, and obtains the power consumption plan of each load unit on the premise of accurately executing the system scheduling instructions.

[0076] Dynamic participation factor

[0077] The grid control center regulates flexible loads when the grid needs ancillary services. Because factors such as the plug-in status, state of charge, and charging power of electric vehicle loads vary during different dispatch periods, the controllable capacity of the electric vehicle load aggregate also changes during these periods. Therefore, a dynamic participation factor (DPF) is introduced to allocate flexible load dispatch instructions to different dispatchable power sources at different times.

[0078] Taking frequency regulation auxiliary services as an example, the basic concept of dynamic participation factors is explained. Maintaining a basically constant power system frequency is an inevitable requirement for reliable and stable operation of the power grid. The power system basically maintains the frequency within the allowable deviation range through primary and secondary frequency adjustments. Load frequency control is an effective control method to adjust the system frequency to stabilize at the rated value and ensure the safe, economical and stable operation of the power system. The traditional single-area system load frequency control model is as follows: Figure 2 shown.

[0079] Figure 2 In the equation, K is the integral control gain; R is the speed regulation of the speed regulator; T g and T ch are the governor time constant and the turbine time constant respectively; ΔP m and ΔP L are the generator mechanical power output deviation and load disturbance respectively; M is the generator inertia constant; D is the load damping coefficient; Δω is the frequency deviation.

[0080] On the basis of the traditional single-area system load frequency control model, the single-area load frequency control model taking into account the charging and discharging of electric vehicles in system frequency regulation is established, such as Figure 3 shown.

[0081] Figure 3 In the equation, K is the integral control gain; R is the speed regulation of the speed regulator; T g and T ch are the governor time constant and the turbine time constant respectively; T EV1 、T EV2 are the charge and discharge time constants of electric vehicle subgroups 1 and 2 respectively; K EV1 , K EV2 are the frequency response coefficients of electric vehicle subgroups 1 and 2 respectively; ΔP m and ΔP L are the generator mechanical power output deviation and load disturbance respectively; M is the generator inertia constant; D is the load damping coefficient; Δω is the frequency deviation.

[0082] Depend on Figure 3As can be seen, electric vehicle aggregates (EVs) and conventional sources (CSs) have different frequency response characteristics. Compared with conventional sources, EV loads are influenced by user travel patterns and electricity usage habits, resulting in significant variations in their controllable capacity within a scheduling period. This variation in EV controllable capacity can be simulated in Figure 1. Therefore, dynamic participation factors can be used to dynamically allocate load scheduling instructions to different dispatchable sources at different times, ensuring accurate frequency regulation while rationally utilizing the frequency regulation capacity of different resources.

[0083] The calculation process of dynamic participation factor is as follows Figure 4 shown.

[0084] ∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t) (1)

[0085] In formula (1), ΔP L (t) is the capacity demand for auxiliary services such as peak regulation and frequency regulation of the power grid at time t, P 1，EV1 (t), P 2，EV2 (t) are the adjustable capacities of electric vehicle load subgroups 1 and 2 at time t, P 3，CS (t) is the controllable capacity of the traditional power generation source at time t, and the control costs of the three resources increase in sequence. Formula (1) indicates that the grid dispatch capacity demand at time t is not greater than the sum of the controllable capacity of the grid generation side and the customer side at time t. Suppose the compensation prices of the three resources participating in the grid ancillary services at time t are η1(t), η2(t), and η3(t). The dynamic participation factors of the three resources at time t are λ 1i (t), λ 2i (t), λ 3i (t) and satisfies λ 1i (t)+λ 2i (t)+λ 3i (t)=1.

[0086] First, the relationship among η1(t), η2(t), and η3(t) is compared, with the basic principle of giving priority to frequency regulation resources with low compensation electricity prices.

[0087] (1.1)η1(t)≤η2(t)≤η3(t)

[0088] Call priority: P 1，EV1 >P 2，EV2 >P 3，CS .

[0089] When the capacity demand ΔPL (t) is less than the first scheduling resource P 1，EV1 (t), that is, ∑ΔP L (t)≤P 1，EV1 (t), let the dynamic participation factor λ 1i = 1. When P 1，EV1 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t), let λ 1i (t)+λ 2i (t) = 1, since the dispatch cost of electric vehicle load subgroup 1 is lower than that of subgroup 2, P is called first during grid regulation. 1，EV1 (t). Then λ 1i (t), λ 2i (t) is calculated by equations (2) and (3) respectively.

[0090]

[0091]

[0092] When P 1，EV1 (t)+P 2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), let λ 1i (t)+λ 2i (t)+λ 3i (t) = 1, λ 1i (t) is still calculated by formula (2), λ 2i (t), λ 3i (t) is calculated by equations (4) and (5) respectively.

[0093]

[0094]

[0095] (1.2)η1(t)≤η3(t)≤η2(t)

[0096] Call priority: P 1，EV1 >P 3，CS >P 2，EV2 .

[0097] When ∑ΔP L (t)≤P 1，EV1 (t), λ 1i =1.

[0098] When P 1，EV1(t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 3，CS (t), let λ 1i (t)+λ 3i (t)=1.

[0099]

[0100]

[0101] When P 1，EV1 (t)+P 3，CS (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), let λ 1i (t)+λ 2i (t)+λ 3i (t) = 1, λ 1i (t) is still calculated by formula (6), λ 2i (t), λ 3i (t) is calculated by equations (8) and (9):

[0102]

[0103]

[0104] (1.3)η2(t)≤η1(t)≤η3(t)

[0105] Call priority: P 2，EV2 >P 1，EV1 >P 3，CS .

[0106] When ∑ΔP L (t)≤P 2，EV2 (t), λ 2i =1.

[0107] When P 2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t), let λ 1i (t)+λ 2i (t)=1.

[0108]

[0109]

[0110] When P 1，EV1 (t)+P2，EV2 (t)≤∑ΔP L (t)≤P 1，EV1 (t)+P 2，EV2 (t)+P 3，CS (t), λ 2i (t) is still calculated by formula (10), let λ 1i (t)+λ 2i (t)+λ 3i (t) = 1, λ 1i (t), λ 3i (t) is calculated by equations (12) and (13) respectively.

[0111]

[0112]

[0113] Based on the above process, the calculation method of dynamic participation factors under other different compensation price scenarios can be deduced accordingly.

[0114] P 1，EV1 (t), P 2，EV2 (t) represents the time-varying controllable capacity of the electric vehicle load. Since electric vehicle control resources have low cost and carbon emissions and are not limited by the ramp rate of traditional generator sets, electric vehicle loads should respond first when participating in grid control. The controllable capacity of electric vehicle loads varies at different times, and it may even be controllable at one moment but uncontrollable at the next. Therefore, equations (14) and (15) are introduced to ensure a smooth transition.

[0115]

[0116]

[0117] Evaluation and prediction of controllable capacity of large-scale electric vehicles

[0118] If charging electric vehicles are allowed to stop to reduce load, which is equivalent to increasing power generation, then these electric vehicles can provide a certain amount of real-time adjustable capacity. If stopped electric vehicles can be dispatched to charge to increase load, which is equivalent to reducing power generation, then these electric vehicles can provide a certain amount of real-time adjustable capacity. Therefore, the controllability and current state of electric vehicles determine the size of real-time dispatchable capacity. The larger the scale of controllable electric vehicles, the greater the dispatchable capacity. The following first introduces the criteria for electric vehicle controllability and then, based on this, introduces a method for evaluating the real-time dispatchable capacity of large-scale electric vehicles.

[0119] Real-time controllability criteria for electric vehicles

[0120] The controllability of electric vehicles is studied from two aspects: grid-connected status and whether it can meet users' travel needs.

[0121] Grid connection criteria

[0122] like Figure 5 As shown in the figure, since the evaluation is performed in units of time interval t, after the electric vehicle is connected to the grid, if it is not connected to the grid at the hour, its dispatchable capacity should be evaluated at the next hour. In addition, during the period when the electric vehicle is about to leave, since it cannot last for the entire time period, the dispatchable capacity of the electric vehicle in this period is ignored. The criterion can be expressed by the following formula.

[0123] INT(t0)+1≤t≤INT(t d )-1(t=1,2,3...24) (16)

[0124] Where t0 is the time when the electric vehicle is connected to the grid, t d is the time when the electric vehicle is off-grid, is the hour, t=1, 2, 3...24, INT is the rounding function.

[0125] If the criterion is not established, it means that the electric vehicle cannot be in the grid-connected state in the entire unit time period, that is, it is uncontrollable. If the criterion is established, it means that the electric vehicle is in the grid-connected state from the first hour INT(t0)+1 to the hour INT(t0) of the off-grid time. d ) are continuously connected to the grid, meeting the grid connection criteria, and their controllability is further judged by the charging requirement criteria.

[0126] User electricity demand criteria:

[0127]

[0128]

[0129] Where, is the user's expected power forecast value, is the predicted value of battery power at each hour, P Ci is the charging power of the electric vehicle, η is the charging efficiency, C i Indicates the battery capacity.

[0130] This criterion indicates that when the electric vehicle is not charged from time t to t+1, and from t+1 until the car leaves time t d If the charging is continued and the user's needs can be met when leaving, the electric vehicle is considered to be controllable during the period from time t to time t+1. Otherwise, it is uncontrollable.

[0131] Electric vehicle dispatchable capacity forecast

[0132] Combined with the prediction of the charging status and controllability of electric vehicles at each moment, the dispatchable capacity of large-scale electric vehicles during this period is evaluated according to the following formula.

[0133] The calculation formula for adjustable capacity is:

[0134] The calculation formula for adjusting the capacity downward is:

[0135] Where, are the “0 / 1” parameters that characterize the load removability and load increaseability of electric vehicles. At time t, when the electric vehicle is controllable and in the charging state, let Indicates that the electric vehicle load can be cut off, otherwise Indicates that the electric vehicle cannot be cut off; when the electric vehicle is controllable and the state is to stop charging, then Indicates that the electric vehicle can be further charged to increase the load, otherwise This means that electric vehicles cannot increase the load further.

[0136] Real-time adjustable capacity evaluation method for electric vehicle loads

[0137] The evaluation process of real-time dispatchable capacity of electric vehicles is as follows: Figure 6 As shown, the specific evaluation steps are as follows:

[0138] Step 1: Define the matrix that represents the controllability of the electric vehicle and initialize it; using ΔT as the time node, define the matrix Indicates burden resectability. It means that the i-th electric vehicle can choose to stop charging at time t according to the demand of the power grid to reduce the load, that is, to provide adjustable capacity; Indicates that the i-th electric vehicle cannot provide adjustable capacity at time t. Define the matrix Characterizes load scalability. It means that the i-th electric vehicle can choose to charge according to the demand of the power grid at time t to increase the load, that is, to provide adjustable capacity; Indicates that the i-th electric vehicle cannot provide adjustable capacity at time t; according to different applications, the duration Δt of the electric vehicle's dispatchable capacity is set.

[0139] Step 2: Obtain the current information of each electric vehicle based on the real-time status; the current information of the electric vehicle includes the estimated departure time of the electric vehicle, the expected power level at departure, the current power level, the charging power, the battery capacity of the electric vehicle, etc. The acquisition of the above information needs to be supported by smart meters, communication equipment, etc.

[0140] Step 3: Determine the controllability of the electric vehicle according to formulas (16)-(18).

[0141] Step 4: Assign values to the matrix representing the controllability of electric vehicles according to the actual situation; determine whether the i-th electric vehicle is charging at time t. If it is charging and controllable, the electric vehicle can stop charging according to the needs of the power grid, which can provide adjustable capacity. If the state is to stop charging and it is controllable, the electric vehicle can be charged according to the needs of the power grid, which can provide adjustable capacity.

[0142] Step 5: Calculate and output the dispatchable capacity of electric vehicles. Calculate the dispatchable capacity of large-scale electric vehicles at the current moment using the adjustable capacity calculation formula and proceed to the calculation for the next period.

[0143] Decomposition and correction of flexible aggregate load scheduling instructions

[0144] Due to the diversity and uncertainty of flexible loads on the customer side and the uneven communication levels, the load aggregation operation system needs to optimize the decomposition and precise regulation of the aggregated load after receiving the dispatch plan or instructions based on the load aggregate from the dispatch center to ensure high-quality execution of the transaction. This link is a concentrated reflection of the load aggregation management and regulation technical capabilities of the load aggregator, and is the key to the smooth implementation of flexible load aggregation in the grid dispatch balance optimization.

[0145] When participating in system energy regulation, different responding entities exhibit varying dynamic energy balancing capabilities due to varying historical or current grid access information. This balancing capability is defined as the dispatchability of the responding entity. Taking electric vehicle loads as an example, the following indicators are used to evaluate the dispatchability of the responding entity.

[0146] Reverse power supply capability

[0147] Based on V2G technology, and by utilizing the reverse power supply capability of EVs to increase the system's backup capacity, the reliability of grid operation can be improved. The reverse power supply capability of EVs can be characterized based on information such as the amount of electricity when the EV is connected to the grid, the discharge power, and the duration of the connection to the grid, as follows:

[0148]

[0149] Where p c，i and η c，i denote the charging power and charging efficiency of the responding subject i, respectively, t d，i , t in，i They represent the starting time and off-grid time of the responding subject i, respectively. s，i is the capacity of the electric vehicle power battery, Sd，i 、S in，i They represent the charge state of responding subject i at the start and end of scheduling.

[0150] Internally regulate power distribution based on the order of charging time margins of all grid-connected electric vehicles

[0151] The participation criteria and charging time margin for electric vehicles in upstream and downstream power regulation are calculated respectively. When the electric vehicle's grid connection, departure time, state of charge, and system regulation duration satisfy equations (13) and (14), the electric vehicle can participate in downward or upward regulation by stopping or starting charging. The charging time margin for downward and upward regulation is shown in equations (16) and (17), respectively:

[0152]

[0153] (t off +Δt agc )≤t e (twenty one)

[0154]

[0155]

[0156]

[0157]

[0158] Where, Indicates the shortest time it takes for an electric vehicle to be fully charged; t in , t e Respectively represent the time when the electric vehicle is connected to the grid and leaves; t off Indicates the time when the power shortage occurs; Δt agc Indicates the period when electric vehicles participate in frequency regulation; when power is adjusted upwards, s down The value is 0, s up When the value is 1, the power is adjusted downwards. down The value is 1, s up The value is 0.

[0159] Since the attribute values of each indicator have different dimensions and value ranges, it is necessary to perform dimensionless processing on the attribute values of each indicator.

[0160]

[0161]

[0162] Based on the obtained comprehensive weight vector and the values of various evaluation indicators, the dispatchability of networked vehicles is comprehensively evaluated. The dispatchability of vehicle i can be expressed as:

[0163]

[0164] The scheduling priority of each responding entity is preliminarily determined based on the dispatchability evaluation value of all responding entities entering the network.

[0165]

[0166] Furthermore, combined with the system power compensation requirements within each sampling period, a generalized power allocation criterion for the responding entity is formulated:

[0167]

[0168] Where p com Indicates the compensation power required by the system, P i It represents the power allocated to responding agent i.

[0169] Set up a case study for analysis: Based on the characteristics of frequency regulation capacity, set the primary, secondary, and tertiary frequency regulation capacity parameters as follows:

[0170] Table 1 Frequency regulation capacity

[0171]

[0172] Considering only private cars providing dispatchable capacity to the power grid, in order to evaluate the real-time dispatchable capacity of a certain electric vehicle cluster, real-time data of electric vehicles is required. To this end, the present invention simulates the parameters of electric vehicles, wherein the grid-connected scenarios and parameters are shown in Table 2. This model considers the scenario of private cars charging in residential areas and units, and simulates the charging status and power of electric vehicles at various times according to the intelligent charging strategy with the goal of smoothing the total load fluctuation, and then simulates the dispatchable capacity of the simulated scenario according to the above-mentioned real-time dispatchable capacity evaluation method. The results are shown in Figure 2. Figure 7 shown.

[0173] Table 2 Parameter settings for charging different types of electric vehicles

[0174]

[0175] Figure 7The figure shows the real-time dispatchable capacity curves for primary, secondary, and tertiary frequency regulation for this electric vehicle cluster over a single day. As can be seen, the dispatchable capacity of electric vehicles varies over time due to the varying controllability of electric vehicles at each moment. In the early stages of grid integration, as the number of connected electric vehicles continues to increase, their dispatchable capacity also increases. However, towards the end of grid integration, user power demands become increasingly urgent, resulting in a decrease in the controllability of electric vehicles and a consequent decrease in dispatchable capacity. Dispatchable capacity varies for different uses due to their varying durations. The longer the duration, the smaller the dispatchable capacity. This is because the longer the duration, the worse the controllability of electric vehicles. Therefore, dispatchable capacity decreases with increasing duration.

[0176] The calculated adjustable capacity is reported to the power grid control center. The control center refers to the adjustable capacity of each load aggregate and issues the overall load control instruction to the regional energy controller. The regional energy controller calculates the dynamic participation factor of each electric vehicle cluster and traditional power source at different scheduling times based on the adjustable capacity of each electric vehicle cluster at each time and its historical control cost, so as to allocate flexible load scheduling instructions to different dispatchable power sources (microgrid energy controller) at different times.

[0177] After receiving the load control instructions initially assigned by the regional energy controller, the microgrid energy controller comprehensively considers influencing factors such as flexible aggregation load historical data, equipment operating status, aggregation mode, etc., and aims to accurately execute the dynamic balance scheduling plan of the power grid. It uses the scheduling instruction allocation method based on state queue sorting to disassemble the scheduling target and perform load distribution, and generate terminal load scheduling control plans and instructions.

[0178] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for calculating a dynamic participation factor taking into account real-time electricity price factors and real-time adjustable capacity, characterized in that: The method includes: Case 1: η1(t)≤η2(t)≤η3(t) If ∑ΔP L (t) ≤ P 1，EV1 (t), then λ 1i = 1, λ 2i = λ 3i = 0; If P 1，EV1 (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 2，EV2 (t), then λ 1i (t) + λ 2i (t) = 1, If P 1，EV1 (t) + P 2，EV2 (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 2，EV2 (t) + P 3，CS (t), then λ 1i (t) + λ 2i (t) + λ 3i (t) = 1, Case 2: η1(t)≤η3(t)≤η2(t) If ∑ΔP L (t) ≤ P 1，EV1 (t), then λ 1i = 1, λ 2i = λ 3i = 0; If P 1，EV1 (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 3，CS (t), then λ 1i (t) + λ 3i (t) = 1, If P 1，EV1 (t) + P 3，CS (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 2，EV2 (t) + P 3，CS (t), then λ 1i (t) + λ 2i (t) + λ 3i (t) = 1, Case 3: η2(t)≤η1(t)≤η3(t) If ∑ΔP L (t) ≤ P 2，EV2 (t), then λ 2i = 1, λ 1i = λ 3i = 0; If P 2，EV2 (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 2，EV2 (t), then λ 1i (t) + λ 2i (t) = 1, If P 1，EV1 (t) + P 2，EV2 (t) ≤ ∑ΔP L (t) ≤ P 1，EV1 (t) + P 2，EV2 (t) + P 3，CS (t), then λ 1i (t) + λ 2i (t) + λ 3i (t) = 1, Among them, η1(t), η2(t), and η3(t) are the P 1，EV1 (t), P 2，EV2 (t), P 3，CS (t) Compensation price for participating in grid ancillary services, ΔP L (t) is the capacity demand of grid auxiliary services at time t, P 1，EV1 (t), P 2，EV2 (t) are the adjustable capacities of electric vehicle load subgroups 1 and 2 at time t, P 3，CS (t) is the adjustable capacity of the traditional power generation source at time t, λ 1i (t), λ 2i (t), λ 3i (t) are the dynamic participation factors of electric vehicle load subgroups 1 and 2 and traditional power generation sources at time t, respectively. The subscript i identifies the dynamic participation factor.

2. The method according to claim 1, wherein The following processing is performed on the dynamic factors between different scheduling periods to obtain the smooth transition dynamic factors: If the jth scheduling resource is available in the T-1th time window but not in the Tth time window, then If the jth scheduling resource is available in both the T-1th and Tth time windows, then in, is the value of the dynamic participation factor of the jth scheduling resource at time t in the T-1th time window, is the adjustable capacity of the jth scheduling resource in the T-1th time window, T is the Tth computing time window, and j = 1, 2, 3.

3. A method for decomposing flexible load aggregate scheduling instructions based on dynamic participation factors, characterized in that: The method is used for a regional energy controller, and the method includes: T1. Receive overall dispatch instructions from the power grid control center; T2. Using the method according to claim 1 or 2, calculate the dynamic participation factors of each electric vehicle cluster and traditional power generation power source at different scheduling times; T3. Decompose the overall dispatch instruction into multiple first-level dispatch instructions based on the dynamic participation factors of each electric vehicle cluster and traditional power generation source at different dispatch times, and send them to the microgrid energy controller. The first-level dispatch instructions correspond one-to-one to the microgrid energy controller.

4. A flexible load cell scheduling instruction decomposition method based on state queue sorting, characterized in that: The method is applied to a microgrid energy controller, and the method includes: S1. Receive a level 1 dispatch instruction issued by a regional energy controller, wherein the regional energy controller adopts the method for decomposing a flexible load aggregate dispatch instruction based on a dynamic participation factor according to claim 3; S2. Based on the state queue sorting algorithm, the first-level scheduling instruction is further decomposed into multiple second-level scheduling instructions, and sent to the flexible load cells. The second-level scheduling instructions correspond to the flexible load cells one by one.

5. The flexible load unit scheduling instruction decomposition method according to claim 4, characterized in that: Step S2 includes: S21. Indicators for evaluating the adjustability of electric vehicles: Reverse power supply capability Charging time margin S22. Perform dimensionless processing on the attribute values of each indicator; S23. Based on the obtained comprehensive weight vector and the values of various evaluation indicators, a comprehensive evaluation is conducted on the dispatchability of electric vehicles connected to the grid. The dispatchability of vehicle i is S24. Preliminary determination of the scheduling priority of each responding entity based on the dispatchable capacity evaluation value of all responding entities entering the network S25. Based on the system power compensation requirements within each sampling period, formulate the generalized power allocation criteria for the responding entities: Among them, P c，i and η c，i are the charging power and charging efficiency of the responding subject i, t d，i and t in，i are the starting time and off-grid time of responding subject i, respectively. s，i is the capacity of the electric vehicle power battery, S d，i and S in，i are the charge states of the responding subject i at the start and end of the scheduling, The charging time margin for electric vehicles to participate in downstream power regulation, The charging time margin for electric vehicles to participate in upstream power regulation, The shortest time to fully charge an electric vehicle; t in , t e are the time when the electric vehicle is connected to the grid and when it leaves the grid; t off is the time when the power shortage occurs; Δt agc The period during which electric vehicles participate in frequency regulation; q ij is the jth evaluation index of flexible load individual i, j = 1, 2, 3, is the average value of the jth evaluation index, ξi is the absolute value of the difference between the jth evaluation index and the average value, w j is the weighted coefficient of the jth evaluation index, n is the total number of evaluation indicators that affect the individual controllable ability of electric vehicle loads, P com is the compensation power required by the system, P i is the power required to respond allocated to responding entity i.

6. An energy controller, characterized in that: include: Computer-readable storage medium and processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium, execute the flexible load aggregate scheduling instruction decomposition method based on dynamic participation factor as described in claim 3, or execute the flexible load monomer scheduling instruction decomposition method based on state queue sorting as described in claim 4 or 5.

7. A computer-readable storage medium, characterized in that The invention comprises a stored computer program; when the computer program is executed by a processor, the device where the computer-readable storage medium is located is controlled to execute the method for calculating the dynamic participation factor considering the real-time electricity price factor and the real-time adjustable capacity as described in claim 1 or 2, or to execute the method for decomposing the flexible load aggregate scheduling instruction based on the dynamic participation factor as described in claim 3, or to execute the method for decomposing the flexible load single unit scheduling instruction based on the state queue sorting as described in claim 4 or 5.

Images