Secondary frequency modulation method for regional power distribution network based on virtual power plant

Abstract

The application provides a regional power distribution network secondary frequency modulation method based on a virtual power plant, and belongs to the field of power distribution network frequency modulation; a specific scheme is as follows: a virtual power plant model of a framework and a frequency modulation market mechanism are obtained; an optimization scheme of the virtual power plant model is obtained under the frequency modulation market mechanism; and the virtual power plant is optimized and dispatched according to the optimization scheme of the virtual power plant model. The application is based on the virtual power plant model of the framework, jointly dispatches, improves resource utilization and economic benefits, reduces the prediction error of the virtual power plant, and guarantees the travel demand and temperature comfort demand of users.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power distribution network frequency modulation, in particular to a regional power distribution network secondary frequency modulation method based on a virtual power plant. BACKGROUND

[0002] The statements in this section merely provide background technology related to the present application and do not necessarily constitute prior art.

[0003] With the development of measurement and communication technology, the number of intelligent buildings in cities is increasing. Electric vehicles and air conditioners have good power real-time response potential and become important demand-side resources in intelligent buildings. Under the background of global environmental change and energy shortage, electric vehicles are rapidly developing with the support of national policies due to their low pollution characteristics; electric vehicles are not only loads of regional power distribution networks, but also can deliver power to regional power distribution networks through V2G technology; at the same time, a large number of air conditioning loads in intelligent buildings are essential and have good power response capability, and have a heat storage effect, which can change the operating power while reducing the impact on human comfort; therefore, intelligent buildings have great peak load reduction potential, and strengthening the operation and regulation will have considerable economic and social benefits.

[0004] At the same time, the demand-side resources in intelligent buildings have the characteristics of disorderly operation, dispersion and small single adjustable capacity. At present, demand-side resources mainly use user self-control, such as random and disorderly charging and discharging of electric vehicles, independent operation of air conditioning loads under different user temperature settings, etc. There are problems such as poor perception ability and slow response speed, which cannot meet the real-time power balance scheduling needs of regional power distribution networks. Various demand-side resources lack coordinated regulation among each other and cannot participate in regional power distribution networks to realize their resource value. How to scale the aggregation of demand-side resources to participate in the regulation of regional power distribution networks has become an important research direction.

[0005] There are studies on the participation of distributed energy and demand response resources in secondary frequency modulation in the prior art. Considering the uncertainty of electric vehicles and temperature-controlled loads and the demand for frequency modulation backup, a backup optimization and real-time scheduling model for participating in frequency modulation auxiliary services is established to optimize the operation of demand-side resources at the day-ahead and intraday scales. Considering electric vehicles and air conditioning resources in intelligent buildings, considering the charging demand of electric vehicles and the power demand of air conditioners, a mixed integer optimization model is established in the day-ahead market to obtain the optimal bidding strategy. The operation strategy of wind power participating in the frequency modulation market is studied, a penalty mechanism for day-ahead scheduling is proposed to reduce the risk, and an estimation method for frequency modulation performance is proposed to maximize the benefit, and the optimal bidding strategy of wind power participating in the frequency modulation market is obtained.

[0006] But in the existing research, the demand side resources such as electric vehicles and air conditioning loads are not jointly dispatched with distributed energy such as wind power and photovoltaic, and the day-ahead / real-time two-stage scheduling problem is not combined, so that the standby space at the day-ahead reporting time is difficult to evaluate, and the prediction error of wind power and photovoltaic and the travel demand and temperature comfort demand of users are not fully considered in the scheduling process. SUMMARY

[0007] In order to solve the problems in the prior art, the application provides a regional power distribution network secondary frequency modulation method based on a virtual power plant, a virtual power plant model based on architecture is used for joint scheduling, resource utilization and economic benefits are improved, prediction errors of the virtual power plant are reduced, and travel demand and temperature comfort demand of users are ensured.

[0008] In order to achieve the above-mentioned purpose, the application adopts the following technical solutions:

[0009] The first aspect of the application provides a regional power distribution network secondary frequency modulation method based on a virtual power plant, including the following steps:

[0010] An architecture-based virtual power plant model and a frequency modulation market mechanism are obtained;

[0011] An optimization scheme of the virtual power plant model is obtained under the frequency modulation market mechanism;

[0012] The virtual power plant is optimized and dispatched according to the optimization scheme of the virtual power plant model.

[0013] Optionally, the component members of the virtual power plant model include:

[0014] Electric vehicles, controllable loads, wind turbine generators, photovoltaic and energy storage; the frequency modulation market mechanism divides the market into a day-ahead market and an intra-day market.

[0015] Optionally, in the day-ahead market, a long-time scale constraint condition is established with the maximum virtual power plant income as an objective function; in the intra-day market, a short-time scale constraint condition is established with the minimum virtual power plant penalty as an objective function.

[0016] Optionally, the optimization scheme of the virtual power plant model obtained under the frequency modulation market mechanism includes:

[0017] In the day-ahead market, an optimization scheme of the virtual power plant model is obtained by using an opportunity constraint programming.

[0018] Optionally, the opportunity constraint programming is a kind of random programming, and the opportunity constraint programming theory allows the objective function to not meet the constraint in a special case, and the mathematical form of the opportunity constraint programming is:

[0019]

[0020] In the formula: f(x) is a target function; g j (x, ξ) is a random constraint function; α j is a probability; j represents the number of constraint conditions; s.t. represents limited; Pr represents the probability of the condition occurring;

[0021] The main feature of the chance constraint programming is to have certain requirements for the probability of the constraint condition, that is:

[0022] Pr{g j (x, ξ)≤0}≥α j

[0023] It is called a chance constraint condition;

[0024] When solving the chance constraint programming, the constraint with a random variable can be converted into a deterministic constraint; in particular, when the variable is an independent normal distribution variable, the chance constraint programming can be converted into:

[0025]

[0026] In the formula: μ is the equivalent expectation; σ is the equivalent variance; Φ is the standard normal distribution function.

[0027] Optionally, the optimization scheme of the virtual power plant model obtained in the frequency modulation market mechanism includes:

[0028] The model predictive control technology MPC (model predictive control) is used in the intraday market to obtain the optimization scheme of the virtual power plant model.

[0029] Optionally, the model predictive control technology refers to a closed-loop optimization control algorithm based on a model, and the idea is to combine the system model, the current state quantity and the constraint condition to solve the optimal control input variable on line.

[0030] Optionally, the optimization scheme of the virtual power plant model is used to optimize and schedule the virtual power plant, including:

[0031] The virtual power plant is optimized and scheduled in the day-ahead market according to the target function of the virtual power plant model.

[0032] Optionally, in the day-ahead market, the target function of the virtual power plant optimization scheduling is to maximize the virtual power plant revenue, and the expression is as follows:

[0033]

[0034] In the formula: T is a scheduling period, which is 24h in the day-ahead; the total revenue is divided into two parts, which are the revenue of the virtual power plant participating in the energy market at t period and the revenue of participating in the frequency modulation market Since the maintenance cost of electric vehicles, energy storage, wind power and photovoltaic is small in the scale of a day, it is not taken into account;

[0035] The specific expression of the VPP (Virtual Power Plant) day-ahead income per part is:

[0036] The VPP income from the electricity market is:

[0037]

[0038] In the formula: and P and P are the selling price and the buying price of the regional power grid in the day-ahead t period, respectively; t sell and P t buy P and P are the selling power and the buying power of the VPP and the regional power grid in the t period, respectively, and P is the sum of the powers of the members; Δt is a period, which is 1h in the day-ahead;

[0039] The VPP income from the frequency modulation market is:

[0040]

[0041] In the formula: and P and P are the frequency modulation capacity price and the mileage price of the VPP in the day-ahead t period predicted according to historical data, respectively; t c and P t M P and P are the frequency modulation capacity and the frequency modulation mileage of the VPP and the regional power grid in the t period, respectively, the frequency modulation capacity is the sum of the frequency modulation capacities of the members, and the frequency modulation mileage is the sum of the frequency modulation signal mileages issued by the regional power grid, in the example of the present chapter, the frequency modulation mileage used in the day-ahead is simulated by using historical mileage data; and The frequency modulation performance index.

[0042] Optionally, the optimization scheme according to the virtual power plant model performs day-ahead optimization scheduling on the virtual power plant by a target function corresponding to the day-ahead market, including:

[0043] obtaining the constraint conditions of each component member in the virtual power plant model in the day-ahead market.

[0044] Optionally, the optimization scheme according to the virtual power plant model performs day-ahead optimization scheduling on the virtual power plant by a target function corresponding to the day-ahead market, further including:

[0045] obtaining the confidence level settings of each component member in the virtual power plant model.

[0046] Optionally, the optimization scheme according to the virtual power plant model optimizes scheduling of the virtual power plant, including:

[0047] The optimization scheme according to the virtual power plant model optimizes scheduling of the virtual power plant by a target function corresponding to an intraday market.

[0048] Optionally, the target function corresponding to the intraday market is:

[0049]

[0050] In the formula, ψ R,t is a deviation of tracking of the frequency modulation signal; ψ E,t is a deviation of the intraday energy market; α cp and α E are penalty coefficients of the frequency modulation market and the energy market, respectively.

[0051] Optionally, the optimization scheme according to the virtual power plant model optimizes scheduling of the virtual power plant by a target function corresponding to an intraday market, including:

[0052] Obtaining constraint conditions of each component member in the virtual power plant model in the intraday market.

[0053] Optionally, the optimization scheme according to the virtual power plant model optimizes scheduling of the virtual power plant by a target function corresponding to an intraday market, further including:

[0054] Obtaining a power distribution strategy of each component member of the virtual power plant in the intraday market.

[0055] Optionally, the power distribution strategy of each component member of the virtual power plant is as follows:

[0056] In the intraday market, the virtual power plant receives a frequency modulation signal issued by a dispatching center and distributes it to each member. In order to better guarantee the travel demand and temperature comfort demand of users, the virtual power plant first issues the frequency modulation signal to wind power, photovoltaic and energy storage, and the insufficient part is borne by electric vehicles and air conditioning loads.

[0057] First, the frequency modulation capacity is allocated according to the frequency modulation capacity ratio of the electric vehicle and the air conditioning load:

[0058]

[0059] In the formula, P up and P dn are the up frequency modulation signal and the down frequency modulation signal allocated to the electric vehicle and the air conditioning load, respectively; P EVup_tar and P EVdn_tar are the up frequency modulation signal and the down frequency modulation signal allocated to the electric vehicle, respectively; P ACup_tar and PACdn_tar The up-regulation frequency signal and the down-regulation frequency signal allocated to the air conditioning load;

[0060] The state of charge of the electric vehicle is sorted, and the electric vehicle with a state of charge higher than the state of charge obtained by the day-ahead optimization scheduling is preferentially selected to respond to the up-regulation frequency signal until the total up-regulation frequency power meets the requirement; similarly, the electric vehicle with a state of charge lower than the state of charge obtained by the day-ahead optimization scheduling is preferentially selected to respond to the down-regulation frequency until the total down-regulation frequency power meets the requirement;

[0061] The frequency regulation capacity required by the air conditioning load is allocated to the air conditioner in the temperature comfort zone; in order to better stabilize the indoor temperature in the temperature comfort zone, the air conditioner with a higher indoor temperature is allocated more down-regulation frequency power to participate in more down-regulation frequency; similarly, the air conditioner with a lower indoor temperature is allocated more up-regulation frequency power to participate in more up-regulation frequency;

[0062] Firstly, the normalized index of room temperature is defined:

[0063]

[0064] In the formula, SOA is the normalized index of room temperature; the closer the room temperature is to the set temperature, the closer SOA is to zero; when the room temperature is in the temperature comfort zone, the range of SOA is [-1, 1];

[0065]

[0066] In the formula: respectively, the up-regulation frequency power and the down-regulation frequency power allocated to the kth air conditioner; ACnum is the total number of air conditioners in the temperature comfort zone; SOA k is the normalized index of the kth air conditioner.

[0067] Compared with the prior art, the application has the beneficial effects that:

[0068] 1. The virtual power plant model established contains two common flexible loads, i.e., electric vehicles and air conditioning loads, and two widely distributed distributed energy sources, i.e., wind power and photovoltaic power, and is jointly scheduled to improve resource utilization and economic benefits.

[0069] 2. The intra-day rolling optimization strategy based on the model predictive control technology effectively reduces the influence of the deviation between the day-ahead scheduling and the actual intra-day situation.

[0070] 3. The influence of the wind power and photovoltaic prediction error is reduced, and the travel demand and temperature comfort demand of users are ensured. DETAILED DESCRIPTION

[0071] Figure 1 The flowchart of the regional power distribution network secondary frequency regulation method based on the virtual power plant provided by Embodiment 1 of the present disclosure;

[0072] Figure 2 A VPP architecture diagram provided for Embodiment 1 of the present disclosure;

[0073] Figure 3 A MPC basic principle diagram provided for Embodiment 1 of the present disclosure;

[0074] Figure 4 A VPP day-ahead optimization scheduling flowchart provided for Embodiment 1 of the present disclosure;

[0075] Figure 5 An opportunity constraint programming confidence level diagram provided for Embodiment 1 of the present disclosure;

[0076] Figure 6 A VPP day-ahead optimization scheduling flowchart provided for Embodiment 1 of the present disclosure. DETAILED DESCRIPTION

[0077] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0078] It should be noted that the terms used herein are only intended to describe specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, they refer to the presence of a feature, step, operation, device, component, and / or combinations thereof.

[0079] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0080] The application scenario of the virtual power plant, the virtual power plant has the complementarity of diversified power source integration and rich regulation and control means, can aggregate the demand side resources into a whole to participate in the operation and dispatch of the regional distribution network, and can play the advantages of the demand side resources to participate in the frequency modulation auxiliary service market. The virtual power plant can centrally control the demand side resources, provide aggregated parameters to the regional distribution network, facilitate the dispatch of the regional distribution network, and distribute the dispatch instructions of the regional distribution network to each demand side resource according to user demand, so as to realize the effective utilization of the demand side resources.

[0081] At the same time, with the gradual reduction of the proportion of thermal power generation in the power system and the continuous increase of the proportion of distributed new energy, the randomness and intermittency of the distributed new energy increase the frequency modulation burden of the power system. Therefore, it is feasible and meaningful to use the virtual power plant to aggregate and control the demand side resources and the distributed energy to participate in the frequency modulation service of the regional distribution network.

[0082] The secondary frequency modulation of the regional power distribution network refers to that the generator set provides sufficient adjustable capacity and certain adjustment rate, and tracks the output of the system to meet the requirement of frequency stability of the system in the allowable adjustment deviation. The process of the secondary frequency modulation of the regional power distribution network based on the virtual power plant is as follows: the regional power distribution network sends the calculated output of the virtual power plant to the virtual power plant through an AGC (Automatic Generation Control) instruction, and then the virtual power plant distributes the output to each component member according to the running state of the virtual power plant.

[0083] Embodiment 1:

[0084] As shown in Figures 1-6 , the embodiment of the present disclosure provides a secondary frequency modulation method of a regional power distribution network based on a virtual power plant, which comprises the following steps: obtaining a virtual power plant model of an architecture and a frequency modulation market mechanism; obtaining an optimization scheme of the virtual power plant model under the frequency modulation market mechanism; and optimizing scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model.

[0085] In one embodiment, the component members of the virtual power plant model comprise an electric vehicle, a controllable load, a wind turbine generator, a photovoltaic device, and energy storage; and the frequency modulation market mechanism divides the market into a day-ahead market and an intra-day market.

[0086] In the embodiment of the present disclosure, the constructed virtual power plant adopts a centralized control architecture, and the component members comprise an electric vehicle, a controllable load, a wind turbine generator, a photovoltaic device, and energy storage, as shown in Figure 2 , the virtual power plant sets up a centralized controller, collects and statistics relevant information between the component members and the regional power distribution network, and uniformly schedules, the information comprising an electric vehicle charging plan and travel time, a predicted load and adjustable amount of a controllable load, a predicted amount of wind power and photovoltaic power, an energy storage amount, a power price, and a frequency modulation price, etc. The virtual power plant uses the wind turbine generator and the photovoltaic device to supply power to the internal load, the energy storage device can reduce the system fluctuation caused by the wind power and the photovoltaic power, and the gap and the excess of the power supply are traded with the regional power distribution network for power and frequency modulation auxiliary service.

[0087] The basic settings of the frequency modulation energy joint market in the related literature are adopted to divide the market into a day-ahead market and an intra-day market, and the frequency modulation auxiliary service market is cleared jointly with the energy market. The market members simultaneously report the information of energy and frequency modulation auxiliary service in the day-ahead market, including mileage bidding, capacity bidding, and provided frequency modulation capacity. In the intra-day market, the frequency modulation power and energy are distributed by maximizing the revenue. Meanwhile, the regional power distribution network punishes the output deviation in the energy market and the frequency modulation market. The virtual power plant is taken as a price receiver, and the bidding strategy is not considered.

[0088] In one embodiment, the above-mentioned target function of maximizing virtual power plant revenue in the day-ahead market establishes long-time scale constraint conditions; and the target function of minimizing virtual power plant penalty in the intra-day market establishes short-time scale constraint conditions.

[0089] In the embodiments of the present disclosure, in the day-ahead market, a target function of maximizing virtual power plant revenue is established to establish long-time scale constraint conditions, including power demand of electric vehicles, temperature comfort demand of air conditioning load, scheduling continuity demand of energy storage, and power constraint of wind and light, to establish an optimal scheduling model. To reduce the impact of prediction error of wind and light and balance economy and reliability, the power of wind and light is constrained by opportunity constraint programming. Meanwhile, considering the feature that wind and light prediction is more accurate as the prediction time is closer, a confidence level curve of opportunity constraint programming is designed;

[0090] In the intra-day market, a target function of minimizing virtual power plant penalty is established to establish short-time scale constraint conditions; model predictive control technology is used to realize rolling optimization operation in the intra-day and ensure consistency of state of charge of electric vehicles and energy storage with day-ahead scheduling; and frequency modulation signal distribution strategy is designed based on state of charge of electric vehicles and temperature of air conditioning load, to ensure travel demand and temperature comfort demand of users to a certain extent.

[0091] In the operation process of the virtual power plant, wind power, photovoltaic power, and load have great uncertainty, and prediction error in the day-ahead market can be as high as 25%. The optimal operation of the virtual power plant has great uncertainty and affects frequency quality of the regional power distribution network. The model predictive control technology is used in the intra-day market and the opportunity constraint programming is used in the day-ahead market to reduce the impact of uncertainty.

[0092] In one embodiment, the above-mentioned optimization scheme of the virtual power plant model obtained in the frequency modulation market mechanism includes: obtaining the optimization scheme of the virtual power plant model by using opportunity constraint programming in the day-ahead market.

[0093] Optionally, the above-mentioned opportunity constraint programming is a kind of random programming, mainly used to solve optimization problems when random variables are included in constraint conditions. The opportunity constraint programming theory allows the target function to not satisfy the constraint in special cases. The mathematical form of the opportunity constraint programming is:

[0094]

[0095] In the formula, f(x) is a target function; g j (x,ξ) is a random constraint function; α j is a probability; j represents the number of constraint conditions; s.t. represents limited by; Pr represents the probability of occurrence of a condition;

[0096] The main feature of the chance constrained programming is that it has a certain requirement for the probability of the constraint condition, that is:

[0097] Pr{g j (x,ξ)≤0}≥α j

[0098] It is called the chance constraint condition; when solving the chance constrained programming, the constraint with the random variable can be converted into a deterministic constraint; in particular, when the variable is an independent normal distribution variable, the chance constrained programming can be converted into:

[0099]

[0100] In the formula: μ is the equivalent expectation; σ is the equivalent variance; Φ is the standard normal distribution function.

[0101] In the embodiment of the present disclosure, when the day-ahead optimization scheduling is performed, due to the great uncertainty of wind power, photovoltaic and load, for the optimization problem with random variables, the traditional method cannot be well solved, and it is necessary to consider applying the stochastic programming theory to establish a suitable dynamic optimization scheduling model.

[0102] In one embodiment, as shown in Figure 3 The optimization scheme of the virtual power plant model obtained in the frequency modulation market mechanism includes: obtaining the optimization scheme of the virtual power plant model by using the model predictive control (MPC) technology in the intraday market.

[0103] Optionally, the model predictive control technology refers to a model-based closed-loop optimization control algorithm, and the idea is to combine the system model, the current state quantity and the constraint condition to solve the optimal control input variable on line.

[0104] In one specific embodiment, in the model predictive control technology, at the current time t, based on a certain prediction model, the system output state prediction y(t+k|t), k=1, 2, …, N in the future finite time domain N is obtained, the system prediction output y(t+k|t) is solved according to the system input and output at the current time and the control signal u(t+k|t), k=0, 1, …, N-1 in the future control time domain, the system output is combined with the reference trajectory, the given performance evaluation standard is followed, the constraint conditions in the current and future time domains are considered, the optimization problem in the future control time domain is solved on line, the control instruction sequence u(t+k|t) in the next finite time domain of the system is obtained, only the first value of the control instruction sequence is issued to the control system for operation, the system output y(t+1) is resampled at the next time, and the above steps are repeated to perform rolling optimization.

[0105] In one embodiment, the above-mentioned optimization scheme according to the virtual power plant model optimizes scheduling of the virtual power plant, including: performing day-ahead optimization scheduling of the virtual power plant according to a target function corresponding to the day-ahead market in the optimization scheme of the virtual power plant model.

[0106] Optionally, in the day-ahead market, the target function for the optimization scheduling of the virtual power plant is to maximize the virtual power plant revenue, and the expression is as follows:

[0107]

[0108] In the formula, T is a scheduling period, which is 24 hours in the day-ahead; the total revenue is divided into two parts, which are the revenue of the virtual power plant participating in the electricity market at time t and the revenue of the virtual power plant participating in the frequency modulation market Since the maintenance costs of electric vehicles, energy storage, wind power, and photovoltaic are small in the scale of one day, they are not considered;

[0109] The specific expression of each part of the day-ahead revenue of the virtual power plant (VPP) is as follows:

[0110] The revenue of the VPP participating in the electricity market is:

[0111]

[0112] In the formula, and are the selling price and the purchasing price of the regional power distribution network in the day-ahead t period, respectively; t sell and P t buy are the selling power and the purchasing power of the VPP and the regional power distribution network in the t period, respectively, and P is the sum of the powers of the members; Δt is a time period, which is 1 hour in the day-ahead;

[0113] The revenue of the VPP participating in the frequency modulation market is:

[0114]

[0115] In the formula, and are the frequency modulation capacity price and the mileage price of the VPP in the day-ahead t period predicted according to historical data, respectively; t c and P t M are the frequency modulation capacity and the frequency modulation mileage of the VPP and the regional power distribution network in the t period, respectively, the frequency modulation capacity is the sum of the frequency modulation capacities of the members, and the frequency modulation mileage is the sum of the frequency modulation signal mileages issued by the regional power distribution network, in the example of the present chapter, the frequency modulation mileage used in the day-ahead is simulated by using historical mileage data; and Frequency regulation performance index.

[0116] In one specific embodiment, the day-ahead optimization scheduling divides a day into 24 periods with 1 hour as a time interval. The centralized controller of the virtual power plant (VPP) makes predictions of wind power, photovoltaic power and load to obtain predicted values of 24 periods. The optimization is performed once a day to obtain a 24-hour operation plan with the maximum VPP revenue as the optimization objective. The VPP day-ahead optimization scheduling process is shown in Figure 4 .

[0117] In one embodiment, the optimization scheme according to the virtual power plant model is day-ahead optimization scheduling of the virtual power plant by a target function corresponding to a day-ahead market, including: obtaining constraint conditions of each component member in the virtual power plant model in the day-ahead market.

[0118] Optionally, the constraints of each component member in the virtual power plant are as follows:

[0119] Electric vehicle constraints

[0120]

[0121]

[0122]

[0123]

[0124]

[0125] In the formula: the i-th electric vehicle discharges to the regional power grid in the t-th period, is the electric vehicle power, a positive value indicates discharging to the regional power grid in the V2G state, and a negative value indicates charging; is the reserved upward frequency regulation power, is the reserved downward frequency regulation power; is the maximum charging and discharging power; is a Boolean variable indicating whether the electric vehicle is connected to the regional power grid; is the state of charge; and are the minimum and maximum states of charge; and are the charging and discharging efficiencies; is the battery capacity; is the amount of electricity when leaving the regional power grid; is the amount of electricity to be reached when leaving the regional power grid.

[0126] Controllable load constraints

[0127] A variable frequency air conditioner is taken as a typical controllable load for study.

[0128] 1. Modeling of variable frequency air conditioners

[0129] Buildings are modeled using the equivalent thermal parameters (ETP) modeling method based on circuit simulation. To simplify the calculation, a first-order ETP model is used for analysis:

[0130]

[0131] where T o is the outside temperature; T i is the indoor air temperature; Q AC is the air conditioning cooling capacity; C a is the equivalent specific heat capacity; and R1 is the equivalent impedance.

[0132] The electrical-thermal conversion model of the air conditioner is used to describe the relationship between the air conditioning cooling capacity and the consumed electrical power. The electrical power P AC of the variable frequency air conditioner and the cooling capacity Q AC are related to the compressor frequency f AC , and the relationship can be expressed as follows:

[0133]

[0134] where k1, k2, l1, and l2 are constant coefficients.

[0135] According to the thermodynamic model and the electrical-thermal conversion model of the variable frequency air conditioner, the relationship between the indoor temperature and the electrical power of the air conditioner can be obtained.

[0136] The predicted mean vote (PMV) index I PMV is used to estimate the temperature comfort, and the main influencing factors are the indoor temperature, air pressure, humidity, etc. The appropriate value range is:

[0137] -1≤I PMV ≤1

[0138] It is generally believed that when the indoor air pressure and humidity are within the appropriate range, the acceptable range of temperature fluctuations for human comfort is ±1℃. Therefore, the temperature comfort zone is set to be within the range of ±1℃ of the most comfortable temperature for the human body. To ensure the temperature comfort requirements of users, the air conditioner control strategy proposed in this paper controls the indoor temperature to change within the temperature comfort zone, and it is assumed that the air conditioner set temperature T iset is the most comfortable temperature for the human body.

[0139] 2. Constraints of variable frequency air conditioners

[0140]

[0141]

[0142]

[0143] T imin ≤T i ≤T imax

[0144] where: the ith air conditioning load in the t period, is the air conditioning power; and are the maximum and minimum air conditioning power; and are the up and down frequency capacity; T i is the indoor temperature; T imax and T imin are the upper and lower limits of the temperature comfort zone, respectively.

[0145] The constraints set guarantee the user's temperature comfort requirements.

[0146] Energy storage constraints:

[0147] 0≤P t es +P t esup ≤P t esmax

[0148] -P t esmax ≤P t es -P t esdn ≤0

[0149]

[0150]

[0151]

[0152] where: the energy storage in the t period, P t es is the energy storage power, positive value represents discharge, negative value represents charge; P t esup is the up frequency power left, P t esdn is the down frequency power left; P t esmax is the maximum charge and discharge power; is the state of charge; and are the minimum and maximum state of charge; ηesch and η esdc are charge and discharge efficiency; E es is battery capacity; is the final power at the end of the dispatch period; is the initial power at the beginning of the dispatch period. The power stored in the battery at the beginning and end of the dispatch period is equal, thus ensuring the continuity of the dispatch.

[0153] Wind turbine constraints:

[0154] The wind turbine is in an unloaded state during operation, i.e. a certain degree of wind curtailment is used to provide the upward frequency modulation power. In the day-ahead prediction, there is a certain error in the prediction of wind power, and the error can be considered to be subject to normal distribution. In order to avoid the penalty due to large errors in the intra-day dispatch, a conservative strategy is adopted to constrain the power below the expected power, however, this will affect the economy of the virtual power plant. However, the occurrence of some extreme cases is particularly low, so the opportunity constraint can be satisfied at a certain confidence level, and the economy and reliability are balanced.

[0155] P{0≤P t wdE +P t wdup ≤P t wd}≥α

[0156] P t wddn ≤P t wdE

[0157] In the formula: P t wdE is the power of the wind turbine participating in the energy market; P t wd is the expected value of the day-ahead wind power prediction; P t wdup and P t wddn are upward and downward frequency modulation powers respectively; and α is the confidence level.

[0158] Since the wind power prediction error satisfies the normal distribution, the opportunity constraint can be converted into a deterministic constraint for solving.

[0159] Photovoltaic constraints:

[0160] The photovoltaic operation strategy is similar to that of the wind turbine.

[0161] P{0≤P t pvE +P t pvup ≤P t pv}≥α

[0162] P t pvdn ≤P t pvE

[0163] In the formula: P t pvE Photovoltaic units participating in the energy market; P t pv P represents the expected value of the day-to-day photovoltaic power forecast. t pvup and P t pvdn α represents the up and down frequency modulation power, respectively; α is the confidence level.

[0164] Power constraints:

[0165]

[0166] In the formula: P t VPP The total power of the virtual power plant during time period t is the power sold to the regional distribution network when it is greater than zero, and the power purchased from the regional distribution network when it is less than zero; EVnum is the number of electric vehicles connected to the regional distribution network; ACnum is the number of variable frequency air conditioners turned on.

[0167] Frequency modulation capacity constraints:

[0168]

[0169]

[0170] P t c =min(P t VPPup ,P t VPPdn )

[0171] In the formula: P t VPPup and P t VPPdn These represent the total up-frequency regulation and down-frequency regulation capacities of the virtual power plant, respectively; P t c The virtual power plant reports the regional distribution network capacity. Since the referenced market mechanism requires that the upper and lower frequency regulation capacities be consistent, the reported capacity is the smaller value between the upper and lower frequency regulation capacities.

[0172] In one embodiment, the above-mentioned day-ahead optimization scheduling of the virtual power plant based on the objective function corresponding to the day-ahead market according to the optimization scheme of the virtual power plant model further includes: obtaining the confidence level settings of each component in the virtual power plant model.

[0173] In the embodiments of the present disclosure, since wind power and photovoltaic have the characteristic that the closer the distance to the prediction time, the more accurate the prediction is, the confidence level of wind power and photovoltaic prediction can be set according to the distance to the prediction time. As shown in FIG. 7. Within 15 minutes, the prediction of wind power and photovoltaic is considered to be accurate, and therefore the confidence level is 1; the longer the distance to the prediction time, the smaller the confidence level. Figure 5

[0174] In one embodiment, the optimization scheme according to the virtual power plant model optimizes the scheduling of the virtual power plant, including: performing intraday optimization scheduling of the virtual power plant according to the objective function corresponding to the intraday market of the optimization scheme of the virtual power plant model.

[0175] In the embodiments of the present disclosure, the intraday rolling optimization is performed at a time interval of 15 minutes and an optimization period of 4 hours. The centralized controller of the virtual power plant predicts the wind power, photovoltaic and load within 4 hours, and performs rolling optimization every 15 minutes to determine the operation plan for the next 4 hours with the target of minimum penalty in the intraday. At this time, the prediction accuracy is higher than that of the day-ahead, and the error within 15 minutes can be ignored, thereby reducing the influence of the uncertainty of the day-ahead prediction value on the intraday operation.

[0176] Although there is a certain error in the day-ahead planning, the day-ahead scheduling ensures the optimal operation in the scale of one day. When the intraday rolling is performed, the day-ahead plan is modified based on the real-time market price and prediction, and the day-ahead scheduling plan is still followed in the long time scale. The period of the rolling optimization is 15 minutes, and the scale is 4 hours. The intraday optimization scheduling process of the VPP is shown in FIG. 8. Figure 6

[0177] Optionally, the objective function corresponding to the intraday market is as follows:

[0178]

[0179] In the formula, ψ f is the penalty price of the frequency response deviation; ψ e is the penalty price of the energy market deviation; ψ f is the deviation of the frequency signal tracking; ψ e is the deviation of the intraday energy market; and α f and α e are the penalty coefficients of the frequency market and the energy market respectively. R,t E,t cp E

[0180] In one embodiment, the optimization scheme according to the virtual power plant model optimizes the scheduling of the virtual power plant in the intraday market, including: obtaining the constraint conditions of each component member in the virtual power plant model in the intraday market.

[0181] ​​​​​​​​Optionally, the day-ahead optimized scheduling result should be the same as the state of charge obtained by the day-ahead optimized scheduling.

[0182] SOC es = SOC esREF

[0183]

[0184] wherein: SOC es is the state of charge of the energy storage in real time in a day; is the state of charge of the electric vehicle in real time in a day; and SOC esREF are the states of charge of the electric vehicle and the energy storage obtained by the day-ahead optimized scheduling. The constraints of the day-ahead rolling optimization are basically the same as those of the day-ahead, and here only the power constraints of the electric vehicle and the energy storage are added.

[0185] In one embodiment, the above optimization scheme according to the virtual power plant model is used for day-ahead optimization scheduling of the virtual power plant by a target function corresponding to the day-ahead market, and further comprises: obtaining a power distribution strategy of each component member of the virtual power plant in the day-ahead market.

[0186] Optionally, the power distribution strategy of each component member of the virtual power plant is as follows:

[0187] In the day-ahead market, the virtual power plant receives a frequency regulation signal issued by a dispatching center and distributes it to each member. In order to better guarantee the travel demand and temperature comfort demand of users, the virtual power plant first issues the frequency regulation signal to wind power, photovoltaic and energy storage, and the insufficient part is borne by the electric vehicle and air conditioning load.

[0188] First, the frequency regulation capacity is distributed according to the frequency regulation capacity ratio of the electric vehicle and the air conditioning load:

[0189]

[0190]

[0191] wherein: P up and P dn are the up and down frequency regulation signals distributed to the electric vehicle and the air conditioning load respectively; P EVup_tar and P EVdn_tar are the up and down frequency regulation signals distributed to the electric vehicle respectively; P ACup_tar and P ACdn_tar are the up and down frequency regulation signals distributed to the air conditioning load respectively.

[0192] The state of charge of the electric vehicle is sorted, and the electric vehicle with the state of charge higher than the state of charge obtained by the day-ahead optimization scheduling is preferentially selected to respond to the frequency increase signal until the total frequency increase power meets the requirement; similarly, the electric vehicle with the state of charge lower than the state of charge obtained by the day-ahead optimization scheduling is preferentially selected to respond to the frequency decrease, until the total frequency decrease power meets the requirement.

[0193] The frequency modulation capacity required by the air conditioning load is allocated to the air conditioners in the temperature comfort zone; in order to better stabilize the indoor temperature in the temperature comfort zone, the air conditioner with higher indoor temperature is allocated more frequency decrease power to participate in more frequency decrease; similarly, the air conditioner with lower indoor temperature is allocated more frequency increase power to participate in more frequency increase;

[0194] Firstly, the normalized index of room temperature is defined:

[0195]

[0196] In the formula: SOA is the normalized index of room temperature; the closer the room temperature is to the set temperature, the closer SOA is to zero; when the room temperature is in the temperature comfort zone, the range of SOA is [-1, 1];

[0197]

[0198] In the formula: respectively, the frequency increase power and the frequency decrease power allocated to the kth air conditioner; ACnum is the total number of air conditioners in the temperature comfort zone; SOA k is the normalized index of the kth air conditioner.

[0199] Embodiment 2:

[0200] The embodiment of the present disclosure provides a secondary frequency modulation system of a regional power distribution network based on a virtual power plant, comprising:

[0201] A data processing module is configured to obtain a virtual power plant model of the architecture and a frequency modulation market mechanism;

[0202] A data optimization module is configured to obtain an optimization scheme of the virtual power plant model under the frequency modulation market mechanism;

[0203] An optimization scheduling module is configured to perform optimization scheduling on the virtual power plant according to the optimization scheme of the virtual power plant model.

[0204] The working method of the above system is the same as the secondary frequency modulation method of the regional power distribution network based on the virtual power plant provided in the above embodiments, and will not be repeated here.

[0205] Embodiment 3:

[0206] The embodiment of the present disclosure provides a storage medium, which stores a program, and the program is executed by a processor to realize steps in the secondary frequency modulation method of the regional power distribution network based on the virtual power plant provided by each of the above embodiments, including:

[0207] obtaining a virtual power plant model of an architecture and a frequency modulation market mechanism;

[0208] obtaining an optimization scheme of the virtual power plant model under the frequency modulation market mechanism;

[0209] optimizing scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model.

[0210] The detailed steps of the method realized by the above program are the same as the secondary frequency modulation method of the regional power distribution network based on the virtual power plant provided by each of the above embodiments, and details are not repeated here.

[0211] Embodiment 4:

[0212] The embodiment of the present disclosure provides an electronic device, including a memory, a processor and a program stored in the memory and executable on the processor, and the processor realizes steps in the secondary frequency modulation method of the regional power distribution network based on the virtual power plant provided by each of the above embodiments when executing the program, including:

[0213] obtaining a virtual power plant model of an architecture and a frequency modulation market mechanism;

[0214] obtaining an optimization scheme of the virtual power plant model under the frequency modulation market mechanism;

[0215] optimizing scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model.

[0216] The detailed steps of the method realized by the above program are the same as the secondary frequency modulation method of the regional power distribution network based on the virtual power plant provided by each of the above embodiments, and details are not repeated here.

[0217] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system or a computer program product. Therefore, the present disclosure can be in the form of a hardware embodiment, a software embodiment or an embodiment combining software and hardware aspects. Moreover, the present disclosure can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer usable program code.

[0218] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure One one or more computer-readable media. Figure One one or more computer-readable media.

[0219] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device that implements the function specified in the flowchart block or blocks. Figure One one or more computer-readable media. Figure One one or more computer-readable media.

[0220] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure One one or more computer-readable media. Figure One one or more computer-readable media.

[0221] Those skilled in the art can understand that all or part of the above-mentioned embodiment methods can be implemented by computer programs instructing relevant hardware, and the programs can be stored in a computer-readable storage medium and can include the processes of the above-mentioned embodiment methods when executed. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), or the like.

[0222] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A secondary frequency regulation method for a regional distribution network based on a virtual power plant, characterized in that, Includes the following steps: Obtain the architecture of the virtual power plant model and the frequency regulation market mechanism; The optimized scheme of the virtual power plant model is obtained under the frequency regulation market mechanism, which divides the market into day-ahead and intraday markets. In the day-ahead market, with the objective function being the maximization of the virtual power plant's revenue, long-term constraints are established. The corresponding objective function in the day-ahead market is: In the formula: T is a scheduling period, which is taken as 24 hours in the day-ahead period; the total revenue is divided into two parts, namely the revenue of the virtual power plant participating in the electricity market during time period t. and the revenue from participating in the FM market In the intraday market, with the objective function being the minimization of the virtual power plant penalty, constraints are established for a short time scale. The corresponding objective function in the intraday market is: In the formula: The penalty price for frequency modulation response deviation; The penalty price for energy market deviations; ψ R,t The deviation in tracking the frequency modulation signal; ψ E,t For intraday energy market deviations; α cp and α E These are the penalty coefficients for the frequency regulation market and the energy market, respectively. The virtual power plant is optimized and scheduled according to the optimization scheme of the virtual power plant model, including: Obtain the constraints of each component in the virtual power plant model within the intraday market; The power allocation strategies of each component of the virtual power plant in the intraday market are obtained, and the power allocation strategies of each component of the virtual power plant are as follows: In the intraday market, the virtual power plant first sends frequency regulation signals to wind power, photovoltaics and energy storage, and the remaining portion is covered by electric vehicles and air conditioning loads; First, allocate frequency regulation capacity according to the ratio of frequency regulation capacity of electric vehicles and air conditioning load: In the formula: P up and P dn These are the up-modulation and down-modulation signals allocated to the electric vehicle and air conditioning load, respectively; P EVup_tar and P EVdn_tar These are the up-modulation and down-modulation signals allocated to the electric vehicle, respectively; P ACup_tar and P ACdn_tar The up-frequency modulation signal and down-frequency modulation signal allocated to the air conditioning load; The electric vehicles are sorted by state of charge (SOC) and electric vehicles with SOCs higher than those obtained from the day-ahead optimized scheduling are selected to respond to the up-frequency modulation signal until the total up-frequency modulation power meets the requirements. Electric vehicles with SOCs lower than those obtained from the day-ahead optimized scheduling are selected to respond to the down-frequency modulation signal until the total down-frequency modulation power meets the requirements. The required frequency regulation capacity for air conditioning load is allocated to air conditioners in the temperature comfort zone; more down-frequency regulation power is allocated to air conditioners with higher indoor temperatures, enabling them to participate in more down-frequency regulation; more up-frequency regulation power is allocated to air conditioners with lower indoor temperatures, enabling them to participate in more up-frequency regulation. First, define a normalized index for room temperature: In the formula: SOA is the normalized index of room temperature; the closer the room temperature is to the set temperature, the closer the SOA is to zero; when the room temperature is within the temperature comfort zone, the SOA ranges from [-1, 1]. In the formula: These represent the up-frequency regulation power and down-frequency regulation power allocated to the k-th air conditioner, respectively; ACnum is the total number of air conditioners in the temperature comfort zone; SOA k Let be the normalized index of the k-th air conditioner.

2. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 1, characterized in that, Obtaining an optimized solution for the virtual power plant model under the frequency regulation market mechanism includes: An optimal solution for the virtual power plant model is obtained using opportunity-constrained programming in the day-ahead market.

3. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 2, characterized in that, The chance-constrained programming described is a type of stochastic programming. Chance-constrained programming theory allows for situations where the objective function does not satisfy the constraints under special circumstances. The mathematical form of chance-constrained programming is: In the formula: f(x) is the objective function; g j (x,ξ) are random constraint functions; α j is the probability; j represents the number of constraints; st indicates that it is subject to; Pr represents the probability of the condition occurring; The main characteristic of chance-constrained programming is that it imposes certain requirements on the probabilities of the constraints, namely: Pr{g j (x,ξ)≤0}≥α j This is called a chance constraint. When solving chance-constrained programming problems, constraints with random variables are transformed into deterministic constraints. Specifically, when the variables are independent and normally distributed, chance-constrained programming can be transformed into: In the formula: μ is the equivalent expectation; σ is the equivalent variance; Φ is the standard normal distribution function.

4. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 1, characterized in that, Obtaining an optimized solution for the virtual power plant model under the frequency regulation market mechanism includes: The optimization scheme of the virtual power plant model is obtained by using model predictive control technology in the intraday market.

5. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 4, characterized in that, The model predictive control technology refers to a model-based closed-loop optimization control algorithm. Its idea is to combine the system model, current state variables, and constraints to solve for the optimal control input variables online in a rolling manner.

6. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 1, characterized in that, The step of optimizing the scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model includes: Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market.

7. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 6, characterized in that, In the aforementioned day-ahead market, the objective function for the optimized scheduling of the virtual power plant is to maximize the revenue of the virtual power plant, as expressed below: In the formula: T is a scheduling period, which is taken as 24 hours in the day-ahead period; the total revenue is divided into two parts, namely the revenue of the virtual power plant participating in the electricity market during time period t. and the revenue from participating in the FM market Since the maintenance costs of electric vehicles, energy storage, wind power, and photovoltaics are relatively small on a daily scale, they are not considered. The specific expression for each part of VPP's day-ahead earnings is as follows: VPP participation in the electricity market benefits: In the formula: and These represent the predicted electricity sales price and purchase price for the regional distribution network during the day-ahead time period t; P t sell and P t buy These represent the electricity sold and purchased by the VPP in the regional distribution network during time period t, respectively, and are the sum of the power of each member; Δt is a time period, which is 1 hour before the day of the month; Benefits of VPP participation in the FM market: In the formula: and These are the frequency regulation capacity price and mileage price predicted by VPP based on historical data for the current day's time period t; P t c and P t M These are the frequency regulation capacity and frequency regulation mileage traded between VPP and the regional distribution network during time period t. The frequency regulation capacity is the sum of the frequency regulation capacities of each member, and the frequency regulation mileage is the sum of the mileage of the frequency regulation signals issued by the regional distribution network. The frequency regulation mileage used recently is obtained by simulation through historical mileage data. and Frequency modulation performance indicators.

8. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 6, characterized in that, Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market, including: Obtain the constraints of each component in the virtual power plant model in the current market.

9. The secondary frequency regulation method for regional distribution networks based on virtual power plants as described in claim 6, characterized in that, Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market, and further includes: Obtain the confidence level settings for each component in the virtual power plant model.

10. A secondary frequency regulation system for a regional distribution network based on a virtual power plant, characterized in that, include: The data processing module is configured to: obtain the virtual power plant model of the architecture and the frequency regulation market mechanism; The data optimization module is configured to: obtain an optimized solution for the virtual power plant model under the frequency regulation market mechanism, which divides the market into a day-ahead market and an intraday market; in the day-ahead market, establish long-term constraints with the objective function of maximizing the virtual power plant's revenue; the corresponding objective function in the day-ahead market is: In the formula: T is a scheduling period, which is taken as 24 hours in the day-ahead period; the total revenue is divided into two parts, namely the revenue of the virtual power plant participating in the electricity market during time period t. and the revenue from participating in the FM market In the intraday market, with the objective function being the minimization of the virtual power plant penalty, constraints are established for a short time scale. The corresponding objective function in the intraday market is: In the formula: The penalty price for frequency modulation response deviation; The penalty price for energy market deviations; ψ R,t The deviation in tracking the frequency modulation signal; ψ E,t For intraday energy market deviations; α cp and α E These are the penalty coefficients for the frequency regulation market and the energy market, respectively. The optimized scheduling module is configured to: perform optimized scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model, including: Obtain the constraints of each component in the virtual power plant model within the intraday market; The power allocation strategies of each component of the virtual power plant in the intraday market are obtained, and the power allocation strategies of each component of the virtual power plant are as follows: In the intraday market, the virtual power plant first sends frequency regulation signals to wind power, photovoltaics and energy storage, and the remaining portion is covered by electric vehicles and air conditioning loads; First, allocate frequency regulation capacity according to the ratio of frequency regulation capacity of electric vehicles and air conditioning load: In the formula: P up and P dn These are the up-modulation and down-modulation signals allocated to the electric vehicle and air conditioning load, respectively; P EVup_tar and P EVdn_tar These are the up-modulation and down-modulation signals allocated to the electric vehicle, respectively; P ACup_tar and P ACdn_tar The up-frequency modulation signal and down-frequency modulation signal allocated to the air conditioning load; The electric vehicles are sorted by state of charge (SOC) and electric vehicles with SOCs higher than those obtained from the day-ahead optimized scheduling are selected to respond to the up-frequency modulation signal until the total up-frequency modulation power meets the requirements. Electric vehicles with SOCs lower than those obtained from the day-ahead optimized scheduling are selected to respond to the down-frequency modulation signal until the total down-frequency modulation power meets the requirements. The required frequency regulation capacity for air conditioning load is allocated to air conditioners in the temperature comfort zone; more down-frequency regulation power is allocated to air conditioners with higher indoor temperatures, enabling them to participate in more down-frequency regulation; more up-frequency regulation power is allocated to air conditioners with lower indoor temperatures, enabling them to participate in more up-frequency regulation. First, define a normalized index for room temperature: In the formula: SOA is the normalized index of room temperature; the closer the room temperature is to the set temperature, the closer the SOA is to zero; when the room temperature is within the temperature comfort zone, the SOA ranges from [-1, 1]. In the formula: These represent the up-frequency regulation power and down-frequency regulation power allocated to the k-th air conditioner, respectively; ACnum is the total number of air conditioners in the temperature comfort zone; SOA k Let be the normalized index of the k-th air conditioner.

11. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 10, characterized in that, Obtaining an optimized solution for the virtual power plant model under the frequency regulation market mechanism includes: An optimal solution for the virtual power plant model is obtained using opportunity-constrained programming in the day-ahead market.

12. The secondary frequency regulation system for regional distribution networks based on a virtual power plant as described in claim 11, characterized in that, The chance-constrained programming described is a type of stochastic programming. Chance-constrained programming theory allows for situations where the objective function does not satisfy the constraints under special circumstances. The mathematical form of chance-constrained programming is: In the formula: f(x) is the objective function; g j (x,ξ) are random constraint functions; α j is the probability; j represents the number of constraints; st indicates that it is subject to; Pr represents the probability of the condition occurring; The main characteristic of chance-constrained programming is that it imposes certain requirements on the probabilities of the constraints, namely: Pr{g j (x,ξ)≤0}≥α j This is called a chance constraint. When solving chance-constrained programming problems, constraints with random variables are transformed into deterministic constraints. Specifically, when the variables are independent and normally distributed, chance-constrained programming can be transformed into: In the formula: μ is the equivalent expectation; σ is the equivalent variance; Φ is the standard normal distribution function.

13. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 10, characterized in that, Obtaining an optimized solution for the virtual power plant model under the frequency regulation market mechanism includes: The optimization scheme of the virtual power plant model is obtained by using model predictive control technology in the intraday market.

14. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 13, characterized in that, The model predictive control technology refers to a model-based closed-loop optimization control algorithm. Its idea is to combine the system model, current state variables, and constraints to solve for the optimal control input variables online in a rolling manner.

15. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 10, characterized in that, The step of optimizing the scheduling of the virtual power plant according to the optimization scheme of the virtual power plant model includes: Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market.

16. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 15, characterized in that, In the aforementioned day-ahead market, the objective function for the optimized scheduling of the virtual power plant is to maximize the revenue of the virtual power plant, as expressed below: In the formula: T is a scheduling period, which is taken as 24 hours in the day-ahead period; the total revenue is divided into two parts, namely the revenue of the virtual power plant participating in the electricity market during time period t. and the revenue from participating in the FM market Since the maintenance costs of electric vehicles, energy storage, wind power, and photovoltaics are relatively small on a daily scale, they are not considered. The specific expression for each part of VPP's day-ahead earnings is as follows: VPP participation in the electricity market benefits: In the formula: and These represent the predicted electricity sales price and purchase price for the regional distribution network during the day-ahead time period t; P t sell and P t buy These represent the electricity sold and purchased by the VPP in the regional distribution network during time period t, respectively, and are the sum of the power of each member; Δt is a time period, which is 1 hour before the day of the month; Benefits of VPP participation in the FM market: In the formula: and These are the frequency regulation capacity price and mileage price predicted by VPP based on historical data for the current day's time period t; P t c and P t M These are the frequency regulation capacity and frequency regulation mileage traded between VPP and the regional distribution network during time period t. The frequency regulation capacity is the sum of the frequency regulation capacities of each member, and the frequency regulation mileage is the sum of the mileage of the frequency regulation signals issued by the regional distribution network. The frequency regulation mileage used recently is obtained by simulation through historical mileage data. and Frequency modulation performance indicators.

17. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 15, characterized in that, Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market, including: Obtain the constraints of each component in the virtual power plant model in the current market.

18. The secondary frequency regulation system for a regional distribution network based on a virtual power plant as described in claim 15, characterized in that, Based on the optimization scheme of the virtual power plant model, the day-ahead optimization scheduling of the virtual power plant is performed using the objective function corresponding to the day-ahead market, and further includes: Obtain the confidence level settings for each component in the virtual power plant model.

19. A storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the secondary frequency regulation method for regional distribution networks based on virtual power plants as provided in any one of claims 1 to 9.

20. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the secondary frequency regulation method for regional distribution networks based on virtual power plants as provided in any one of claims 1 to 9.

Images