Power distribution network coordination control method and device based on average field game

Abstract

The invention discloses a power distribution network coordination control method and device based on an average field game. The method comprises the following steps: giving a power constraint of each inverter; and based on the inverter model, constructing an aggregation model corresponding to the plurality of inverters in the power distribution network. Constructing a first objective function according to supply and demand balance of the power distribution network; and constructing a second objective function based on the power gain, the frequency adjustment gain and the operation loss of the inverter. And obtaining a mean field game objective function by combining a mean field theory, the first objective function and the second objective function. And solving the average field game objective function by adopting a strategy iterative algorithm to obtain a power control strategy of the inverter in the power distribution network. An aggregation model of a plurality of inverters is constructed, a mean field game objective function is obtained based on supply and demand balance of a power distribution network and economic benefits generated by the inverters and in combination with a mean field theory, then a power control strategy of the inverters is solved, and the economic benefits are maximized while stable operation of the power grid is ensured when the inverters are connected to the power grid.

Description

Technical Field

[0001] The present invention belongs to the technical field of coordinated control of distribution networks, and in particular relates to a coordinated control method and device for distribution networks based on mean field game. Background Art

[0002] In a power system that includes renewable energy, as the amount of renewable energy connected to the grid through inverters increases, the grid structure will change.

[0003] The grid connection mode and control strategy of the inverter not only determine the safe and stable operation of the inverter, but also have an important impact on the overall stability of the power grid. According to the different control methods, the inverter can be divided into two types: grid-forming inverter (GFM) and grid-following inverter (GFL). The grid-forming inverter adopts a power synchronization strategy similar to that of a synchronous generator, which can provide frequency and voltage support.

[0004] Patent CN115173472A discloses a novel hybrid control method and system for wind power photovoltaic grid-connected converters, including constructing a grid-following outer loop control strategy with a DC voltage-reactive power outer loop, and introducing a GFL control participation coefficient; constructing a grid-following inner loop control strategy, and introducing a GFL control participation coefficient to obtain a GFL control voltage reference value; constructing a DC voltage-frequency droop grid-connected control strategy with a stable control strategy, and introducing a GFM control participation coefficient to obtain a GFM control reference value; synthesizing the GFL control voltage reference value and the GFM control reference value to obtain a converter output three-phase voltage reference value. By mixing the two control modes of GFL and GFM and introducing a participation coefficient, a converter hybrid control strategy with an adjustable participation ratio is realized, thereby better coordinating the response speed and stability of the converter and ensuring stable operation of the converter.

[0005] The relevant technology only considers the stability of inverter and grid operation when a large number of inverters are connected to the distribution network. How to maximize the economic benefits of the grid while ensuring the stable operation of the grid when the inverters are connected to the grid is a problem that needs to be solved at present. Summary of the invention

[0006] In view of the defects existing in the above-mentioned prior art, the present invention provides a distribution network coordinated control method and device based on mean field game, including: providing a power constraint for each inverter. Based on the inverter model, constructing an aggregation model corresponding to multiple inverters in the distribution network. Constructing a first objective function according to the supply and demand balance of the distribution network. Constructing a second objective function based on the power gain, frequency regulation benefit and operating loss of the inverter. Combining the mean field theory, the first objective function and the second objective function, the mean field game objective function is obtained. Using a strategy iteration algorithm, the mean field game objective function is solved to obtain a power control strategy of the inverter in the distribution network. By constructing an aggregation model of multiple inverters, based on the supply and demand balance of the distribution network and the economic benefits generated by the inverter and combined with the mean field theory, the mean field game objective function is obtained, and then the power control strategy of the inverter is solved, so as to maximize the economic benefits while ensuring the stable operation of the power grid when the inverter is connected to the power grid.

[0007] In a first aspect, the present invention provides a distribution network coordinated control method based on mean field game, comprising:

[0008] Give the power constraints of the inverters in the distribution network;

[0009] Based on the inverter model, an aggregation model corresponding to multiple inverters in the distribution network is constructed;

[0010] According to the supply and demand balance of the distribution network, a first objective function is constructed;

[0011] A second objective function is constructed based on the power gain, frequency regulation benefit and operation loss of the inverter;

[0012] Combining the mean field theory, the first objective function and the second objective function, we get the mean field game objective function;

[0013] The strategy iteration algorithm is used to solve the mean field game objective function and obtain the power control strategy of the inverter in the distribution network.

[0014] It can be seen from the above technical solution that by constructing an aggregation model of multiple inverters, the mutual influence between inverters is taken into account. Based on the supply and demand balance of the distribution network and the power gain, frequency regulation benefit and operating loss of the inverter, and combined with the mean field theory, the mean field game objective function is obtained. On the basis of considering the economic benefits generated by the inverter, the power control strategy of the inverter is solved to maximize the economic benefits of the power grid while ensuring the stable operation of the power grid when the inverter is connected to the power grid.

[0015] In a possible implementation manner, the power constraint of the inverter in the distribution network is specifically expressed as:

[0016] P i ∈[P i,min ,Pi,max ]

[0017] P i,min =min((P i +ΔP i ),P N,i )

[0018] P i,max =max((P i -ΔP i ),0)

[0019]

[0020] Among them, P i is the active power output by the ith inverter, P i,min is the minimum value of the active power output by the ith inverter, P i,max is the maximum value of the active power output by the i-th inverter, P N,i is the rated active power of the i-th inverter, ΔP i is the maximum value of the active power output by the ith inverter within the frequency fluctuation range allowed by the power system, Δf max is the maximum frequency fluctuation within the frequency fluctuation range allowed by the power system, m p,i is the first droop coefficient of the i-th inverter.

[0021] In a possible implementation, based on the inverter model, an aggregation model corresponding to multiple inverters in the distribution network is constructed, specifically including:

[0022] The first inverter model of the relationship between the frequency of the inverter controlled by the first droop coefficient and the output active power is given, which is specifically expressed as:

[0023] P i =P i,ref -mp ,i (f i -f i,ref )

[0024] Among them, P i is the active power output by the ith inverter, P i,ref is the active power setting value of the i-th inverter, m p,i is the first droop coefficient of the i-th inverter, f i is the frequency of the ith inverter, f i,ref is the rated frequency of the i-th inverter;

[0025] A second inverter model showing the relationship between the voltage amplitude of the inverter controlled by the second droop coefficient and the output reactive power is given, which is specifically expressed as:

[0026] Q i =Q i,ref -m q,i (V i -V i,ref )

[0027] Among them, Q i is the reactive power output by the i-th inverter, Q i,ref is the reactive power setting value of the i-th inverter, m q,i is the second droop coefficient of the ith inverter, V i is the actual voltage of the ith inverter, V i,ref is the rated voltage of the i-th inverter;

[0028] The first inverter model and the second inverter model are integrated to form an inverter model;

[0029] Combine the connection relationship of the multiple inverters and obtain a first aggregation model by aggregating the active powers output by the multiple inverters;

[0030] Based on the first droop coefficients of the multiple inverters, analyzing and obtaining a first combined droop coefficient after the multiple inverters are connected;

[0031] In combination with the first aggregation model and the first combined droop coefficient, an aggregation model corresponding to multiple inverters in the distribution network is obtained.

[0032] In a possible implementation, a first objective function is constructed according to the supply and demand balance of the distribution network, specifically including:

[0033] The maximum demand and real-time actual demand of the distribution network are obtained, and the first objective function is given in combination with the demand price. The first objective function is specifically expressed as:

[0034] MIN PR=(D max -D)-λ(r price -r price0 ) ξ

[0035] Among them, MIN PR is the first objective function, ξ is the demand elasticity coefficient, D max is the maximum demand, r price is the actual demand price, r price0 is the benchmark demand price, λ is the demand response sensitivity coefficient, and D is the actual demand.

[0036] In a possible implementation, a second objective function is constructed based on the power gain, frequency regulation benefit and operation loss of the inverter, which is specifically expressed as:

[0037] Analyze the relationship between actual demand and demand price, obtain the price function, and give the power gain of the inverter active power;

[0038] Based on the frequency regulation change of the inverter, the frequency regulation benefit of the inverter is given;

[0039] Combined with the operating loss, the second objective function is constructed, which is specifically expressed as:

[0040] MIN J(P i ,P -i )=H(P i ,P -i )+B(P i )-L(P i )

[0041] Among them, MIN J(P i ,P -i ) is the second objective function, H(P i ,P -i ) is the power gain of the active power of the ith inverter, B(P i ) is the frequency regulation benefit of the i-th inverter, L(P i ) is the operating loss of the i-th inverter, P -i is the active power set excluding the output of the i-th inverter.

[0042] In a possible implementation, the mean field game objective function is specifically expressed as:

[0043] J i (P i ,z)=(ξD max -ξz)P i +B(P i )-L(P i )

[0044]

[0045] Among them, J i (P i ,z) is the mean field game function, and z is the mean field.

[0046] In a possible implementation, a strategy iteration algorithm is used to solve the mean field game objective function to obtain a power control strategy for an inverter in a distribution network, which specifically includes:

[0047] Initialize power control strategy;

[0048] Iterate the power control strategy based on the policy iteration equation corresponding to the policy iteration algorithm;

[0049] The power control strategy obtained in each iteration is judged to determine the power control strategy of the inverter.

[0050] In a possible implementation, the strategy iteration equation is specifically expressed as:

[0051]

[0052] Where k is the number of iterations and k ≥ 1, P i,k+1 is the active power output of the ith inverter obtained by the k+1th iteration, P i is the active power output by the i-th inverter, is the average active power output of all inverters in the kth iteration, C i is the cost function of the ith inverter, λ i is the demand response sensitivity coefficient of the i-th inverter, P i d is the required active power output of the i-th inverter, is the average active power output of all inverters in the k+1th iteration, and N is the total number of inverters.

[0053] In a possible implementation manner, judging the power control strategy obtained in each iteration to determine the power control strategy of the inverter specifically includes:

[0054] According to the power control strategy P of the kth iteration i,k and the power control strategy P for the k+1th iteration i,k+1 The relationship between the difference and the preset convergence threshold is used to obtain a first judgment result, wherein k≥1;

[0055] According to the first judgment result, the k+2th iteration or judgment P i,k+1 Whether it is within the preset active power range, the power control strategy of the inverter is obtained.

[0056] In a second aspect, the present invention further discloses a distribution network coordination control device based on mean field game, comprising:

[0057] A model building module, used to build an aggregation model corresponding to multiple inverters in the distribution network based on the inverter model;

[0058] A constraint determination module, used for providing power constraints of inverters in the distribution network;

[0059] A first function building module, used to build a first objective function according to the supply and demand balance of the distribution network;

[0060] A second function building module is used to build a second objective function based on the power gain, frequency regulation benefit and operation loss of the inverter;

[0061] An objective function construction module is used to combine the mean field theory, the first objective function and the second objective function to obtain the mean field game objective function;

[0062] The strategy iteration module is used to solve the mean field game objective function by using the strategy iteration algorithm to obtain the power control strategy of the inverter in the distribution network.

[0063] In a possible implementation manner, the power constraint of the inverter in the distribution network is specifically expressed as:

[0064] P i ∈[P i,min ,P i,max ]

[0065] P i,min =min((P i +ΔP i ),P N,i )

[0066] P i,max =max((P i -ΔP i ),0)

[0067]

[0068] Among them, P i is the active power output by the ith inverter, P i,min is the minimum value of the active power output by the ith inverter, P i,max is the maximum value of the active power output by the i-th inverter, P N,i is the rated active power of the i-th inverter, ΔP i is the maximum value of the active power output by the ith inverter within the frequency fluctuation range allowed by the power system, Δf max is the maximum frequency fluctuation within the frequency fluctuation range allowed by the power system, m p,i is the first droop coefficient of the i-th inverter.

[0069] In a possible implementation, based on the inverter model, an aggregation model corresponding to multiple inverters in the distribution network is constructed, specifically including:

[0070] The first inverter model of the relationship between the frequency of the inverter controlled by the first droop coefficient and the output active power is given, which is specifically expressed as:

[0071] P i =P i,ref -m p,i (f i -f i,ref )

[0072] Among them, P i is the active power output by the ith inverter, P i,ref is the active power setting value of the i-th inverter, m p,i is the first droop coefficient of the i-th inverter, f i is the frequency of the ith inverter, f i,ref is the rated frequency of the i-th inverter;

[0073] A second inverter model showing the relationship between the voltage amplitude of the inverter controlled by the second droop coefficient and the output reactive power is given, which is specifically expressed as:

[0074] Q i =Q i,ref -m q,i (V i -V i,ref )

[0075] Among them, Q i is the reactive power output by the ith inverter, Q i,ref is the reactive power setting value of the i-th inverter, m q,i is the second droop coefficient of the ith inverter, V i is the actual voltage of the ith inverter, V i,ref is the rated voltage of the i-th inverter;

[0076] The first inverter model and the second inverter model are integrated to form an inverter model;

[0077] Combine the connection relationship of the multiple inverters and obtain a first aggregation model by aggregating the active powers output by the multiple inverters;

[0078] Based on the first droop coefficients of the multiple inverters, analyzing and obtaining a first combined droop coefficient after the multiple inverters are connected;

[0079] In combination with the first aggregation model and the first combined droop coefficient, an aggregation model corresponding to multiple inverters in the distribution network is obtained.

[0080] In a possible implementation, a first objective function is constructed according to the supply and demand balance of the distribution network, specifically including:

[0081] The maximum demand and real-time actual demand of the distribution network are obtained, and the first objective function is given in combination with the demand price. The first objective function is specifically expressed as:

[0082] MIN PR=(D max -D)-λ(r price -r price0 ) ξ

[0083] Among them, MIN PR is the first objective function, ξ is the demand elasticity coefficient, D max is the maximum demand, r price is the actual demand price, r price0 is the benchmark demand price, λ is the demand response sensitivity coefficient, and D is the actual demand.

[0084] In a possible implementation, a second objective function is constructed based on the power gain, frequency regulation benefit and operation loss of the inverter, which is specifically expressed as:

[0085] Analyze the relationship between actual demand and demand price, obtain the price function, and give the power gain of the inverter active power;

[0086] Based on the frequency regulation change of the inverter, the frequency regulation benefit of the inverter is given;

[0087] Combined with the operating loss, the second objective function is constructed, which is specifically expressed as:

[0088] MIN J(P i ,P -i )=H(P i ,P -i )+B(P i )-L(P i )

[0089] Among them, MIN J(P i ,P -i ) is the second objective function, H(P i ,P -i ) is the power gain of the active power of the ith inverter, B(P i ) is the frequency regulation benefit of the i-th inverter, L(P i ) is the operating loss of the i-th inverter, P -i is the active power set excluding the output of the i-th inverter.

[0090] In a possible implementation, the mean field game objective function is specifically expressed as:

[0091] J i (P i ,z)=(ξD max -ξz)P i +B(P i )-L(P i )

[0092]

[0093] Among them, J i (P i,z) is the mean field game function, and z is the mean field.

[0094] In a possible implementation, a strategy iteration algorithm is used to solve the mean field game objective function to obtain a power control strategy for an inverter in a distribution network, which specifically includes:

[0095] Initialize power control strategy;

[0096] Iterate the power control strategy based on the policy iteration equation corresponding to the policy iteration algorithm;

[0097] The power control strategy obtained in each iteration is judged to determine the power control strategy of the inverter.

[0098] In a possible implementation, the strategy iteration equation is specifically expressed as:

[0099]

[0100] Where k is the number of iterations and k ≥ 1, P i,k+1 is the active power output of the ith inverter obtained by the k+1th iteration, P i is the active power output by the i-th inverter, is the average active power output of all inverters in the kth iteration, C i is the cost function of the ith inverter, λ i is the demand response sensitivity coefficient of the i-th inverter, P i d is the required active power output of the i-th inverter, is the average active power output of all inverters in the k+1th iteration, and N is the total number of inverters.

[0101] In a possible implementation manner, judging the power control strategy obtained in each iteration to determine the power control strategy of the inverter specifically includes:

[0102] According to the power control strategy P of the kth iteration i,k and the power control strategy P for the k+1th iteration i,k+1 The relationship between the difference and the preset convergence threshold is used to obtain a first judgment result, wherein k≥1;

[0103] According to the first judgment result, the k+2th iteration or judgment P i,k+1 Whether it is within the preset active power range, the power control strategy of the inverter is obtained.

[0104] The distribution network coordinated control method and device based on mean field game provided by the present invention have at least the following beneficial effects:

[0105] By constructing an aggregation model of multiple inverters, the mutual influence between inverters is taken into account. Based on the supply and demand balance of the distribution network and the power gain, frequency regulation benefit and operating loss of the inverter, and combined with the mean field theory, the mean field game objective function is obtained. On the basis of considering the economic benefits generated by the inverter, the power control strategy of the inverter is solved to maximize the economic benefits of the power grid while ensuring the stable operation of the power grid when the inverter is connected to the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 A flow chart of a distribution network coordinated control method based on mean field game provided in an embodiment of the present invention;

[0107] Figure 2 A flowchart of constructing a second objective function provided by an embodiment of the present invention;

[0108] Figure 3 A flowchart of solving a power control strategy provided by an embodiment of the present invention;

[0109] Figure 4 A flow chart of a strategy iteration algorithm provided by an embodiment of the present invention;

[0110] Figure 5 A structural block diagram of a distribution network coordinated control device based on mean field game provided in an embodiment of the present invention.

[0111] Among them, 201, model construction module; 202, constraint determination module; 203, first function construction module; 204, second function construction module; 205, objective function construction module; 206, strategy iteration module. DETAILED DESCRIPTION

[0112] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0113] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0114] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0115] The research on the coordinated control strategy of dual inverters in microgrids is booming, mainly including the following:

[0116] Droop control strategy: The stable operation of the system is achieved through the relationship between the output power of the inverter and the voltage and frequency. In the microgrid, the main function of the energy storage system is to compensate for the instability and intermittency of renewable energy generation, match the power difference between power generation and load, and improve the stability of the voltage and frequency of the microgrid. The droop control strategy achieves output active and reactive power decoupling through the inductive reactance of the inverter filter, and uses the output active and reactive currents to droop and adjust the output frequency and voltage amplitude of the inverter, respectively, so as to achieve inverter grid-connected and isolated grid parallel operation.

[0117] Virtual impedance technology: used to adjust the resistance-sensitivity characteristics of the inverter output impedance and solve the control coupling phenomenon. By introducing a variable virtual impedance strategy, the voltage drop problem caused by the fixed value virtual impedance is optimized, and the stability and response speed of the system are improved.

[0118] Dual-mode switching: In order to ensure uninterrupted power supply to the load, the dual-mode inverter needs to switch seamlessly between the two working modes of grid-connected and isolated grid. This requires the inverter to avoid sudden changes in load voltage during the mode conversion process, which can be achieved by reasonably controlling the grid-connected switch moment and the controller conversion moment.

[0119] Distributed energy resource management system (DERMS): DERMS can monitor, control and optimize distributed energy resources, monitor energy production and consumption in real time, and achieve efficient, reliable and sustainable operation of the energy system through remote control and energy optimization algorithms.

[0120] A distributed control scheme for voltage regulation based on range consistency has been proposed, using power coupling and extended state observers to coordinate the different internal dynamics and control structures of the inverters. By designing the same distributed controller, the voltage of different inverters can be regulated. A distributed secondary control scheme based on the leader-follower consensus framework is proposed, where GFM acts as the leader and directly controls the frequency and voltage, and GFL acts as the follower and measures the frequency and voltage for power modulation. This coordinated control helps to reasonably allocate power among all inverters and ensures frequency / voltage recovery after system disturbances to achieve accurate power sharing between GFM and GFL inverters. Consensus control is also the most widely used cooperative control strategy for GFM and GFL inverters. In view of the problem that droop control under nonlinear loads will lead to steady-state voltage / frequency deviation, a unified droop-free distributed secondary control framework for GFM and GFL inverters is proposed. The goal of this scheme is to achieve zero deviation in frequency and voltage, and to achieve proportional sharing of active and reactive power among all inverters, and to design the control link according to the partial derivative of the objective function.

[0121] However, these methods do not adequately consider the situation where a large number of inverters are connected to the distribution network, and there is still potential for research on approximate optimal control strategies for a large number of inverters in the distribution network. In addition, the relevant technologies do not consider the impact of inverters on the economic efficiency of the grid when they are connected to the grid.

[0122] After a large number of inverters are connected to the grid, system-level stable control becomes more challenging due to the diversity of inverter structures. Among the many optimal control strategies for power systems, mean-field games (MFG) are considered to be the control strategy for exploring large-scale networked inverter systems. MFG studies the existence of approximate Nash equilibrium (NE) and the corresponding individual strategies in games involving a large number of agents. As the number of agents increases, the influence and importance of each agent decreases, paving the way for the application of mean field theory. The mean field (MF) can replace the influence of all other agents on a given agent, and each agent updates the control strategy according to the MF. The control strategy reflects the best response of each agent to the behavior of a large number of agents, thereby achieving a Nash equilibrium. Nash equilibrium refers to a strategy combination under which any player will not increase his or her own benefits if he or she unilaterally changes his or her strategy (the strategies of other players remain unchanged).

[0123] like Figure 1 As shown, an embodiment of the present invention provides a distribution network coordinated control method based on mean field game, and the specific steps of the method are as follows:

[0124] S101: Provide power constraints of inverters in the distribution network.

[0125] P i ∈[P i,min ,P i,max ]

[0126] P i,min =min((P i +ΔP i ),P N,i )

[0127] P i,max =max((P i -ΔP i ),0)

[0128]

[0129] Among them, P i is the active power output by the ith inverter, P i,min is the minimum value of the active power output by the ith inverter, P i,max is the maximum value of the active power output by the i-th inverter, P N,i is the rated active power of the i-th inverter, ΔP i is the maximum value of the active power output by the ith inverter within the frequency fluctuation range allowed by the power system, Δf max is the maximum frequency fluctuation within the frequency fluctuation range allowed by the power system, m p,i is the first droop coefficient of the i-th inverter.

[0130] S102: Based on the inverter model, construct an aggregation model corresponding to multiple inverters in the distribution network.

[0131] Specifically, a first inverter model of the relationship between the frequency of the inverter controlled by the first droop coefficient and the output active power is given, which is specifically expressed as:

[0132] P i =P i,ref -m p,i (f i -f i,ref )

[0133] Among them, P i is the active power output by the ith inverter, P i,ref is the active power setting value of the i-th inverter, m p,i is the first droop coefficient of the i-th inverter, f i is the frequency of the ith inverter, f i,ref is the rated frequency of the i-th inverter;

[0134] A second inverter model showing the relationship between the voltage amplitude of the inverter controlled by the second droop coefficient and the output reactive power is given, which is specifically expressed as:

[0135] Q i =Q i,ref -m q,i (V i -V i,ref )

[0136] Among them, Q i is the reactive power output by the ith inverter, Q i,ref is the reactive power setting value of the i-th inverter, m q,i is the second droop coefficient of the ith inverter, V i is the actual voltage of the ith inverter, V i,ref is the rated voltage of the i-th inverter;

[0137] The first inverter model and the second inverter model are integrated to form an inverter model;

[0138] Combine the connection relationship of the multiple inverters and obtain a first aggregation model by aggregating the active powers output by the multiple inverters;

[0139] Based on the first droop coefficients of the multiple inverters, analyzing and obtaining a first combined droop coefficient after the multiple inverters are connected;

[0140] In combination with the first aggregation model and the first combined droop coefficient, an aggregation model corresponding to multiple inverters in the distribution network is obtained.

[0141] Aggregation model, specifically expressed as:

[0142]

[0143] Where N is the number of inverters, P a is the active power output by multiple inverters combined, m p,a It is the first combined droop coefficient after multiple inverters are combined.

[0144] In a specific implementation, considering that multiple inverters can be connected in parallel to the same common connection point, it is necessary to derive an aggregation model of multiple inverters, that is, the voltage of N inverters connected in parallel at the common connection point is constant, and the current is the sum of the outputs of the N inverters.

[0145] In the embodiment provided by the present invention, active frequency control is used as a specific embodiment of the grid-type inverter cooperative control framework. In the grid-type inverter control framework, two parameters of a single inverter are required, including the active power output value P i and the first droop coefficient m of active power-frequency p,i .

[0146] In the embodiment provided by the present invention, a group of inverters N is introduced, and the power control strategy P i , the second objective function J i , with the goal of achieving optimal control of the inverter active output, a competitive game G = (N, {Y i} i∈N},{ J i} i∈N ), Y i =[P i,min ,P i,max ].

[0147] Understandably, P i is the active power output by the ith inverter. When the active power output by the ith inverter is determined, the power control strategy of the ith inverter is also determined, that is, the active power output by the ith inverter is controlled to be P i .

[0148] S103: Constructing a first objective function according to the supply and demand balance of the distribution network.

[0149] Specifically, the maximum demand and real-time actual demand of the distribution network are obtained, and the first objective function is given in combination with the demand price. The first objective function is specifically expressed as:

[0150] MIN PR=(D max -D)-λ(r pice -r price0 ) ξ

[0151] Among them, MIN PR is the first objective function, ξ is the demand elasticity coefficient, D max is the maximum demand, r price is the actual demand price, r price0 is the benchmark demand price, λ is the demand response sensitivity coefficient, and D is the actual demand.

[0152] In a specific implementation, the optimal demand is the minimum value of PR. Under the condition of satisfying the supply-demand balance constraint, the influence of the demand price and the demand elasticity coefficient on the actual demand is considered at the same time.

[0153] S104: Constructing a second objective function based on the power gain, frequency regulation benefit and operation loss of the inverter.

[0154] Specifically, refer to Figure 2, analyze the relationship between actual demand and demand price, obtain the price function, and give the power gain of the inverter active power. Based on the frequency regulation change of the inverter, the frequency regulation benefit of the inverter is given. Combined with the operating loss, the second objective function is constructed, which is specifically expressed as:

[0155] MIN J(P i ,P -i )=H(P i ,P -i )+B(P i )-L(P i )

[0156] Among them, MIN J(P i ,P -i ) is the second objective function, H(P i ,P -i ) is the power gain of the active power of the ith inverter, B(P i ) is the frequency regulation benefit of the i-th inverter, L(P i ) is the operating loss of the i-th inverter, P -i is the active power set excluding the output of the i-th inverter.

[0157] Furthermore, from the supply price function, it can be seen that the market price reflects the control strategy of each inverter. Therefore, the gain of active power provided by the inverter can be expressed as:

[0158] H(P i ,P -i )=r price (P i ,P -i )×P i

[0159] It is important to understand that as traditional synchronous generators are gradually replaced by renewable energy generators, the problem of insufficient frequency response will become more serious in future power systems. To solve this problem, a market mechanism should be designed to alleviate the shortage of frequency regulatory resources. Frequency regulation services are provided by adding each inverter to the ancillary service market. The ancillary service market is a market in the power market where ancillary services required to ensure safe and stable operation of the system are traded. Ancillary services include: active reserve services for sudden failures of the power system, frequency regulation services for maintaining system frequency stability, reactive power support and voltage control services for maintaining the voltage of each node of the system, and black start services after a major blackout.

[0160] Furthermore, each inverter is added to the ancillary service market to provide frequency regulation services. The income generated by each inverter, namely the frequency regulation income, is specifically expressed as:

[0161]

[0162] Wherein, χ is the quotation corresponding to the first droop coefficient of each unit i-th inverter.

[0163] In addition, the inverter will generate power loss during operation. The cost of inverter operation loss is specifically expressed as:

[0164]

[0165] Among them, α i , β i and γ i are the power loss coefficients of the i-th inverter, and α i >0, β i >0,γ i >0.

[0166] The second objective function J i It aims to maximize the benefit of the ith inverter while satisfying the power demand and various operating constraints. Therefore, the power control strategy for the optimal active power of the ith inverter considering the frequency constraint can be expressed as:

[0167]

[0168] Among them, the argmax(f(x)) function is used to find the independent variable when f(x) reaches the maximum value, that is, argmaxJ i (P i ) means that when the second objective function J i The power control strategy P when taking the maximum value i The value of .

[0169] S105: Combining the mean field theory, the first objective function and the second objective function, a mean field game objective function is obtained.

[0170] Specifically, the mean field game objective function is expressed as:

[0171] J i (P i ,z)=(ξD max -ξz)P i +B(P i )-L(P i )

[0172]

[0173] Among them, J i (P i ,z) is the mean field game function, and z is the mean field.

[0174] It should be understood that the second objective function is usually referred to as an N-person game, and the grid-connected inverter has the ability to control the number of inverters to N→∞. In order to apply the N-person game scenario to the grid-connected inverter, it is necessary to assume that individuals are indistinguishable, that is, all inverters in the same group have similar power control capabilities and are mathematically identical. The mean field is then expressed as: Substituting the mean field into the second objective function, we get the mean field game objective function:

[0175] Based on mean field game theory, the power control strategy of the optimal active power of the ith inverter considering frequency constraints is adjusted as follows:

[0176] S106: Using a strategy iteration algorithm to solve the mean field game objective function, and obtain a power control strategy for the inverter in the distribution network.

[0177] Specifically, refer to Figure 3 , initialize the power control strategy. Iterate the power control strategy based on the strategy iteration equation corresponding to the strategy iteration algorithm. Judge the power control strategy obtained in each iteration to determine the power control strategy of the inverter.

[0178] Furthermore, the strategy iteration equation is specifically expressed as:

[0179]

[0180] Where k is the number of iterations and k ≥ 1, P i,k+1 is the active power output of the ith inverter obtained by the k+1th iteration, P i is the active power output by the i-th inverter, is the average active power output of all inverters in the kth iteration, C i is the cost function of the ith inverter, λ i is the demand response sensitivity coefficient of the i-th inverter, P i d is the required active power output of the i-th inverter, is the average active power output of all inverters in the k+1th iteration, and N is the total number of inverters.

[0181] It should be understood that the above supply price function shows that the average field game objective function of the ith inverter is coupled through the supply price affected by the power control strategies of all inverters. Therefore, from the perspective of game theory, each inverter can determine its optimal response strategy to the power control strategies of other inverters. Then, the average field point of the game is obtained by calculating the optimal power control strategy of all inverters.

[0182] Further, according to the power control strategy P of the kth iteration i,k and the power control strategy P for the k+1th iteration i,k+1 The relationship between the difference of the value of the convergence threshold and the preset convergence threshold is obtained, where k≥1. According to the first judgment result, the k+2th iteration or judgment P is performed. i,k+1 Whether it is within the preset active power range, the power control strategy of the inverter is obtained.

[0183] In the embodiment provided by the present invention, a decentralized strategy iteration algorithm based on memory augmented neural network (MANN) iteration is proposed to solve the mean field game objective function and obtain the power control strategy of the inverter in the distribution network.

[0184] In a specific example, referring to Figure 4 First, the power control strategy is initialized. The power control strategy, that is, the initial value of the active power output of the i-th inverter is obtained by the optimal power flow calculation. Then the power control strategy is iterated using the strategy iteration equation. According to the power control strategy P of the k-th iteration i,k and the power control strategy P for the k+1th iteration i,k+1 The relationship between the difference of and the preset convergence threshold is used to obtain a first judgment result. When the first judgment result is that the difference is greater than or equal to the preset convergence threshold, it indicates that the power control strategy P i,k+1 The convergence condition is not met and the k+2th iteration is required. When the first judgment result is that the difference is less than the preset convergence threshold, it means that the power control strategy P i,k+1 To meet the convergence conditions, we need to further judge P i,k+1 Is it within the preset active power range? If so, the power control strategy of the inverter is obtained. Otherwise, the power control strategy P is changed. i,k+1 Set to be equal to the maximum or minimum value in the preset active power range, and continue to iterate until the power control strategy meets the convergence condition. The preset convergence threshold is used to control the iterative convergence error of the power control strategy. During the iteration process, the power control strategy of each inverter will be updated as the mean field is updated. After the iteration is completed, each inverter has the optimal power control strategy under the control of the mean field game theory.

[0185] In each iteration, the inverter responds optimally to the broadcasted intermediate frequency signal and updates the power control strategy. The broadcast center then updates the mean field entries based on the inverter's updated power control strategy and broadcasts the updated mean field. This information must continue to interact and iterate until convergence is achieved to obtain a set of network element solutions, namely the inverter's power control strategy.

[0186] The above-mentioned optimal power flow (OPF) refers to the power flow distribution that can meet all operating constraints and make a certain performance indicator of the system (such as power generation cost or network loss) reach the optimal value by adjusting the available control variables (such as generator output power, adjustable transformer taps, etc.) when the structural parameters and load conditions of the power system are given.

[0187] Optimal power flow refers to the optimization process of adjusting the parameters of various control devices in the power system from the perspective of optimizing the operation of the power system, and minimizing the objective function under the constraints of satisfying the normal power balance of the nodes and various safety indicators. Usually, the optimization power flow is divided into active power optimization and reactive power optimization. The objective function of active power optimization is the power generation cost or power consumption, and the objective function of reactive power optimization is the network loss of the entire network. Since optimal power flow is an analysis method that considers both the security and economy of the network, it is widely used in the safe operation of the power system, economic dispatch, power grid planning, reliability analysis of complex power systems, and economic control of transmission congestion.

[0188] A distribution network coordinated control method based on mean field game provided by an embodiment of the present invention is based on mean field game theory and creates a highly competitive game environment for inverters. The interaction between inverters is approximated by the influence caused by the power control strategy of the average inverter. The inverter independently establishes a power control strategy based on its own control output and the broadcast intermediate frequency information. The decision-making process is decentralized, and the shared information mean field project promotes consistency between inverters. In addition, since there is no information exchange between inverters, the communication burden can be greatly reduced. In the embodiment provided by the present invention, all meshing inverters are allowed to participate in frequency regulation, and frequency regulation and active power distribution can be achieved simultaneously.

[0189] Reference Figure 5 The embodiment of the present invention provides a distribution network coordination control device based on mean field game, comprising:

[0190] A model building module 201 is used to build an aggregation model corresponding to multiple inverters in the distribution network based on the inverter model;

[0191] The constraint determination module 202 is used to provide a power constraint of each inverter in the distribution network;

[0192] A first function construction module 203, configured to construct a first objective function according to the supply and demand balance of the distribution network;

[0193] A second function building module 204, configured to build a second objective function based on the power gain, frequency regulation benefit and operation loss of the inverter;

[0194] The objective function construction module 205 is used to combine the mean field theory, the first objective function and the second objective function to obtain the mean field game objective function;

[0195] The strategy iteration module 206 is used to adopt a strategy iteration algorithm to solve the mean field game objective function to obtain a power control strategy for the inverter in the distribution network.

[0196] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0197] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0198] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0199] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0200] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0201] The units involved in the embodiments of the present disclosure may be implemented by software or hardware, wherein the name of a unit does not limit the unit itself in some cases.

[0202] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A distribution network coordinated control method based on mean field game, characterized in that: include: Give the power constraints of the inverters in the distribution network; Based on the inverter model, an aggregation model corresponding to multiple inverters in the distribution network is constructed; According to the supply and demand balance of the distribution network, a first objective function is constructed; A second objective function is constructed based on the power gain, frequency regulation benefit and operation loss of the inverter; Combining the mean field theory, the first objective function and the second objective function, we get the mean field game objective function; The strategy iteration algorithm is used to solve the mean field game objective function and obtain the power control strategy of the inverter in the distribution network.

2. The coordinated control method for distribution network based on mean field game according to claim 1, characterized in that: The power constraint of the inverter in the distribution network is specifically expressed as: P i ∈[P i,min ,P i,max ] P i,min =min((P i +ΔP i ),P N,i ) P i,max =max((P i -ΔP i ),0) Among them, P i is the active power output by the ith inverter, P i,min is the minimum value of the active power output by the ith inverter, P i,max is the maximum value of the active power output by the i-th inverter, P N,i is the rated active power of the i-th inverter, ΔP i is the maximum value of the active power output by the ith inverter within the frequency fluctuation range allowed by the power system, Δf max is the maximum frequency fluctuation within the frequency fluctuation range allowed by the power system, m p,i is the first droop coefficient of the i-th inverter.

3. The coordinated control method for distribution network based on mean field game according to claim 1, characterized in that: Based on the inverter model, an aggregation model corresponding to multiple inverters in the distribution network is constructed, including: A first inverter model showing the relationship between the frequency of the inverter controlled by the first droop coefficient and the output active power is given, which is specifically expressed as: P i =P i,ref -m p,i (f i -f i,ref ) Among them, P i is the active power output by the ith inverter, P i,ref is the active power setting value of the i-th inverter, m p,i is the first droop coefficient of the i-th inverter, f i is the frequency of the ith inverter, f i,ref is the rated frequency of the i-th inverter; The second inverter model showing the relationship between the voltage amplitude of the inverter controlled by the second droop coefficient and the output reactive power is given, which is specifically expressed as: Q i =Q i,ref -m q,i (V i -V i,ref ) Among them, Q i is the reactive power output by the i-th inverter, Q i,ref is the reactive power setting value of the i-th inverter, m q,i is the second droop coefficient of the ith inverter, V i is the actual voltage of the ith inverter, V i,ref is the rated voltage of the i-th inverter; The first inverter model and the second inverter model are integrated to form an inverter model; Combine the connection relationship of the multiple inverters and obtain a first aggregation model by aggregating the active powers output by the multiple inverters; Based on the first droop coefficients of the multiple inverters, analyzing and obtaining a first combined droop coefficient after the multiple inverters are connected; In combination with the first aggregation model and the first combined droop coefficient, an aggregation model corresponding to multiple inverters in the distribution network is obtained.

4. The coordinated control method for distribution network based on mean field game according to claim 1, characterized in that: According to the supply and demand balance of the distribution network, the first objective function is constructed, which specifically includes: The maximum demand and real-time actual demand of the distribution network are obtained, and the first objective function is given in combination with the demand price. The first objective function is specifically expressed as: MIN PR=(D max -D)-λ(r price -r price0 ) ξ Among them, MIN PR is the first objective function, ξ is the demand elasticity coefficient, D max is the maximum demand, r price is the actual demand price, r price0 is the benchmark demand price, λ is the demand response sensitivity coefficient, and D is the actual demand.

5. The coordinated control method for distribution network based on mean field game according to claim 4, characterized in that: Based on the power gain, frequency regulation benefit and operation loss of the inverter, the second objective function is constructed, which is specifically expressed as: Analyze the relationship between actual demand and demand price, obtain the price function, and give the power gain of the inverter active power; Based on the frequency regulation change of the inverter, the frequency regulation benefit of the inverter is given; Combined with the operating loss, the second objective function is constructed, which is specifically expressed as: MIN J(P i ,P -i )=H(P i ,P -i )+B(P i )-L(P i ) Among them, MIN J(P i ,P -i ) is the second objective function, H(P i ,P -i ) is the power gain of the active power of the ith inverter, B(P i ) is the frequency regulation benefit of the i-th inverter, L(P i ) is the operating loss of the i-th inverter, P -i is the active power set excluding the output of the i-th inverter.

6. The coordinated control method for distribution network based on mean field game according to claim 5, characterized in that: The mean field game objective function is specifically expressed as: J i (P i ,z)=(ξD max -ξz)P i +B(P i )-L(P i ) Among them, J i (P i ,z) is the mean field game function, and z is the mean field.

7. The coordinated control method for distribution network based on mean field game according to claim 1, characterized in that: The strategy iteration algorithm is used to solve the mean field game objective function and obtain the power control strategy of the inverter in the distribution network, which includes: Initialize power control strategy; Iterate the power control strategy based on the strategy iteration equation corresponding to the strategy iteration algorithm; The power control strategy obtained in each iteration is judged to determine the power control strategy of the inverter.

8. The coordinated control method for distribution network based on mean field game according to claim 7, characterized in that: The strategy iteration equation is specifically expressed as: Where k is the number of iterations and k ≥ 1, P i,k+1 is the active power output of the ith inverter obtained by the k+1th iteration, P i is the active power output by the i-th inverter, is the average active power output of all inverters in the kth iteration, C i is the cost function of the ith inverter, λ i is the demand response sensitivity coefficient of the i-th inverter, P i d is the required active power output of the i-th inverter, is the average active power output of all inverters in the k+1th iteration, and N is the total number of inverters.

9. The coordinated control method for distribution network based on mean field game according to claim 7, characterized in that: The power control strategy obtained in each iteration is judged to determine the power control strategy of the inverter, which specifically includes: According to the power control strategy P of the kth iteration i,k and the power control strategy P for the k+1th iteration i,k+1 The relationship between the difference and the preset convergence threshold is used to obtain a first judgment result, wherein k≥1; According to the first judgment result, the k+2th iteration or judgment P i,k+1 Whether it is within the preset active power range, the power control strategy of the inverter is obtained.

10. A distribution network coordination control device based on mean field game, characterized in that: A distribution network coordinated control method based on mean field game as claimed in any one of claims 1 to 9 is adopted, comprising: A model building module, used to build an aggregation model corresponding to multiple inverters in the distribution network based on the inverter model; A constraint determination module, used for providing power constraints of inverters in the distribution network; A first function building module, used to build a first objective function according to the supply and demand balance of the distribution network; A second function building module is used to build a second objective function based on the power gain, frequency regulation benefit and operation loss of the inverter; An objective function construction module is used to combine the mean field theory, the first objective function and the second objective function to obtain the mean field game objective function; The strategy iteration module is used to solve the mean field game objective function by using the strategy iteration algorithm to obtain the power control strategy of the inverter in the distribution network.

Images