CN108648024A - A kind of power distribution network distributed generation resource deploying node computational methods - Google Patents

Abstract

The invention discloses a kind of power distribution network distributed generation resource deploying node computational methods, using in cooperative game for handle unlimited more player's Game with Coalitions method --- Aumann Shapley values methods share network loss and polluted gas discharge capacity, propose a kind of computational methods being used to calculate DLMP values in power distribution network based on this.Fair determining each DG of the invention is contributed made by network loss and disposal of pollutants reducing, positive economic incentives signal can be provided for it, distribution company can adjust weight factor according to the priority of network loss and discharge capacity simultaneously, have the different types of DG of the excitation stressed to reduce network loss or discharge capacity.

Description

Marginal electricity price calculation method for distributed power supply nodes of power distribution network

Technical Field

The invention discloses a marginal electricity price calculation method for a distributed power supply node of a power distribution network, and belongs to the field of power distribution network control of a power system.

Background

A Distributed Generation (DG) is connected to a power distribution network, the injected active power and reactive power can change the size and flow direction of the power flow of the power distribution network, and influence is brought to the voltage quality, network loss and reliability of the power distribution network. In the power market environment, in order to maintain safe and reliable operation of a power distribution network, a distribution network company can indirectly manage and control the output of a grid-connected DG in a power rate incentive mode, and therefore a node marginal power rate (LMP) which is the most effective power rate mechanism of the current power transmission network can be introduced. In recent years, scholars at home and abroad carry out systematic research on DLMP (distribution network node marginal price), but the problem of computing combination explosion caused by the increase of the number of DGs connected in a grid is not solved, and zero sale margin cannot be realized.

Disclosure of Invention

The purpose of the invention is as follows: the invention provides a marginal price calculation method for a distributed power supply node of a power distribution network, which reduces the network loss and pollution emission of the power distribution network under the condition that a plurality of DGs are connected into the power distribution network.

The technical scheme is as follows: the technical scheme adopted by the invention is a method for calculating the marginal electricity price of a distributed power supply node of a power distribution network, which comprises the following steps:

1) setting DLMPs at all DG nodes to a unified market clearing price rho^a；

2) Cost function according to DGObtaining the active power P of each DG_i；

3) Calculating the grid loss P of the power distribution network after DG grid connection_lossAnd pollution emission E;

4) calculating the network loss and emission reduction apportionment amount of each DG based on an A-S value method;

5) calculating the active DLMP correction quantity and the reactive DLMP correction quantity of the DG node;

6) correcting original active and reactive DLMP values of the DG nodes;

7) judging whether a termination condition is met, and if so, ending iteration; if not, returning to the step 2) to repeatedly modify the active DLMP and the reactive DLMP of each DG.

The cost function of each DG in the step 2)Obtain an active power output P_iThe calculation formula of (2) is as follows:

wherein,denotes DG at the j +1 th iteration_iThe active power of (1) and (2), wherein a_i,b_i,c_iAre respectively DG_iThe cost function parameter of (a) is,is DG_iDLMP value obtained at the j-th iteration.

The network loss P of the power distribution network in the step 3)_lossAnd the amount of emission of pollutants E was calculated as follows:

in the formulae (3) and (4), N_brIs the number of branches, N_DGAs a result of the total number of DG,and EF_gridAre respectively DG_iAnd pollution emission factor (kg/kW) of the power plant.

The step 4) is implemented with DG in the power grid_iThe generated output at a certain moment isThen the DG_iThe net loss/emission reduction caused isFor DG_iThe output is divided infinitely, if DG is at this time_iThe output is increased by an infinitesimal small amount Delta b_i(Δb_i→ 0), then Δ b_iThe marginal contribution to reducing grid loss/emissions is:

when DG is reached_iGenerated output b_iWhen increasing from 0 to its maximum, DG is available_iNet loss/emission reduction apportionment amount psi_i：

In the formula (6), λ is an integral variable, b_iIs DG_iPower generation output of f^k(. cndot.) is the loss/emission reduction of the distribution grid at a given lambda value.

The active DLMP correction quantity and the reactive DLMP correction quantity of the DG node in the step 5) are as follows:

in the above formula, the first and second carbon atoms are,andare respectively DG_iActive and reactive DLMP corrections;andare respectively DG_iThe corresponding weights of the network loss and emission reduction correction components of the active DLMP are respectively omega₁And ω₂And satisfy omega₁+ω₂＝1；Andare respectively DG_iThe network loss and emission reduction share at the jth iteration;and E^jRespectively the network loss and the emission of the j iteration after the DG is accessed,and E^baseThe loss and the emission when the DG is not connected are respectively.

DG is processed in the step 6) according to the following formula_iThe original active and reactive DLMP values are corrected:

in the above equation, the reactive power rates at the balance nodes are less than 1% of the active power rates and can be ignored, so ρ^r≈0。

Has the advantages that: according to the invention, the network loss and the pollution emission reduction are shared by an A-S value method, the contribution of each DG to the reduction of the network loss and the pollution emission can be fairly determined, and a positive economic excitation signal can be provided for the DG; meanwhile, the network loss and pollution emission reduction of the power distribution network can bring extra benefits to the distribution network company, and the extra benefits are completely distributed to each DG by the A-S value method, namely, zero sale surplus of the distribution scheme is realized; due to the incentive effect of electricity price, each DG adjusts the power generation output of the DG to increase the income as much as possible, thereby being beneficial to the management and control of the DG by a distribution network company; the distribution network company can adjust the weight factor according to the priority of the network loss and the emission, and has a strong emphasis on exciting DGs of different types to reduce the network loss or the emission.

Drawings

FIG. 1 is a flow chart of the operation of the present invention;

FIG. 2 is a diagram of an IEEE33 node power distribution system topology;

FIG. 3 is a diagram of an IEEE69 node power distribution system topology;

FIG. 4 is a graph of network loss results obtained by the conventional method and the method of the present invention for different market clearing prices based on IEEE69 node power distribution system solutions;

fig. 5 is a graph of active DLMP results at DG nodes obtained by the conventional method and the method of the present invention when the market clearing price solved by the IEEE69 node power distribution system is ρ ═ 23($/MW) and ρ ═ 26($/MW), respectively;

FIG. 6 is a graph of sales margins obtained by the conventional method and the method of the present invention for different market clearing prices solved based on an IEEE69 node power distribution system.

Detailed Description

The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.

As shown in fig. 1, the present invention comprises the steps of:

1) setting DLMPs at all DG nodes to be uniform market clearing price rho^a(Balancing the price of electricity at a node), i.e. (π^a)_i＝ρ^a，(π^a)_iRepresenting an active DLMP value at the node i to obtain the initial running states of all DGs;

2) obtaining the active output P of each DG according to the cost function of the DG_iThe calculation formula is as shown in formula (12):

P_i ^j+1denotes DG at the j +1 th iteration_iThe active power of (11) and (12), wherein a_i,b_i,c_iAre respectively DG_iThe cost function parameter of (a) is,is DG_iAt j iterationThe DLMP value obtained;

3) calculating the grid loss P of the power distribution network after DG grid connection_lossAnd emission of pollutants E:

in formulae (13) and (14), N_brIs the number of branches, N_DGAs a result of the total number of DG,and EF_gridAre respectively DG_iAnd pollution emission factor (kg/kW) of the power plant;

4) calculating the sharing amount of the loss and emission reduction of each DG based on an A-S value method, and comparing the sharing amount with the sharing result of the Shapley value:

shapley value method: calculating the network loss/emission reduction apportionment amount of each DG in the power distribution network based on a Shapley value method, wherein the calculation formula is as follows:

in the formulae (15) and (16),representation is apportioned to the ith distributed power supply DG_iLoss (l)/emission (e) reduction of (a), S means comprising a plurality of DGs_iThe | S | refers to the DG number in the alliance S, and n refers to the total DG number participating in the distribution network loss and emission reduction amount allocation，[v(S)-v(S-{i})]Calculate the DG due to the ith distributed generator_iThe added value of the profit brought to the alliance by joining the alliance S, namely the marginal profit of the alliance S, W (S) represents a weight value which represents the ith distributed power supply DG_iThe share of the marginal profit of federation S that should be allocated, n! Representing the ordering of all possible DG additions to a large federation (including all DG).

A-S value method (Aumann-Shapley value method):

the basic idea of the A-S value method is to divide each person in the station into an infinite number of persons in the station, and then calculate the apportionment of the persons in each infinite small station by adopting the Shapley value method. The essence of the method is to calculate the average value of the marginal contribution of the middle-aged people in each office to each cooperative alliance, so that the influence of the alliance adding sequence of the middle-aged people in each office on the result can be ignored, and the method has economic consistency and equality. Fair and reasonable allocation can be achieved.

Apparently, the A-S value method will be much more computationally expensive than the Shapley value method due to the greatly increased number of leagues. However, the a-S value method divides each person in each office indefinitely, so the apportionment results are independent of the order in which each person in each office joins the coalition, and the apportionment amount can be calculated by an analytical method.

Suppose DG in a distribution network_iThe generated output at a certain moment isThen the DG_iThe net loss/emission reduction caused isFor DG_iThe output is divided infinitely, if DG is at this time_iThe output is increased by an infinitesimal small amount Delta b_i(Δb_i→ 0), then Δ b_iThe marginal contribution to reducing grid loss/emissions is:

when DG is reached_iGenerated output b_iWhen increasing from 0 to its maximum, DG is available_iNet loss/emission reduction apportionment amount psi_i：

In the formula (18), λ is an integral variable, b_iIs DG_iPower generation output of f^k(. cndot.) is the loss and emission reduction of the distribution grid at a given lambda value.

Due to the complexity of the distribution network loss and emission problems, it is difficult to express the loss/emission split into the canonical Aumann-sharey form. For this purpose, we use the simplified A-S value method to convert DG_iPerforming finite division, and respectively calculating DG of a small part of the last access system in the S alliance_iThe network loss and emission reduction brought to the system are summed to obtain the DG_iThe result is split.

5) And calculating the active DLMP correction quantity and the reactive DLMP correction quantity of the DG node. The active DLMP correction consists of two components: a grid loss reduction component and an emission reduction component.

In the formulae (19), (20) and (21),andare respectively DG_iActive and reactive DLMP corrections;andare respectively DG_iThe corresponding weights of the network loss and emission reduction correction components of the active DLMP are respectively omega₁And ω₂And satisfy omega₁+ω₂＝1；Andare respectively DG_iThe network loss and emission reduction share at the jth iteration;and E^jRespectively the network loss and the emission of the j iteration after the DG is accessed,and E^baseThe loss and the emission when the DG is not connected are respectively.

6) For DG_iThe original active and reactive DLMP values are corrected:

in equation (22), the reactive power rate at the balance node is less than 1% of the active power rate and can be ignored, so ρ^r≈0。

7) Judging whether a termination condition is met, and if so, ending iteration; if not, returning to the step 2) to repeatedly modify the active DLMP and the reactive DLMP of each DG:

wherein ε is a minimum value. When the formula (23) is satisfied, it means that the DG is increased_iThe active power output of the transformer cannot reduce the network loss and the pollution emission, namely the effect of electrovalence excitation is lost, so that the iteration can be terminated.

The revenue for a distribution network company can be expressed as:

in formula (24), benefit^jRepresents the benefit gained by the distribution network company at the j iteration, Demand represents the total load of the distribution system, gamma_eIs the unit pollutant emission cost ($/kg), P_i ^jAndrespectively representing the active and reactive power, pi, at the jth iteration^c,π^a,π^rRespectively corresponding to a user DLMP, an active DLMP and a reactive DLMP. When the active and reactive DLMPs of each DG are determined by the uniform electricity price method, the two middle terms of equation (24) are both 0, i.e.:

in the formula (25), γ_eAnd is the unit pollutant emission cost ($/kg).

The formula (24) and the formula (25) are subtracted to obtain the extra income of the distribution network company, namely the sales surplus is:

in the formulae (27) and (28), Δ benefit^jIndicates the extra revenue, MS, of the distribution network company at the jth iteration_lAnd MS_eRespectively, additional revenue for the distribution company due to network loss and emission reductions.

The invention respectively verifies the effectiveness and the simplicity of the A-S value method for apportionment compared with the Shapley value method and the feasibility and the rationality of the DLMP calculation model through two embodiments.

EXAMPLE 1

Fig. 2 shows an IEEE33 node power distribution network, which contains 33 nodes and 3 branch lines. DGs are connected at nodes 8, 12 and 23 respectively, the emission factors of the DGs to be connected are shown in Table 1, and the power factors are all lagged by 0.9.

TABLE 1

And respectively adopting Shapley value and A-S value methods to share the loss of the network and the emission reduction of the IEEE-33 node system. Assuming that the active power of all DGs is 500kW, three DGs can be regarded as three local people, denoted as N ═ 1,2,3, and the total local people set and each non-empty subset form a coalition, and the network loss and emission reduction of each coalition are shown in table 2.

TABLE 2

The share of the loss and emission reduction of three DG were calculated from the sharley values and the results are shown in table 3.

When the A-S value method is adopted to calculate the apportionment amount, because of the complexity of the distribution network loss and the emission apportionment problem, the simplified A-S value method is adopted to calculate the DG_iPerforming a finite division, i.e. DG_iEqually dividing into n small parts, and then calculating DG of each small part respectively_iThe loss and emission reduction of the system caused by the last access of the system are summed to obtain the DG_iThe result is split. The results of the A-S value method are shown in Table 3.

TABLE 3

Compare the sharey values in table 3 with the split results of the simplified a-S value method: the error of the network loss and emission reduction apportionment amount of the three DGs obtained by simplifying the A-S value method is very small compared with the apportionment result of the Shapley value, so the calculation result of the simplified A-S value method accords with the fairness principle; however, the calculation amount of the sharley value method is increased sharply with the increase of the number of DGs in the power distribution network (for example, when the number of DGs is 10, the number of unions to be calculated is 3628800), and the a-S value method introduces a limit processing and analysis method, so that the problem of combination explosion can be solved well, and the calculation is simpler.

EXAMPLE 2

As shown in fig. 3, the IEEE69 node power distribution network includes three types of DGs, four of each type of DG, and all the DG power factors lag by 0.9; the generation capacity, cost function parameters and emission factor of each DG are shown in tables 4 and 5, respectively.

TABLE 4

TABLE 5

The network loss and emission reduction apportionment amount of each DG calculated based on the A-S value method can obtain the DG active power output, the DLMP value at the DG node and the corresponding network loss and emission amount corresponding to different weight factors under different market clearing prices, and the results are respectively shown in tables 6 and 7.

TABLE 6

TABLE 7

The traditional DLMP calculation model mainly comprises a unified power price method and a marginal grid loss method: DLMPs at all nodes in the unified electricity price method are unified market clearing prices; the DLMP at the node i in the marginal loss method is determined by the marginal contribution of the node to the loss:

equations (29) and (30) are the active and reactive DLMP values, ρ, at node i, respectively^aClearing price for uniform market, P_iAnd Q_iRespectively the active and reactive power at node i.

To verify the superiority of the method of the present invention over other conventional methods, the current ω is calculated₁＝ω₂When the power rate method, the marginal grid loss method and the system grid loss value obtained by the a-S value allocation method are unified under different market prices, and when the market clearing price is ρ 23 and ρ 26($/MW), respectively, the active DLMP value at each DG node is obtained by the three methods (the marginal grid loss method and the a-S value allocation method both adopt the iterative model in the method of the present invention), and the results are respectively shown in fig. 4 and fig. 5.

As can be seen from fig. 4 and 5, compared with the two conventional methods, the DLMP determined by the method of the present invention can provide more power rate excitation for each DG, thereby reducing the system network loss to a greater extent.

FIG. 6 is a calculated sales margin based on the method of the present invention and two conventional methods for different market clearing prices. The DG is incorporated into the power networks and has reduced system's loss of network and emission, has brought extra income for the distribution network system: the part of the income under the unified electricity price method is completely owned by a distribution network company, and the marginal network loss method can only distribute part of the income to DGs, so that the two traditional methods can not realize zero sale surplus; the method of the invention fairly distributes the network loss and the emission reduction to each DG through an A-S value method, namely, completely distributes the extra income of the distribution network system to each DG, thereby realizing the zero sale surplus of the distribution scheme.

Generally, the LMP calculation method for the power distribution network based on the distribution of the loss and the emission reduction by the A-S values can effectively overcome the problem of combined explosion of the traditional Shapley value distribution method, and has certain applicability to the power distribution network with more DGs; the DLMP calculation method not only can reduce the network loss and the emission simultaneously, but also can realize the management and the control of a distribution network company on the DG output through the electrovalence excitation; compared with the traditional method, the method can stimulate the electricity price of each DG to a greater extent, and reduces the system network loss and the emission.

It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modifications, equivalents, improvements and the like which come within the spirit and scope of the invention are included.

Claims (6)

1. A marginal electricity price calculation method for a distributed power supply node of a power distribution network is characterized by comprising the following steps:

1) setting DLMPs at all DG nodes to a unified market clearing price rho^a；

2) Cost function according to DGObtaining the active power P of each DG_i；

3) Calculating the grid loss P of the power distribution network after DG grid connection_lossAnd pollution emission E;

4) calculating the network loss and emission reduction apportionment amount of each DG based on an A-S value method;

5) calculating the active DLMP correction quantity and the reactive DLMP correction quantity of the DG node;

6) correcting original active and reactive DLMP values of the DG nodes;

7) judging whether a termination condition is met, and if so, ending iteration; if not, returning to the step 2) to repeatedly modify the active DLMP and the reactive DLMP of each DG.

2. The method for calculating the marginal electricity price of the distributed power nodes of the power distribution network according to claim 1, wherein the step 2) is performed by using cost functions of all DGsObtain an active power output P_iThe calculation formula of (2) is as follows:

P_i ^j+1denotes DG at the j +1 th iteration_iThe active power of (1) and (2), wherein a_i,b_i,c_iAre respectively DG_iThe cost function parameter of (a) is,is DG_iDLMP value obtained at the j-th iteration.

3. The marginal electricity price calculation method for the distributed power nodes of the power distribution network according to claim 1, wherein the power distribution network loss P in the step 3) is_lossAnd the amount of emission of pollutants E was calculated as follows:

in the formulae (3) and (4), N_brIs the number of branches, N_DGAs a result of the total number of DG,and EF_gridAre respectively DG_iAnd pollution emission factor (kg/kW) of the power plant.

4. The marginal electricity price calculation method for the distributed power supply nodes of the power distribution network according to claim 1, wherein DGs in the power distribution network are arranged in the step 4)_iThe generated output at a certain moment isThen the DG_iThe net loss/emission reduction caused isFor DG_iThe output is divided infinitely, if DG is at this time_iThe output is increased by an infinitesimal small amount Delta b_i(Δb_i→ 0), then Δ b_iThe marginal contribution to reducing grid loss/emissions is:

when DG is reached_iGenerated output b_iWhen increasing from 0 to its maximum, DG is available_iNet loss/emission reduction apportionment amount psi_i：

In the formula (6), λ isIntegral variable, b_iIs DG_iPower generation output of f^k(. cndot.) is the loss/emission reduction of the distribution grid at a given lambda value.

5. The marginal electricity price calculation method for the distributed power supply nodes of the power distribution network according to claim 1, wherein the active and reactive DLMP corrections of the DG nodes in the step 5) are as follows:

in the above formula, the first and second carbon atoms are,andare respectively DG_iActive and reactive DLMP corrections;andare respectively DG_iThe corresponding weights of the network loss and emission reduction correction components of the active DLMP are respectively omega₁And ω₂And satisfy omega₁+ω₂＝1；Andare respectively DG_iThe network loss and emission reduction share at the jth iteration;and E^jRespectively the network loss and the emission of the j iteration after the DG is accessed,and E^baseThe loss and the emission when the DG is not connected are respectively.

6. The method for calculating the marginal electricity price of the distributed power supply node of the power distribution network according to claim 1, wherein the DG is calculated in step 6) according to the following formula_iThe original active and reactive DLMP values are corrected:

in the above equation, the reactive power rates at the balance nodes are less than 1% of the active power rates and can be ignored, so ρ^r≈0。