A method for allocating costs of auxiliary services

Abstract

The present invention discloses an auxiliary service cost allocation method, comprising: S1: obtaining a first auxiliary service cost to be borne by a target participant using the EANS method; S2: obtaining a second auxiliary service cost to be borne by the target participant using the Shapley value method; S3: correcting the second auxiliary service cost using a risk factor to obtain a third auxiliary service cost based on the improved Shapley value method; S4: weighting the first and third auxiliary service costs to obtain a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the second auxiliary service cost; S5: obtaining an auxiliary service cost allocation model using an auxiliary service cost calculation formula based on the first and second weight values; and S6: allocating the target auxiliary service cost using the auxiliary service cost allocation model. The present invention can truly reflect the impact of different risk levels on the allocation of auxiliary service costs.

Description

Technical Field

[0001] The present invention relates to the technical field of auxiliary services, and in particular to a method for allocating auxiliary service costs. Background Art

[0002] Due to the volatility and uncertainty of renewable energy generation, and the current high requirements for renewable energy consumption in China, the grid-connected operation of renewable energy will prompt the power grid to improve its ancillary service capabilities and provide corresponding ancillary services to support the integration of renewable energy. Therefore, the grid's ancillary service capabilities need to increase its investment costs due to the integration of renewable energy. At the same time, the large-scale integration of renewable energy into the grid will lead to an increase in the grid's ancillary service costs. According to the current ancillary service cost compensation mechanism, it would be unfair for the thermal power system to bear the increased overall ancillary service costs. Therefore, a new mechanism for sharing the ancillary service costs brought about by the integration of renewable energy has been urgently needed to incentivize the power grid to provide corresponding ancillary services for the integration of renewable energy.

[0003] Traditional ancillary service cost allocation methods include the EANS ancillary service cost allocation method and the Shapley value ancillary service cost allocation method. The EANS method, also known as the equal allocation of nonseparable costs method, is widely used in practice. A shortcoming of the EANS allocation method is that it fails to take into account alliances other than large alliances and alliances containing (n-1) participants, which also have an impact on the allocation results. The Shapley value allocation method addresses this shortcoming of the EANS allocation method and is widely used in benefit and cost allocation problems. It considers the impact of all possible alliances on the allocation results and allocates the allocation results according to the marginal contributions of the participants to all alliances. However, the Shapley value method assumes that all participants bear equal risk, which is obviously an ideal situation and cannot truly reflect the impact of different risk levels on the allocation of ancillary service costs. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for allocating auxiliary service costs, so as to truly reflect the impact of different risk levels on the allocation of auxiliary service costs.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides a method for allocating auxiliary service costs, comprising:

[0007] S1: Use the EANS method to obtain the first auxiliary service cost that the target participant should bear;

[0008] S2: Use the Shapley value method to obtain the second auxiliary service cost that the target participant should bear;

[0009] S3: Using the risk factor to modify the second ancillary service cost to obtain a third ancillary service cost based on the improved Shapley value method;

[0010] S4: performing weighted allocation on the first auxiliary service cost and the third auxiliary service cost to obtain a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the third auxiliary service cost;

[0011] S5: Obtaining an auxiliary service cost allocation model using an auxiliary service cost calculation formula based on the first weight value and the second weight value;

[0012] S6: Allocate the target ancillary service cost using the ancillary service cost allocation model.

[0013] Optionally, the step S1 includes:

[0014] S11: Obtain the divisible cost of the auxiliary services for the target participant based on the total cost of the auxiliary services borne by all participants N and the total cost of the auxiliary services borne by all participants N1 except the target participant;

[0015] S12: Obtain the total indivisible cost of auxiliary services incurred by all participants based on the total cost of auxiliary services borne by all participants N and the divisible cost of auxiliary services of all target participants;

[0016] S13: Allocate the total indivisible cost of the auxiliary services to each participant to obtain the indivisible cost of the auxiliary services corresponding to the target participant;

[0017] S14: Obtain the first auxiliary service cost that the target participant should bear based on the indivisible auxiliary service cost and the divisible auxiliary service cost corresponding to the target participant.

[0018] Alternatively, the auxiliary service of the target participant i may be divided into costs SC i for:

[0019] SC i =C(N)-C(N1 / i)

[0020] The total non-divisible cost of ancillary services NSC is:

[0021]

[0022] Among them, C(N) represents the total cost of auxiliary services borne by all participants N = {1, 2, …, n}, and C(N1 / i) represents the total cost of auxiliary services borne by all participants N1 = {1, 2, …i-1, i+1, …, n} except the target participant i.

[0023] Optionally, the first auxiliary service cost that the target participant i should bear is for:

[0024]

[0025] Among them, SC i represents the divisible cost of ancillary services for target participant i, n represents the number of all participants, and NSC represents the total indivisible cost of ancillary services.

[0026] Optionally, the second auxiliary service cost h i [C] is:

[0027]

[0028] Where S represents all alliances in which the target participant i participates; s represents the number of participants in the alliance S in which the target participant i participates; C represents the auxiliary service cost function; C(S) represents the auxiliary service cost value that should be shared by the alliance S; C(S-{i}) represents the auxiliary service cost value that should be shared by the alliance S excluding participant i; C(S)-C(S-{i}) represents the marginal contribution of the target participant i to the alliance S in which he participates; q(s) represents the weighting factor for how to evenly distribute the marginal contribution of the auxiliary service cost, that is, the probability of C(S)-C(S-{i}) occurring and n represents the total number of participants, and ! represents the factorial symbol.

[0029] Optionally, the third auxiliary service cost for:

[0030]

[0031] in, represents the Shapley value of ancillary service cost allocation after risk factor correction and h i (C) represents the second ancillary service cost, Δh i [C] represents the auxiliary service cost correction amount that target participant i should share and represents the risk factor and represents the actual risk borne by each participant, n represents the number of all participants, NSC represents the total indivisible cost of auxiliary services, E iIndicates the corresponding online power consumption of target participant i, E 总 Represents the sum of the corresponding online electricity consumption of all participants.

[0032] Optionally, step S4 includes:

[0033] S41: According to the number of participants and the number of auxiliary service cost attribute indicators, a multi-attribute decision matrix is ​​obtained;

[0034] S42: Calculating the weight of the first auxiliary service cost allocation solution and the weight of the third auxiliary service cost allocation solution according to the multi-attribute decision matrix;

[0035] S43: Calculating an entropy value of the EANS method and an entropy value of the improved Shapley value method according to the weight of the first ancillary service cost allocation solution and the weight of the third ancillary service cost allocation solution respectively;

[0036] S44: calculating the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method according to the entropy value of the EANS method and the entropy value of the improved Shapley value method respectively;

[0037] S45: Obtaining a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the second auxiliary service cost according to the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method.

[0038] Optionally, in step S5, the auxiliary service cost allocation model x i for:

[0039]

[0040] Wherein, w1 represents the first weight value corresponding to the first auxiliary service cost and represents the first auxiliary service cost that target participant i should bear, and w2 represents the second weight value corresponding to the second auxiliary service cost represents the third auxiliary service cost, d1 represents the difference coefficient of the target item attribute index under the EANS method, and d1 = 1-Y1, Y1 represents the entropy value of the EANS method, and p i1 represents the weight of the first ancillary service cost allocation solution and x i1is the parameter in the first column of the i-th row in the multi-attribute decision-making matrix and represents the auxiliary service cost sharing solution of the i-th participant using the EANS method; d2 represents the difference coefficient of the target item attribute index under the improved Shapley value method and d2 = 1-Y2, Y2 represents the entropy value of the improved Shapley value method and ε represents a constant coefficient and ε=1 / ln(n), p i2 represents the weight of the third ancillary service cost allocation solution and x i2 is the parameter in the ith row and second column of the multi-attribute decision-making matrix and represents the ancillary service cost allocation solution of the ith participant using the improved Shapley value method, SC i represents the divisible cost of auxiliary services for target participant i, n represents the number of all participants, NSC represents the indivisible total cost of auxiliary services, S represents all alliances in which target participant i participates; s represents the number of participants in the alliance S in which target participant i participates, C represents the auxiliary service cost function, C(S) represents the auxiliary service cost value that should be shared by alliance S, C(S-{i}) represents the auxiliary service cost value that should be shared by alliance S excluding participant i, C(S)-C(S-{i}) represents the marginal contribution of target participant i to the alliance S in which he participates, ! represents the factorial symbol, and E i Indicates the corresponding online power consumption of target participant i, E 总 Represents the sum of the corresponding online electricity consumption of all participants.

[0041] The present invention has the following beneficial effects:

[0042] 1. The present invention takes into account other alliances except the grand alliance and the alliance containing (n-1) participants, and considers the impact of these alliances on the allocation results.

[0043] 2. The present invention takes into account the risk sharing factors of alliance members and can truly reflect the impact of different risk levels on the sharing of auxiliary service costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the auxiliary service cost allocation method of the present invention. DETAILED DESCRIPTION

[0045] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0046] Example 1

[0047] The present invention provides an auxiliary service cost allocation method based on EANS-improved Shapley value method, referring to Figure 1 Shown, including:

[0048] S1: Use the EANS method to obtain the first auxiliary service cost that the target participant should bear;

[0049] Optionally, the step S1 includes:

[0050] S11: Obtain the divisible cost of the auxiliary services for the target participant based on the total cost of the auxiliary services borne by all participants N and the total cost of the auxiliary services borne by all participants N1 except the target participant;

[0051] In the present invention, the divisible cost of auxiliary services for a target participant is defined as the cost increase due to the target participant joining. Therefore, the divisible cost of auxiliary services for the target participant i, SC i for:

[0052] SC i =C(N)-C(N1 / i)

[0053] Among them, C(N) represents the total cost of auxiliary services borne by all participants N = {1, 2, …, n}, and C(N1 / i) represents the total cost of auxiliary services borne by all participants N1 = {1, 2, …i-1, i+1, …, n} except the target participant i.

[0054] S12: Obtain the total indivisible cost of auxiliary services incurred by all participants based on the total cost of auxiliary services borne by all participants N and the divisible cost of auxiliary services of all target participants;

[0055] For the total auxiliary service cost C(N) corresponding to the set N of all participants, there is also an indivisible cost NSC. Therefore, the present invention obtains the indivisible total auxiliary service cost incurred by all participants based on the total auxiliary service cost borne by all participants N and the divisible auxiliary service costs of all target participants, namely:

[0056] The total non-divisible cost of ancillary services NSC is:

[0057]

[0058] S13: Allocate the total indivisible cost of the auxiliary services to each participant to obtain the indivisible cost of the auxiliary services corresponding to the target participant;

[0059] S14: Obtain the first auxiliary service cost that the target participant should bear based on the indivisible auxiliary service cost and the divisible auxiliary service cost corresponding to the target participant.

[0060] The EANS allocation method is that each participant bears their own divisible cost SC iBased on this, the indivisible cost NSC is evenly allocated among all participants, and the auxiliary service cost allocation solution of target participant i based on the EANS method can be obtained, that is, the first auxiliary service cost that target participant i should bear is for:

[0061]

[0062] Among them, SC i represents the divisible cost of ancillary services for target participant i, n represents the number of all participants, and NSC represents the total indivisible cost of ancillary services.

[0063] when When NSC = 0, each participant's auxiliary service can be divided into costs SC i , is ideal, and there is no indivisible cost; when When the indivisible cost of ancillary services is positive, that is, NSC>0, the ancillary service cost share of each participant is deducted from their respective divisible costs SC i In addition to the apportioned value, a portion of the indivisible cost must be apportioned; when When the indivisible cost of ancillary services is negative, that is, NSC<0, the ancillary service cost sharing value of each participant is within their respective divisible costs SC i On the basis of the apportioned value, a smaller portion of the indivisible costs can be apportioned.

[0064] S2: Use the Shapley value method to obtain the second auxiliary service cost that the target participant should bear;

[0065] Here, first, let h i [C] is the expected shared cost of participant i based on the Shapley value method (i.e., the second ancillary service cost), and C is the ancillary service cost function. The ancillary service cost shared value of participant i based on the Shapley value method is as follows:

[0066]

[0067] Where S represents all alliances in which the target participant i participates; s represents the number of participants in the alliance S in which the target participant i participates; C represents the auxiliary service cost function; C(S) represents the auxiliary service cost value that should be shared by the alliance S; C(S-{i}) represents the auxiliary service cost value that should be shared by the alliance S excluding participant i; C(S)-C(S-{i}) represents the marginal contribution of the target participant i to the alliance S in which he participates; q(s) represents the weighting factor for how to evenly distribute the marginal contribution of the auxiliary service cost, that is, the probability of C(S)-C(S-{i}) occurring and n represents the total number of participants, and ! represents the factorial symbol.

[0068] The Shapley value allocation method takes into account many influencing factors in the allocation of ancillary service costs. It is neither a simple average cost distribution nor a distribution based solely on the installed capacity or power generation of each participant. Instead, it allocates according to the contribution of the participants. The calculation is relatively simple, meets the Pareto improvement property, and can meet the three properties required for alliance formation - individual rationality, collective rationality and effectiveness, thereby obtaining an ancillary service cost allocation plan that is relatively satisfactory to everyone.

[0069] S3: Using the risk factor to modify the second ancillary service cost to obtain a third ancillary service cost based on the improved Shapley value method;

[0070] Since the original Shapley value method used in step S2 does not fully consider the risks of participants in the alliance, it assumes that the risks borne by alliance participants are the same. However, this is only an ideal situation and does not exist in reality. The present invention considers various risk factors and proposes a Shapley value method based on risk improvement as follows:

[0071] Assume that the auxiliary service costs to be shared by different participants in practice are Due to the influence of different risk factors, the actual risk borne by each participant is The difference from the average risk is the risk factor, which is:

[0072]

[0073] and:

[0074]

[0075] in, It represents the difference between the actual risk borne by participant i and the ideal result; when When , it means that the actual risk that participant i should share is lower than the ideal risk, and the auxiliary service cost sharing amount should be reduced; when When , it means that the actual risk that participant i should share is higher than the ideal risk, and the share of ancillary service costs should be increased.

[0076] Therefore, the cost allocation correction amount Δh based on the risk factor i [C] is:

[0077]

[0078] NSC represents the non-divisible total cost of ancillary services. represents the risk factor and It represents the actual risk borne by each participant, and n represents the number of all participants.

[0079] Shapley value of ancillary service cost allocation after introducing risk factor correction

[0080]

[0081] Among them, the risk factors meet

[0082] It can be seen that each participant should share the auxiliary service cost satisfy:

[0083]

[0084] In solving the problem of ancillary service cost allocation, the Shapley value method is used to allocate the costs without considering the inconsistent risks faced by alliance members. By introducing the different actual risks of different participants under the influence of many factors, the Thus constructing risk factors This method characterizes the degree to which participants' actual ancillary service cost allocations deviate from the ideal scenario, enabling a more reasonable and accurate characterization of ancillary service cost allocations across different participants. This improved approach refines and adjusts the ancillary service cost allocations based on differences in risk factors, while maintaining the overall alliance cost. This allocation allows participants with varying risk profiles to more rationally share their respective costs, effectively stabilizing the alliance.

[0085] The above-mentioned Shapley value method based on risk factor correction can more effectively solve the problem of ancillary service cost allocation. However, since the various risk factors affecting the actual ancillary service costs cannot be intuitively obtained, and since there is a certain correlation between the differences and uncertainties in the power generation of new energy power stations of different sizes and the differences and instability in their power generation revenue and their online power generation, the present invention uses the actual online power generation ratio of each participant to calculate their risk value, that is, taking:

[0086]

[0087] Among them, E i Indicates the corresponding online power consumption of target participant i, E 总 Represents the sum of the corresponding online electricity consumption of all participants.

[0088] In summary, the third auxiliary service cost for:

[0089]

[0090] in, represents the Shapley value of ancillary service cost allocation after risk factor correction and h i (C) represents the second ancillary service cost, Δh i [C] represents the auxiliary service cost correction amount that target participant i should share and represents the risk factor and represents the actual risk borne by each participant, n represents the number of all participants, NSC represents the total indivisible cost of auxiliary services, E i Indicates the corresponding online power consumption of target participant i, E 总 Represents the sum of the corresponding online electricity consumption of all participants.

[0091] S4: performing weighted allocation on the first auxiliary service cost and the third auxiliary service cost to obtain a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the third auxiliary service cost;

[0092] Optionally, step S4 includes:

[0093] S41: According to the number of participants and the number of auxiliary service cost attribute indicators, a multi-attribute decision matrix is ​​obtained;

[0094] The multi-attribute decision matrix A is:

[0095]

[0096] Among them, each row represents a plan, and n rows in the multi-attribute decision matrix represent the existence of n members participating in the ancillary service cost sharing; each column represents an attribute indicator, and the two columns in the multi-attribute decision matrix represent two ancillary service cost sharing solutions, where the first column represents the ancillary service cost sharing solution for n members based on the EANS method, and the second column represents the ancillary service cost sharing solution for n members based on the improved Shapley value method.

[0097] Quantify each attribute index in the same way, that is, calculate the weight of the ancillary service cost allocation solution of the i-th member based on the EANS method or the improved Shapley value method:

[0098] S42: Calculating the weight of the first auxiliary service cost allocation solution and the weight of the third auxiliary service cost allocation solution according to the multi-attribute decision matrix;

[0099]

[0100] S43: Calculating an entropy value of the EANS method and an entropy value of the improved Shapley value method according to the weight of the first ancillary service cost allocation solution and the weight of the third ancillary service cost allocation solution respectively;

[0101]

[0102] Among them, Y j represents the total contribution of all members to the different ancillary service cost sharing solutions, i.e., the entropy value; ε represents a constant coefficient and ε=1 / ln(n), thus ensuring that 0≤Y j ≤1, that is, Y j The maximum value is 1.

[0103] S44: calculating the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method according to the entropy value of the EANS method and the entropy value of the improved Shapley value method respectively;

[0104]

[0105] When the contribution of each member under a certain attribute index tends to be consistent, Y j Close to 1, the difference coefficient d j tends to 0. Therefore, the coefficient of variation d j It represents the degree of consistency of contribution of each member under the j-th indicator, that is, the degree to which the auxiliary service cost sharing values ​​of each member are similar under a certain auxiliary service cost sharing solution.

[0106] S45: Obtaining a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the second auxiliary service cost according to the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method.

[0107]

[0108] Among them, w1+w2=1, when d j =0, since its weight is 0, the j-th indicator can be removed, indicating that it has no effect on the result, that is, the auxiliary service cost allocation solution can be ignored.

[0109] S5: Obtaining an auxiliary service cost allocation model using an auxiliary service cost calculation formula based on the first weight value and the second weight value;

[0110] Therefore, it can be determined that the auxiliary service cost allocation model x i for:

[0111]

[0112] Wherein, w1 represents the first weight value corresponding to the first auxiliary service cost and represents the first auxiliary service cost that target participant i should bear, and w2 represents the second weight value corresponding to the second auxiliary service cost represents the third auxiliary service cost, d1 represents the difference coefficient of the target item attribute index under the EANS method, and d1 = 1-Y1, Y1 represents the entropy value of the EANS method, and p i1 represents the weight of the first ancillary service cost allocation solution and x i1 is the parameter in the first column of the i-th row in the multi-attribute decision-making matrix and represents the auxiliary service cost sharing solution of the i-th participant using the EANS method; d2 represents the difference coefficient of the target item attribute index under the improved Shapley value method and d2 = 1-Y2, Y2 represents the entropy value of the improved Shapley value method and p i2 represents the weight of the third ancillary service cost allocation solution and x i2 is the parameter in the ith row and second column of the multi-attribute decision-making matrix and represents the ancillary service cost allocation solution of the ith participant using the improved Shapley value method, SC i represents the divisible cost of auxiliary services for target participant i, n represents the number of all participants, NSC represents the indivisible total cost of auxiliary services, S represents all alliances in which target participant i participates; s represents the number of participants in the alliance S in which target participant i participates, C represents the auxiliary service cost function, C(S) represents the auxiliary service cost value that should be shared by alliance S, C(S-{i}) represents the auxiliary service cost value that should be shared by alliance S excluding participant i, C(S)-C(S-{i}) represents the marginal contribution of target participant i to the alliance S in which he participates, ! represents the factorial symbol, and E i Indicates the corresponding online power consumption of target participant i, E 总 Represents the sum of the corresponding online electricity consumption of all participants.

[0113] S6: Allocate the target ancillary service cost using the ancillary service cost allocation model.

[0114] Example 2

[0115] This embodiment takes the sharing of the peak-shaving auxiliary service costs caused by the connection of new energy to the power grid as an example. Therefore, the entities participating in the sharing are the new energy and the load that participate in the peak-shaving auxiliary service cost sharing.

[0116] The cost of peak load ancillary services C TF The calculation formula is as follows:

[0117] C TF =AFC TF ·K TF +PNA·TF+αΔt·TF 2

[0118] First, we analyze the cost of peak load regulation ancillary services directly caused by load fluctuations without the access of new energy. price Δt-c N ·Δt+P N ·α·Δt,P N Indicates the rated generating capacity of the unit, p price represents the on-grid electricity price of the generator set, c N represents the unit variable cost under rated conditions, Δt represents the time period length in each dispatch cycle, α represents the slope of the unit variable cost function, that is, the increase in unit variable cost corresponding to a 1MW power reduction, and the peak load auxiliary service capacity directly caused by load fluctuation is TF D , then the cost of peak load regulation auxiliary services directly caused by load fluctuation is:

[0119]

[0120] in, represents the total installed capacity of peak-shaving auxiliary service units directly caused by load fluctuations. The following analyzes the cost of peak-shaving auxiliary services directly caused by renewable energy. Assume that the peak-shaving auxiliary service capacity directly caused by renewable energy fluctuations is From this, the cost of peak load ancillary services directly caused by the fluctuation of renewable energy can be obtained as follows:

[0121]

[0122] Among them, AFC TF represents the annual average fixed cost of peak load ancillary service installed capacity, Represents the total installed capacity of peak-shaving auxiliary service units directly caused by renewable energy fluctuations. Next, we analyze the cost of peak-shaving auxiliary services caused by renewable energy access to the grid. Similar to load, renewable energy output can be considered as a "negative" load. After it is connected to the grid, it is combined with the load to form an "equivalent net load curve", and then its peak-shaving auxiliary service cost is analyzed. Suppose the peak-shaving auxiliary service capacity caused by renewable energy access to the grid is According to the peak load ancillary service cost calculation formula, the total cost of peak load ancillary services caused by the access of new energy to the grid can be obtained as follows:

[0123]

[0124] in, It represents the total installed capacity of peak-shaving auxiliary service units caused by the access of new energy to the power grid.

[0125] The following uses the EANS-improved Shapley value method ancillary service cost allocation model to study and analyze the peak-shaving ancillary service cost allocation problem caused by the access of new energy to the power grid.

[0126] The two entities that participate in the cost sharing of peak load auxiliary services are new energy and load, so n = 2. Therefore, the peak load auxiliary service cost that new energy should share is for:

[0127]

[0128] Among them, E NE Indicates the amount of electricity connected to the grid by renewable energy, E D Indicates the amount of electricity connected to the load side and the cost of the load-side peak load auxiliary services for:

[0129]

[0130] Among them, the cost of new energy peak regulation can be divided into for:

[0131]

[0132] Divisible cost of load peak regulation for:

[0133]

[0134] Non-divisible cost NSC TF for:

[0135]

[0136] The cost of the peak load auxiliary service of new energy is thus shared. By deduction, we can get:

[0137]

[0138] Cost sharing of load peak load shaving ancillary services By deduction, we can get:

[0139]

[0140] Finally, it can be determined that new energy units should share the responsibility The peak load ancillary service cost is The cost will be shared by thermal power units.

[0141] The cost of backup ancillary services caused by the integration of new energy into the grid is then apportioned. Therefore, the main subjects of this study are still the two individuals that participate in the cost sharing of backup ancillary services, namely, the new energy and the load. The cost calculation formula for backup ancillary services is as follows:

[0142] C R =AFC R ·KR +PN·R

[0143] Assuming that there is no access to new energy, the reserve auxiliary service capacity directly caused by load fluctuation is The cost of standby auxiliary services directly caused by load fluctuation is:

[0144]

[0145] in, represents the total installed capacity of the standby auxiliary service units directly caused by load fluctuations. Assume that the standby auxiliary service capacity directly caused by the fluctuation of renewable energy is The cost of standby auxiliary services directly caused by the fluctuation of new energy is:

[0146]

[0147] in, represents the total installed capacity of the standby auxiliary service units directly caused by the fluctuation of renewable energy. Assume that the standby auxiliary service capacity caused by the access of renewable energy to the grid is The total cost of backup ancillary services caused by the access of new energy to the grid is:

[0148]

[0149] in, Represents the total installed capacity of standby auxiliary service units caused by the access of renewable energy to the grid. Similarly, the EANS-Improved Shapley Value method auxiliary service cost allocation model is used to study and analyze the standby auxiliary service cost allocation problem caused by the access of renewable energy to the grid. The EANS-Improved Shapley Value method allocation solution for the standby auxiliary service cost of renewable energy can be obtained. for:

[0150]

[0151] Load reserve ancillary service cost EANS-improved Shapley value method allocation solution for:

[0152]

[0153] Among them, the divisible cost of new energy standby for:

[0154]

[0155] Load reserve divisible cost for:

[0156]

[0157] Non-divisible cost NSC R for:

[0158]

[0159] The cost of backup auxiliary services for new energy is thus allocated By deduction, we can get:

[0160]

[0161] Share of backup ancillary service costs for load By deduction, we can get:

[0162]

[0163] Similar to the cost sharing of peak load auxiliary services, it can be determined that new energy units should share the cost. The backup ancillary service cost on the load side is The cost will be shared by thermal power units.

[0164] The ancillary service cost formulas for wind power connected to the grid and caused by the equivalent net load, and the ancillary service cost formula for photovoltaic connected to the grid combined with the equivalent net load, and the ancillary service cost allocation model of the present invention are used respectively to derive a wind power and photovoltaic ancillary service cost allocation model based on the EANS-improved Shapley value method, and then the wind power and photovoltaic ancillary service costs are allocated.

[0165] The specific process is:

[0166] The cost of peak-shaving auxiliary services caused by the connection of wind power and photovoltaic power to the grid is shared, so the main body of sharing is the wind power and photovoltaic power that participate in the sharing of peak-shaving auxiliary service costs. The total cost of peak-shaving auxiliary services caused by equivalent net load after wind power is connected to the grid is determined as:

[0167]

[0168] in, It represents the total installed capacity of peak load-shaving auxiliary service units caused by the connection of wind power to the grid.

[0169] The total cost of peak load regulation ancillary services caused by equivalent net load after photovoltaic power is connected to the grid is:

[0170]

[0171] in, It represents the total installed capacity of peak-shaving auxiliary service units caused by photovoltaic power grid access.

[0172] The total cost of peak load regulation ancillary services caused by the equivalent net load after photovoltaic and wind power are connected to the grid is:

[0173]

[0174] in, It represents the total installed capacity of peak load-shaving auxiliary service units caused by the joint access of photovoltaic and wind power to the grid.

[0175] The contribution value of peak-shaving auxiliary service costs caused by wind power access to the grid can be defined as the difference between the peak-shaving auxiliary service costs caused by the equivalent net load after access to the grid and the peak-shaving auxiliary service costs directly caused by load fluctuations before access.

[0176] Therefore, combined with the formula C in the previous article R =AFC R ·K R + PN·R, the contribution of peak load ancillary service cost due to wind power access to the grid is:

[0177]

[0178] Similarly, the contribution value of the peak-shaving auxiliary service cost caused by PV grid access can be defined as the difference between the peak-shaving auxiliary service cost caused by the equivalent net load after grid access and the peak-shaving auxiliary service cost directly caused by load fluctuations before grid access. Therefore, the contribution value of the peak-shaving auxiliary service cost caused by PV grid access is:

[0179]

[0180] Similarly, the contribution value of peak load ancillary service cost caused by the joint access of wind power and photovoltaic power to the grid is:

[0181]

[0182] The EANS-improved Shapley value method ancillary service cost allocation model mentioned above is used to study and analyze the problem of peak load ancillary service cost allocation caused by the separate connection of wind power and photovoltaic power to the grid. The peak load ancillary service cost that wind power should share is obtained. for:

[0183]

[0184] Among them, E W Indicates the amount of wind power on the grid, MW; E PV Indicates the photovoltaic grid-connected power, MW. The photovoltaic peak load auxiliary service cost to be shared for:

[0185]

[0186] Among them, the cost of wind power peak regulation can be divided into for:

[0187]

[0188] Photovoltaic peak regulation separable costs for:

[0189]

[0190] Non-divisible cost NSC TFWPV for:

[0191]

[0192] The cost of wind power peak load ancillary services is thus allocated By deduction, we can get:

[0193]

[0194] Cost sharing of photovoltaic peak-shaving auxiliary services By deduction, we can get:

[0195]

[0196] It can be determined that wind turbines should share the responsibility The peak load auxiliary service cost should be shared by the photovoltaic units. The cost of peak load ancillary services.

[0197] Then, the cost of standby auxiliary services caused by the separate connection of wind power and photovoltaic power to the power grid is shared. Therefore, the main bodies of this sharing are the two individuals, wind power and photovoltaic power, that participate in the sharing of standby auxiliary service costs.

[0198] The contribution of standby auxiliary service costs caused by wind power access to the grid is:

[0199]

[0200] The contribution of standby auxiliary service costs due to PV grid access is:

[0201]

[0202] The contribution of standby auxiliary service costs caused by the joint access of wind power and photovoltaic power to the grid is:

[0203]

[0204] The backup ancillary service cost that wind power should share is for:

[0205]

[0206] PV's share of backup ancillary service costs for:

[0207]

[0208] Among them, the divisible cost of wind power reserve for:

[0209]

[0210] Photovoltaic backup divisible costs for:

[0211]

[0212] Non-divisible cost NSC RWPV for:

[0213]

[0214] Therefore, the cost of wind power backup ancillary services is allocated By deduction, we can get:

[0215]

[0216] Share of the cost of photovoltaic backup ancillary services By deduction, we can get:

[0217]

[0218] Wind turbines should share the responsibility The backup auxiliary service costs should be shared by the PV units. The cost of standby ancillary services.

[0219] First, the auxiliary service cost allocation model of the present invention is used to obtain the auxiliary service cost allocation model of the new energy individual auxiliary service based on the EANS-improved Shapley value method after the new energy power station is connected to the grid, and then the individual auxiliary service cost of the new energy is allocated. The specific contents of the correction and allocation process are as follows

[0220] The cost of peak-shaving auxiliary services is shared among individual renewable energy sources, that is, how to share the cost of peak-shaving auxiliary services caused by the connection of renewable energy power stations to the grid. Therefore, the main participants in the sharing are the n renewable energy power stations that participate in the sharing of peak-shaving auxiliary service costs. According to the cost calculation formula for peak-shaving auxiliary services, the total cost of peak-shaving auxiliary services caused by the equivalent net load after renewable energy power station i is connected to the grid is:

[0221]

[0222] in, It represents the total installed capacity of peak load-shaving auxiliary service units caused by the access of new energy power station i to the grid.

[0223] Then the contribution value of the peak load ancillary service cost caused by the access of new energy power station i to the grid is:

[0224]

[0225] Similarly, the contribution value of the peak load ancillary service cost caused by all new energy power stations forming a large alliance and connecting to the power grid is:

[0226]

[0227] in, It represents the total installed capacity of peak load ancillary service units caused by all new energy power stations forming a large alliance and connecting to the grid. The contribution value of peak load ancillary service cost caused by the joint connection of all other new energy power stations to the grid, except for new energy power station i, is:

[0228]

[0229] in, represents the total installed capacity of peak-shaving auxiliary service units caused by the joint access of all other new energy power stations except new energy power station i to the grid. From this, the divisible peak-shaving cost of new energy power station i can be obtained. for:

[0230]

[0231] Non-divisible cost NSC TFn for:

[0232]

[0233] The cost allocation solution for peak load ancillary services of renewable energy power station i based on the EANS method is as follows:

[0234]

[0235] Next, we calculate the cost allocation solution for peak-shaving auxiliary services of renewable energy power station i based on the improved Shapley value method. First, we determine the contribution value of peak-shaving auxiliary services cost caused by the access of any alliance S containing renewable energy power station i to the grid:

[0236]

[0237] in, It represents the total installed capacity of peak load auxiliary service units caused by the access of any alliance S containing new energy power station i to the grid.

[0238] In the alliance S, except for the new energy power station i, the contribution value of the peak load ancillary service cost caused by the joint access of the remaining new energy power stations to the grid is:

[0239]

[0240] in, represents the total installed capacity of peak-shaving auxiliary service units caused by the joint access of all new energy power stations except new energy power station i to the grid. From this, we can determine the marginal contribution of new energy power station i to the peak-shaving auxiliary service cost of the alliance S as:

[0241]

[0242] From the modeling analysis of the improved Shapley value method in the previous article, it can be seen that in the cost allocation solution based on the improved Shapley value method, the correction term of the peak load ancillary service cost allocation solution of the new energy power station i with the introduction of risk factors is:

[0243]

[0244] Among them, E i represents the on-grid power of new energy power station i, MW; E 总 represents the sum of the grid-connected power generated by all renewable energy power stations in a large alliance, in MW. Thus, the cost allocation solution for peak-shaving auxiliary services of renewable energy power station i based on the improved Shapley value method is as follows:

[0245]

[0246] Based on the EANS-Improved Shapley Value method ancillary service cost allocation model mentioned above, the EANS-Improved Shapley Value method peak load ancillary service cost allocation solution for renewable energy power station i is:

[0247]

[0248] Finally, it can be determined that the new energy power station i should bear The cost of peak load ancillary services.

[0249] Similarly, the backup auxiliary service costs among individual renewable energy sources are allocated based on the EANS-improved Shapley value method. for:

[0250]

[0251] Non-divisible cost NSC Rn for:

[0252]

[0253] The cost allocation solution of standby auxiliary services of renewable energy power station i based on EANS method is as follows:

[0254]

[0255] The marginal contribution of the backup auxiliary service cost of the new energy power station i to the alliance S is:

[0256]

[0257] Based on the improved Shapley value method, the correction term of the standby auxiliary service cost allocation solution of the new energy power station i with the risk factor is obtained as follows:

[0258]

[0259] Therefore, the cost allocation solution of standby auxiliary services of renewable energy power station i based on the improved Shapley value method is as follows:

[0260]

[0261] Similarly, based on the EANS-Improved Shapley Value method ancillary service cost allocation model mentioned above, the EANS-Improved Shapley Value method standby ancillary service cost allocation solution for renewable energy power station i is:

[0262]

[0263] Finally, it can be determined that the new energy power station i should bear The cost of standby ancillary services.

[0264] Working Principle: This invention constructs an ancillary service cost allocation model based on the EANS-modified Shapley value method. The EANS method is first used to determine the divisible ancillary service costs to be borne by each participant i. The non-divisible ancillary service costs (NSC) are then evenly allocated among all participants. The Shapley value method is then used to calculate the average marginal contribution of participant i to each alliance in which it participates. Risk factors are then used to characterize the various risk factors affecting ancillary service cost allocation, and the results obtained by the Shapley value method are corrected. A multi-attribute decision matrix is ​​then used to determine the weights of the two items in the EANS-modified Shapley value method. Finally, the EANS-modified Shapley value method is used to allocate ancillary service costs.

[0265] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for allocating costs of auxiliary services, characterized in that: include: S1: Use the EANS method to obtain the first auxiliary service cost that the target participant should bear; The target participants i First ancillary service costs to be borne for: in, Indicates the target participant i The divisible costs of ancillary services are n Indicates the number of all participants, represents the total indivisible cost of ancillary services; S2: Use the Shapley value method to obtain the second auxiliary service cost that the target participant should bear; The second ancillary service cost for: in, S Indicates the target participant i All alliances involved; s Indicates the target participant i Alliances participated in S The number of participants, C represents the ancillary service cost function, Represents an alliance S The value of ancillary service costs to be shared, Represents an alliance S Excluding participants i The auxiliary service cost value that should be shared by the foreign exchange is Indicates the target participant i Alliances S The marginal contribution of The weighting factor that represents how to evenly distribute the marginal contribution of ancillary service costs is: The probability of occurrence and , n represents the number of all participants, ! represents the factorial symbol; S3: Using the risk factor to modify the second ancillary service cost to obtain a third ancillary service cost based on the improved Shapley value method; The third ancillary service cost for: in, represents the Shapley value of ancillary service cost allocation after risk factor correction and , represents the second ancillary service cost, Indicates the target participant i The amount of ancillary service cost correction to be allocated and , represents the risk factor and , Indicates the actual risk borne by each participant. represents the total indivisible cost of ancillary services, Indicates the target participant i The corresponding Internet power consumption, n Indicates the number of all participants, Represents the sum of the corresponding online electricity consumption of all participants; S4: performing weighted allocation on the first auxiliary service cost and the third auxiliary service cost to obtain a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the third auxiliary service cost; S5: Obtaining an auxiliary service cost allocation model using an auxiliary service cost calculation formula based on the first weight value and the second weight value; S6: Allocate the target ancillary service cost using the ancillary service cost allocation model.

2. The auxiliary service cost allocation method according to claim 1, characterized in that: The step S1 comprises: S11: According to all participants N Total ancillary service costs borne by all participants except the target participant N 1. The total cost of auxiliary services borne by the target participant is used to obtain the divisible cost of auxiliary services for the target participant; S12: According to all participants N The total cost of ancillary services undertaken and the divisible cost of ancillary services of all target participants are used to obtain the total indivisible cost of ancillary services incurred by all participants; S13: Allocate the total indivisible cost of the auxiliary services to each participant to obtain the indivisible cost of the auxiliary services corresponding to the target participant; S14: Obtain the first auxiliary service cost that the target participant should bear based on the indivisible auxiliary service cost and the divisible auxiliary service cost corresponding to the target participant.

3. The auxiliary service cost allocation method according to claim 2, characterized in that: The target participants i Separable costs of ancillary services for: The total indivisible cost of ancillary services for: in, Indicates all participants Total ancillary service costs incurred, Indicates that except the target participant All other participants Total costs of ancillary services incurred.

4. The auxiliary service cost allocation method according to claim 1, characterized in that: The step S4 comprises: S41: According to the number of participants and the number of auxiliary service cost attribute indicators, a multi-attribute decision matrix is ​​obtained; S42: Calculating the weight of the first auxiliary service cost allocation solution and the weight of the third auxiliary service cost allocation solution according to the multi-attribute decision matrix; S43: Calculating an entropy value of the EANS method and an entropy value of the improved Shapley value method according to the weight of the first ancillary service cost allocation solution and the weight of the third ancillary service cost allocation solution respectively; S44: calculating the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method according to the entropy value of the EANS method and the entropy value of the improved Shapley value method respectively; S45: Obtaining a first weight value corresponding to the first auxiliary service cost and a second weight value corresponding to the second auxiliary service cost according to the difference coefficient of the target item attribute index under the EANS method and the difference coefficient under the improved Shapley value method.

5. The auxiliary service cost allocation method according to claim 1, characterized in that: In step S5, the auxiliary service cost sharing model for: in, represents the first weight value corresponding to the first auxiliary service cost and , Indicates the target participant i The first auxiliary service costs to be borne, Indicates the second weight value corresponding to the second auxiliary service cost , represents the third ancillary service cost, represents the difference coefficient of the target item attribute index under the EANS method and , represents the entropy value of the EANS method and , represents the weight of the first ancillary service cost allocation solution and , is the first i The first column of the row is the parameter and represents the i The cost sharing solution for ancillary services among participants using the EANS method; It represents the difference coefficient of the target item attribute index under the improved Shapley value method and , represents the entropy value of the improved Shapley value method and , represents a constant coefficient and , represents the weight of the third ancillary service cost allocation solution and , is the first i The second column of the row is the parameter and represents the i The cost sharing solution of ancillary services among participants using the improved Shapley value method is: Indicates the target participant i The divisible costs of ancillary services are n Indicates the number of all participants, represents the total indivisible cost of ancillary services, S Indicates the target participant i All alliances involved; s Indicates the target participant i Alliances participated in S The number of participants, C represents the ancillary service cost function, Represents an alliance S The value of ancillary service costs to be shared, Represents an alliance S Excluding participants i The auxiliary service cost value that should be shared by the foreign exchange is Indicates the target participant i Alliances S The marginal contribution of , ! represents the factorial symbol, Indicates the target participant i The corresponding Internet power consumption, Represents the sum of the corresponding online electricity consumption of all participants.

Images