Coordination mechanism utility incentive and user payment capacity market hierarchical design method

Abstract

A capacity market hierarchical design method for coordination mechanism utility incentive and user payment comprises the following steps: 1) determining a capacity demand sum according to a power system planning target in combination with a load prediction condition of a target year; 2) constructing a plurality of sub-capacity market platforms; 3) constructing a sub-market division decision model to which the power supply belongs; 4) solving the sub-market division decision model to which the power supply belongs, classifying the power supplies with different cost structures into corresponding sub-capacity market platforms, and determining the capacity demand of each sub-capacity market platform; 5) determining key boundary parameters of each sub-capacity market platform through a capacity market parameter optimization method considering a mechanism utility incentive target; and 6) feeding back the key boundary parameters of each sub-capacity market platform to each sub-capacity market platform, and constructing a hierarchical capacity market of coordination mechanism utility excitation and user payment. According to the invention, the market riding phenomenon caused by the multi-element main body cost structure difference is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power systems and their automation technology, in particular to a capacity market hierarchical design method for coordinating mechanism utility incentive and user payment. BACKGROUND

[0002] The capacity market mainly solves the problem of fixed investment cost recovery of power sources in the electricity market based on variable cost bidding, thereby generating investment incentive signals and guiding power source investment. In a single type of power system (such as a traditional power system dominated by thermal power), the fixed cost and variable cost of power sources are similar, and a single capacity market can effectively encourage market participants to invest and plan and encourage market participants to eliminate high-cost units. However, with the continuous acceleration of the green and low-carbon transformation of the power system, the proportion of new energy generation continues to grow, and the main body on the generation side is showing a trend of diversification, while the cost structures of different types of power sources are different, and the single capacity market mechanism has been difficult to coordinate the needs of market participants. Practice shows that it is difficult for a single capacity market mechanism to provide accurate and effective price signals for commodities with large differences in cost structure.

[0003] Currently, the power system is mainly composed of new energy (mainly wind power and photovoltaic) and thermal power, and the cost structures of the power sources are significantly different. Under normal circumstances, the fixed cost proportion of wind power and photovoltaic is much higher than that of thermal power. If the capacity market lacks consideration of the differences in cost structure and cost recovery of power sources, it will be difficult to take into account the different cost recovery demands of various types of power sources, affecting the accurate generation of investment incentive signals. For example, when new energy and flexible power sources such as thermal power participate in the electricity spot market at the same time, based on the theory of nodal price, after sorting by price from low to high, the last thermal power unit that meets the power supply requirement will be the marginal unit, and its marginal cost will be set as the spot market clearing price. In this case, new energy has a higher generation profit space due to its low generation cost. At the same time, new energy can obtain additional subsidies through green certificate trading, and the generation income of new energy can almost cover all costs or be higher than costs.

[0004] In addition, in the capacity market, the capacity resources are quoted according to the cost shortage, that is, the unrecouped cost is reported. For new energy, although the investment cost is higher than that of conventional flexible power supply, since it has a higher income space in the spot market, it can report a lower price in the capacity market. According to the supply and demand matching principle, usually, high-priced units (such as gas) become marginal units, and their capacity quotes are set as the capacity market clearing price. Therefore, new energy has little demand to recover the cost through the capacity market, but it obtains excess profits through 'piggybacking', causing the capacity market to compensate for high fees. This is not conducive to guiding the smooth transformation of power supply structure and affecting the true utility of the capacity market. In the long run, it will form a single and unreasonable power supply structure, leading to the deviation of the capacity market incentive signal from the original intention of market mechanism design. Although the existing actual market reduces the new energy income space by calculating the equivalent reliable capacity, it still cannot avoid the new energy units that have achieved profits in the energy market from obtaining additional capacity income in the capacity market. SUMMARY

[0005] The purpose of the present application is to provide a capacity market hierarchical design method that coordinates mechanism utility incentive and user payment, comprising the following steps:

[0006] 1) According to the power system planning target, combined with the load prediction situation in the target year, determine the total capacity demand.

[0007] 2) Build multiple sub-capacity market platforms.

[0008] 3) Based on the total capacity demand, build a power source belonging to sub-market division decision model.

[0009] 4) Solve the power source belonging to sub-market division decision model, and classify the power sources with different cost structures into the corresponding sub-capacity market platform, and determine the capacity demand of each sub-capacity market platform.

[0010] 5) Based on the power sources and capacity demand divided by each sub-capacity market platform, determine the key boundary parameters of each sub-capacity market platform through a capacity market parameter optimization method considering the mechanism utility incentive target.

[0011] 6) Feedback the key boundary parameters of each sub-capacity market platform to each sub-capacity market platform, and build a hierarchical capacity market that coordinates mechanism utility incentive and user payment.

[0012] Further, the power sources include wind power sources, coal-fired power sources, and photovoltaic power sources.

[0013] Further, the power source belonging to sub-market division decision model includes a double-layer optimization model.

[0014] Further, when the power source belonging to sub-market division decision model is a double-layer optimization model, the objective function of the upper layer optimization model is as follows:

[0015]

[0016] In the formula, This represents the total payment amount for each sub-capacity market platform. p and i are the indices of the sub-capacity market and the power supply, respectively. These are the clearing price and clearing capacity of the i-th power source in the p-th sub-capacity market, respectively.

[0017] The objective function of the lower-level optimization model is shown below:

[0018]

[0019] In the formula, Quote the capacity of the i-th power source.

[0020] Furthermore, when the decision model for allocating a power source to a sub-market is a two-level optimization model, the constraints of the upper-level optimization model include capacity market demand constraints, revenue equilibrium constraints, and the unique constraint of the power source's sub-market affiliation.

[0021] The constraints of the lower-level optimization model include capacity balance constraints and upper and lower limits of the winning bid capacity.

[0022] Furthermore, the capacity market demand constraint is as follows:

[0023]

[0024] In the formula, p is the index of the sub-capacity market. This represents the total capacity demand. Let p be the capacity demand of the p-th sub-capacity market.

[0025] The benefit equilibrium constraint is as follows:

[0026]

[0027] In the formula, i is the index of the power source. This represents the average yield of all power supplies. NI is the total number of power supplies. This represents the upper limit of the equilibrium of returns.

[0028] Wherein, the rate of return η of the i-th power source i As shown below:

[0029]

[0030] In the formula, CT i Let represent the total annual revenue from the energy market, the total annual revenue from the capacity market, and the total annual cost of the i-th power source, respectively. Let i be the initial capacity provided by the i-th power source in the target year. For the i-th power source, according to the initial capacity The projected total energy market revenue for the target year. CT i p For the i-th power source, according to the initial capacity The total cost for the projected target year. These are the clearing price and clearing capacity of the i-th power source in the p-th sub-capacity market, respectively.

[0031] The unique constraint on the power supply ownership submarket is as follows:

[0032]

[0033] In the formula, To characterize the state variable representing the i-th power source belonging to the p-th sub-capacity market, if the i-th power source belongs to the p-th sub-capacity market, then... Conversely,

[0034] The capacity balance constraints are as follows:

[0035]

[0036] In the formula, This is the dual multiplier corresponding to the capacity balance constraint.

[0037] The upper and lower limits of the winning bid capacity are constrained as follows:

[0038]

[0039] In the formula, These represent the upper and lower limits of the winning bid capacity for the i-th power source in the p-th sub-capacity market, respectively. For the i-th power source, the dual multiplier corresponding to the upper and lower limits of the winning bid capacity in the p-th sub-capacity market. These represent the upper and lower limits of the capacity for the i-th power source to participate in the sub-capacity market. It is the dual multiplier with the unique constraint of the power supply submarket.

[0040] Furthermore, when the decision model for classifying the power source's sub-market is a two-level optimization model, the steps for solving the decision model for classifying the power source's sub-market are as follows:

[0041] 4.1) Based on KKT conditions, the two-level optimization model is transformed into a single-level MPEC model, that is, the KKT conditions of the lower-level optimization model are added as constraints to the upper-level optimization model, as shown below:

[0042]

[0043] where p, i are the indices of sub-capacity market, power source, respectively. is the dual multiplier corresponding to the capacity balance constraint. is the dual multiplier corresponding to the upper and lower bound of the winning capacity. are the upper and lower bound of the winning capacity of the ith power source in the pth sub-capacity market. is the clearing capacity of the ith power source in the pth sub-capacity market. is the capacity bid of the ith power source.

[0044] 4.2) Convert the single-level MPEC model into a MILP model based on linearization, as follows:

[0045]

[0046]

[0047] where, is a 0-1 variable. is a positive number. q is the index of the linearization segment point. is the clearing price of the ith power source in the pth sub-capacity market. is the capacity corresponding to the segment point q. is the distance between any two consecutive segment points. is a 0-1 variable. is a continuous variable, when , is the clearing price M q is a positive number.

[0048] 4.3) Solve the MILP model using a solver to classify the power sources with different cost structures into the corresponding sub-capacity market platforms and determine the capacity demand of each sub-capacity market platform.

[0049] Further, in step 5), the steps for determining the key boundary parameters of each sub-capacity market platform are as follows:

[0050] 5.1) Construct a capacity investment simulation model.

[0051] 5.2) Based on the power sources and capacity demand classified by each sub-capacity market platform, run the capacity investment simulation model and evaluate the running results of the capacity investment simulation model through mechanism utility.

[0052] 5.3) Based on the evaluated running results, feedback and correct the key boundary parameters of each sub-capacity market platform.

[0053] Furthermore, the capacity investment simulation model includes an initial investment decision module, a capacity market pricing decision module, a capacity market clearing module, and a final investment decision module.

[0054] The initial investment decision module is shown below:

[0055]

[0056] In the formula, y is the year index. For e-commerce indexing. Indicates the first The initial investment capacity of a power generator in year y. Indicates the first The net profit of a power generator in year y-1. i is the index of the power source. For the first A collection of power supplies for a power generator. i,y-1 Let represent the net profit of the i-th power source in year y-1. These are the upper and lower limits for permitted investment. This represents the upper limit of investment intentions. Investment rate. For the first The total capacity of a generator in year y. For the first The total fixed cost of the power supply type invested by each power generator. is the coupling coefficient.

[0057] The capacity market pricing decision module is shown below:

[0058]

[0059] In the formula, m represents the delivery period. Pl y This represents the load in year y. Let f represent the predicted load for year y+m. Let f be the actual load growth rate. Let g be the percentage of prediction error. Let be the estimated total annual cost of the i-th power source in year y+m. Let represent the projected market revenue for the i-th power source in year y+m, excluding the capacity market. Let be the capacity of the i-th power source. Let represent the market price for the capacity of the i-th power source in year y.

[0060] The capacity market clearing module is shown below:

[0061]

[0062] In the formula, Q represents capacity. d (Q) represents the capacity demand curve. is the winning capacity of the i-th power source in the y-th year capacity market. C_market,y is the set of power sources participating in the y-th year capacity market auction. is the upper limit of the winning capacity of the i-th power source. is the capacity market clearing price. is the capacity market annual revenue of the i-th power source in the y-th year. θ represents the capacity market utility.

[0063] The final investment decision module is shown as follows:

[0064] When the net revenue rate is greater than the set benchmark revenue rate value , the investment decision is not adjusted.

[0065] When the net revenue rate is greater than the set benchmark revenue rate value , the investment decision sets the maximum allowed investment upper limit

[0066] When the net revenue rate is less than the set benchmark revenue rate value , the power producer reduces the investment amount until it completely gives up capacity investment.

[0067] The operation results of the capacity investment simulation model through the mechanism utility assessment include the system installed capacity in the target year

[0068] Further, the step of feedback correction of the key boundary parameters of each sub-capacity market platform is shown as follows:

[0069] 5.3.1) Set the initial value of the capacity adjustment coefficient and the price adjustment coefficient to 1, the iteration number k = 0, and set the target installed capacity

[0070] 5.3.2) Let the iteration number k = k + 1, and calculate the capacity demand parameter X C and the capacity price parameter X P based on the capacity adjustment coefficient and the price adjustment coefficient as follows:

[0071]

[0072] In the formula, is the capacity demand parameter value. L max is the capacity delivery year peak load. IRM is the system reserve margin. e is the average equivalent availability rate of power sources in the region. is the capacity price parameter value. C G is the total cost of the reference power source. I EThe estimated annual revenue of the reference power source in the capacity delivery year in the electricity energy market. A The estimated annual revenue of the reference power source in the capacity delivery year in the ancillary service market. g The power source capacity.

[0073] 5.3.3) Based on the capacity demand parameter X C and the capacity price parameter X P , determine the capacity demand curve F d (Q), run the capacity investment simulation model to obtain the system installed capacity in the target year

[0074] 5.3.4) Calculate the difference between the system installed capacity in the target year and the target installed capacity , and determine whether the absolute value of the difference is less than a preset threshold value, if yes, output the capacity adjustment coefficient and the price adjustment coefficient , if not, go to step 5.3.5).

[0075] 5.3.5) Update the capacity adjustment coefficient and the price adjustment coefficient , and return to step 5.3.2).

[0076] The update strategy of the capacity adjustment coefficient and the price adjustment coefficient is as follows:

[0077]

[0078] In the formula, x (k) represents the capacity adjustment coefficient or the price adjustment coefficient x (k+1) in the kth iteration, x (k-1) represents the capacity adjustment coefficient or the price adjustment coefficient x (k) in the k+1th iteration, and x (k) represents the capacity adjustment coefficient or the price adjustment coefficient x (k) in the k-1th iteration. f(x (k) ) represents the adjustment coefficient update strategy. f'(x initial ) represents the derivative of the update strategy f(x I ) with respect to x (k) . represents the system installed capacity in the target year. Q initial represents the initial system capacity. Q I (x (k) ), Q I (x (k-1)respectively represent the system investment increment at the kth, k-1th iteration. S (k) represents the change rate at the kth iteration. represents the change rate at the initial iteration.

[0079] The technical effect of the present application is self-evident. The present application can effectively improve the market hitch phenomenon caused by the cost structure difference of multiple subjects, reduce the capacity market fee payment under the premise of ensuring that the capacity market plays the expected investment incentive role, effectively alleviate the user payment burden, and realize the orderly coordination of investment incentive utility and user payment.

[0080] In the capacity market mechanism based on marginal clearing, the present application provides a capacity market mechanism design scheme that can adapt to the change trend of multiple subjects on the power generation side. The method improves the uneven capacity compensation caused by the cost structure difference of power sources by reasonably reconstructing the market mechanism, further improves the economy on the basis of ensuring that the capacity market plays an incentive role, and promotes the capacity market to generate effective and economic investment incentive signals.

[0081] The present application proposes a hierarchical capacity market bidding mode based on the current capacity market mechanism. The power sources are classified into multiple sub-capacity markets according to the similarity of the cost structure, and a hierarchical capacity cost recovery bidding market is formed. The hierarchical capacity market proposed in the present application maintains the basic framework of the existing capacity market, and usually adopts a capacity price market pricing mechanism based on marginal clearing to avoid high implicit costs caused by changes in the market mechanism. The method proposed in the present application classifies power sources with similar cost structures into the same layer, realizes the same bidding of power sources with similar cost structures, solves or alleviates the capacity market hitch phenomenon, realizes the market mechanism utility incentive target, and reduces the excessive payment of the capacity market. The core of the hierarchical design scheme of the capacity market is the refinement of the bidding platform, that is, the single capacity market platform is divided into multiple sub-capacity market platforms according to the cost structure difference of the power sources. The sub-capacity market platforms are organized and operated by the market operation agency in the same period in units of years, and are respectively auctioned according to the capacity demand target of each sub-capacity market platform in a certain capacity payment year in the future. Each sub-capacity market platform adopts a centralized bidding mechanism, and the market operation agency assumes the roles of capacity fee payment and capacity fee diversion respectively towards the power generation side and the user side. BRIEF DESCRIPTION OF DRAWINGS

[0082] Figure 1 FIG. 1 is a schematic diagram of the hierarchical design method for the capacity market;

[0083] Figure 2 FIG. 2 is a schematic diagram of the cost value difference of each type of subject;

[0084] Figure 3 FIG. 3 is a schematic diagram of the capacity market hitch phenomenon caused by the cost structure difference of power sources;

[0085] Figure 4 A parameter feedback correction method flow chart. DETAILED DESCRIPTION

[0086] The application will be further described in conjunction with the examples below, but should not be understood as limiting the above-mentioned subject matter of the application to the following examples. Various substitutions and modifications can be made according to ordinary technical knowledge and conventional means without departing from the above-mentioned technical idea of the application, and all should be included in the protection scope of the application.

[0087] Example 1

[0088] Referring to Figures 1 to 4 A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, comprising the following steps:

[0089] 1) According to the power system planning target, combined with the load prediction situation of the target year, the total capacity demand is determined.

[0090] The determination method of the total capacity demand is as follows: according to the net load curve obtained from the load prediction situation and new energy output, and the expected value of loss of load calculated by combining the system loss of load probability, the total installed capacity demand of the specific system is obtained (the total installed capacity of the system is continuously increased in an iterative manner to make the loss of load and other data meet the demand).

[0091] 2) Construct multiple sub-capacity market platforms.

[0092] 3) Based on the total capacity demand, a power source sub-market division decision model is constructed.

[0093] 4) The power source sub-market division decision model is solved, the power sources with different cost structures are classified into the corresponding sub-capacity market platforms, and the capacity demand of each sub-capacity market platform is determined.

[0094] 5) Based on the power sources and capacity demand classified by each sub-capacity market platform, the key boundary parameters of each sub-capacity market platform are determined through a capacity market parameter optimization method considering the mechanism utility incentive target.

[0095] 6) The key boundary parameters of each sub-capacity market platform are fed back to each sub-capacity market platform, and a hierarchical capacity market for coordinating mechanism utility incentive and user payment is constructed.

[0096] Example 2

[0097] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in example 1, further, the power source includes wind power source, coal-fired power source, photovoltaic power source.

[0098] Example 3

[0099] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in any one of embodiments 1 to 2, further, the sub-market division decision model of the power source includes a bi-level optimization model.

[0100] Embodiment 4:

[0101] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in any one of embodiments 1 to 3, further, when the sub-market division decision model of the power source is a bi-level optimization model, the objective function of the upper optimization model is shown in the formula (1) of the summary.

[0102] The objective function of the lower optimization model is shown in the formula (2) of the summary.

[0103] Embodiment 5:

[0104] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in any one of embodiments 1 to 4, further, when the sub-market division decision model of the power source is a bi-level optimization model, the constraint conditions of the upper optimization model include capacity market demand constraints, revenue equilibrium constraints, and power source belonging to a unique sub-market constraint.

[0105] The constraint conditions of the lower optimization model include capacity balance constraints and upper and lower limits of winning capacity constraints.

[0106] Embodiment 6:

[0107] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in any one of embodiments 1 to 5, further, the capacity market demand constraints are shown in the formula (3) of the summary.

[0108] The revenue equilibrium constraints are shown in the formulas (3)-(8) of the summary.

[0109] The power source belonging to a unique sub-market constraint is shown in the formula (9) of the summary.

[0110] The capacity balance constraints are shown in the formula (10) of the summary.

[0111] The upper and lower limits of winning capacity constraints are shown in the formulas (11)-(13) of the summary.

[0112] Embodiment 7:

[0113] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content is seen in any one of embodiments 1 to 6, further, when the sub-market division decision model of the power source is a bi-level optimization model, the steps for solving the sub-market division decision model of the power source are as follows:

[0114] 4.1) Transform the bi-level optimization model into a single-level MPEC model based on KKT conditions, i.e. add the KKT conditions of the lower-level optimization model as constraints to the upper-level optimization model, as shown in the summary formula (14)-(18).

[0115] 4.2) Transform the single-level MPEC model into a MILP model based on linearization processing, as shown in the summary formula (19)-(26).

[0116] wherein the distance between any two consecutive segment points is a constant.

[0117] 4.3) Solve the MILP model by using a solver, classify the power sources with different cost structures into corresponding sub-capacity market platforms, and determine the capacity demand of each sub-capacity market platform.

[0118] The solver includes a CPLEX solver.

[0119] Embodiment 8:

[0120] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content of which is any one of embodiments 1 to 7, further, in step 5), the steps of determining the key boundary parameters of each sub-capacity market platform are as follows:

[0121] 5.1) Build a capacity investment simulation model.

[0122] 5.2) Based on the power sources and capacity demand divided by each sub-capacity market platform, run the capacity investment simulation model, and evaluate the running results of the capacity investment simulation model through mechanism utility.

[0123] 5.3) Based on the evaluated running results, feedback correct the key boundary parameters of each sub-capacity market platform.

[0124] Embodiment 9:

[0125] A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, the main technical content of which is any one of embodiments 1 to 8, further, the capacity investment simulation model includes an initial investment decision module, a capacity market bidding decision module, a capacity market clearing module, and a final investment decision module.

[0126] The initial investment decision module is shown in the summary formula (27)-(30).

[0127] The capacity market bidding decision module is shown in the summary formula (30)-(32).

[0128] The capacity market clearing module is shown in the summary formula (33)-(36).

[0129] The final investment decision module is shown as follows:

[0130] When the net benefit rate is greater than the set benchmark benefit rate value , the investment decision is not adjusted.

[0131] When the net benefit rate is greater than the set benchmark benefit rate value , the investment decision is set to the maximum allowed investment upper limit

[0132] When the net benefit rate is less than the set benchmark benefit rate value , the power producer reduces the investment amount until it completely gives up capacity investment.

[0133] The operating results of the capacity investment simulation model through the mechanism utility evaluation include the system installed capacity in the target year

[0134] Embodiment 10:

[0135] A capacity market hierarchical design method for coordinating mechanism utility incentives and user payments, the main technical content of any one of embodiments 1 to 9, further, the step of feedback correction of the key boundary parameters of each sub-capacity market platform is shown as follows:

[0136] 5.3.1) Set the initial value of the capacity adjustment coefficient and the price adjustment coefficient to 1, the iteration number k = 0, and set the target installed capacity

[0137] 5.3.2) Let the iteration number k = k + 1, calculate the capacity demand parameter X C and the capacity price parameter X P based on the capacity adjustment coefficient C and the price adjustment coefficient P , as shown in the summary formula (37)-(40).

[0138] 5.3.3) Based on the capacity demand parameter X d and the capacity price parameter X i , determine the capacity demand curve F i (Q), run the capacity investment simulation model to obtain the system installed capacity in the target year

[0139] 5.3.4) Calculate the system installed capacity in the target year i and the target installed capacity i .the difference, and determining whether the absolute value of the difference is less than a preset threshold value, if yes, outputting the capacity adjustment coefficient and the price adjustment coefficient if no, entering step 5.3.5).

[0140] 5.3.5) updating the capacity adjustment coefficient and the price adjustment coefficient returning to step 5.3.2).

[0141] the capacity adjustment coefficient and the price adjustment coefficient The updating strategy of the capacity adjustment coefficient

[0142] Embodiment 11:

[0143] Referring to Figures 1 to 4 A capacity market hierarchical design method for coordinating mechanism utility incentive and user payment, comprising the following steps:

[0144] Firstly, a capacity market hierarchical design scheme considering the cost structure difference of multiple subjects is proposed, and power sources with similar cost structures are classified into the same layer of the capacity market bidding platform, so that the power sources with consistent cost recovery demands produce consistent price signals for investment incentive; then, a power source sub-capacity market division decision model considering income balance is proposed, so as to determine the demand of each layer market capacity market; finally, a capacity market parameter optimization scheme considering the mechanism utility incentive target is proposed, so as to further determine the key market boundary parameters of each layer market to meet the expected mechanism utility. The example study based on the Cplex solver in the Matlab software calling the IEEE standard network shows that the hierarchical capacity market design scheme proposed in the application can effectively improve the power carpooling phenomenon, can reduce the capacity payment cost under the premise of ensuring the expected investment incentive effect of the capacity market, effectively alleviate the user payment burden, and realize the effective coordination of mechanism utility incentive and user payment.

[0145] (1) a capacity market hierarchical design scheme considering the cost structure of multiple subjects

[0146] The application is based on the current capacity market mechanism, and proposes a hierarchical capacity market bidding mode. According to the similarity of power supply cost structure, it is divided into multiple sub-capacity markets, forming a hierarchical capacity cost recovery bidding market. The hierarchical capacity market proposed in the application maintains the basic framework of the existing capacity market, and generally adopts a marginal clearing-based capacity price market pricing mechanism to avoid high implicit costs caused by changes in market mechanisms. The proposed method divides power sources with similar cost structures into the same layer, allowing power sources with similar cost structures to bid on the same platform, thereby solving or alleviating the capacity market piggybacking phenomenon, achieving market mechanism utility incentive goals, and reducing excessive capacity market payments. The core of the capacity market hierarchical design scheme is the refinement of the bidding platform, that is, a single capacity market platform is divided into multiple sub-capacity market platforms according to the cost structure difference of power sources. The sub-capacity market platform is organized and operated by the market operation agency in the same period in units of years, and each sub-capacity market platform is auctioned according to the capacity demand target of a certain capacity payment year in the future. Each sub-capacity market platform adopts a centralized bidding mechanism, and the market operation agency assumes the roles of capacity cost payment and capacity cost diversion for the generation side and the user side, respectively. The hierarchical capacity market establishment steps are as follows:

[0147] 1) Determine the total capacity demand: according to the system planning target, combined with the load prediction of the target year, i.e. the system reliability requirement and other related information, determine the total capacity demand of the hierarchical capacity market.

[0148] 2) Decide the sub-capacity market to which the power source belongs and determine the capacity demand of each sub-capacity market: based on the power source sub-capacity market division method considering income balance, divide the power sources participating in capacity into the corresponding sub-capacity market platform, and determine the capacity demand of each sub-capacity market.

[0149] 3) Determine the key market boundary parameters of each sub-capacity market: based on the capacity market parameter optimization method considering the mechanism utility incentive target, determine the key boundary parameters of each sub-capacity market to ensure the realization of the expected incentive effect of the capacity market. The parameter optimization method will realize the configuration of the capacity market parameters by establishing a capacity investment simulation under each sub-capacity market, forming a mechanism utility evaluation according to the mechanism utility incentive target and feeding back to correct the capacity market parameters.

[0150] (2) Capacity market hierarchical decision-making model considering market main body income balance

[0151] The application establishes a power bidding sub-platform division decision model considering income balance to determine various types of power participating sub-capacity market and determine its capacity demand. In a certain period, all members participating in the capacity market are determined only once to belong to the sub-capacity market. For example, for power i, the sub-market to which it belongs is determined when it participates in the hierarchical capacity market for the first time and is announced to the market members. The sub-market participated by the power remains unchanged until the next market operator re-divides the sub-market participated by all market members in the market.

[0152] 1) Power sub-market division decision model

[0153] The power sub-capacity market division decision problem is described by a bi-level optimization model. The upper layer takes the minimization of the total cost of the capacity market as the target, decides the sub-market to which each power belongs and its capacity demand, and passes it to the lower layer model. The lower layer is a sub-capacity market clearing model, which obtains the sub-market clearing result and returns it to the upper layer model.

[0154] 1) Upper layer model

[0155] The objective function of the upper layer model is to minimize the total cost of the hierarchical capacity market, as shown in equation (1).

[0156]

[0157] In the formula: is the total payment cost of the hierarchical capacity market; p and i are the indexes of the sub-capacity market and the power, respectively; and are the clearing price and clearing capacity of power i in sub-capacity market p, respectively.

[0158] The constraints of the upper layer model are as follows:

[0159] a) Capacity market demand constraint: the sum of the capacity demand of each sub-market should meet the total capacity demand, as shown in equation (2).

[0160]

[0161] In the formula: is the total capacity demand of the hierarchical capacity market, which is determined based on the system planning target. is the capacity demand of sub-market p, which is the decision variable of the upper layer model.

[0162] b) Income balance constraint: the income of each power should meet the requirement of balance degree, which is used to limit the degree of power hitchhiking, as follows.

[0163]

[0164] In the formula: η i represents the income rate of power i; represents the average of all power source revenues; N1 is the total number of power sources; is the upper limit of the revenue balance. The power source revenue η i are calculated as shown in equations (4)-(7).

[0165]

[0166] where: CT i respectively represent the total annual revenue of power source i in the energy market, the total annual revenue of power source i in the capacity market, and the total annual cost of power source i; is the initial available capacity of power source i in the target year, including the built capacity and the capacity that is initially planned to be invested and built in the target year; is the total predicted revenue of power source i in the target year in the energy market according to the initial available capacity; CT i p is the total predicted cost of power source i in the target year according to the initial available capacity.

[0167] c) Power source belonging sub-market unique constraint: each power source should only participate in bidding in a certain sub-capacity market, as shown in equation (8).

[0168]

[0169] where: is a 0-1 variable representing the state of power source i belonging to sub-market p, and is a decision variable of the upper model, if power source i participates in the market auction of sub-market p, is 1, otherwise is 0. If the sub-market to which power source i belongs has been determined in the capacity market organized in the previous time, is fixed as a constant and no longer needs to be decided.

[0170] 2. Lower model

[0171] The lower model is the clearing model of each sub-capacity market. Considering that the modeling accuracy of the capacity market has less influence on the decision result of determining the sub-market to which the power source belongs, in order to reduce the calculation complexity, the capacity demand curve in the capacity market is described by using a vertical curve model. For the clearing model of the capacity market of sub-market p, the objective function is the minimization of the capacity purchase cost, as follows.

[0172]

[0173] The corresponding constraint equations are shown in equations (10)-(13).

[0174]

[0175] wherein: formula (10) is a capacity balance constraint, is a dual multiplier corresponding to the capacity balance constraint; formula (11) is an upper and lower limit constraint of the winning capacity in sub-market p, and respectively are upper and lower limits of the winning capacity of power supply i in sub-market p, and respectively are dual multipliers corresponding to the upper and lower limit constraints; and respectively are upper and lower limits of the capacity of power supply i participating in the capacity market; and respectively are dual multipliers corresponding to formula (12) and formula (13).

[0176] 2) Model transformation and solution

[0177] For the above double-layer optimization model, the double-layer model is transformed into a single-layer mathematical optimization problem with equilibrium constraints based on the KKT condition, and then linearized to transform into a mixed integer linear programming model for solution. Based on the KKT condition, the constraint formula (10)-(13) of the original problem, the Lagrange function coefficient constraint formula (14), the inequality constraint coefficient constraint formula (15)-(16), and the complementary relaxation constraint formula (17)-(18) are equivalent to replace the capacity market clearing model of the lower layer sub-market p. The KKT conditions of each sub-market model in the lower layer are added as constraints to the upper layer model, and the double-layer optimization model is transformed into a single-layer MPEC model.

[0178]

[0179] In the transformed MPEC model, the complementary relaxation conditions formula (17)-(18), the objective function formula (1), and the calculation formula (6) of the capacity revenue used in the revenue balance constraint are nonlinear, which affects the model solution. The big M method is used for linearization, and the complementary relaxation conditions formula (17)-(18) are transformed into linear convex constraints as formula (19)-(22).

[0180]

[0181] wherein: is a 0-1 variable; is a large enough positive number.

[0182] The nonlinear terms in the objective function formula (1) and the capacity revenue calculation formula (6) are transformed into linear terms by piecewise linearization technology The corresponding constraints are shown in formula-.

[0183]

[0184] where q is the index of the linear segment; is the value of the segment point q; is the distance between two segment points; is the introduced 0-1 variable; is the introduced continuous variable, when is is the capacity market clearing price M q is a large enough positive number.

[0185] After linearization, the MPEC model is converted into a MILP model, which is solved by a solver to obtain the results of the division of each power supply submarket and the capacity demand of the submarket.

[0186] (3) Capacity market parameter optimization model for utility incentive target of security mechanism

[0187] After obtaining the participants and capacity demand of each submarket, the key market boundary parameters of each submarket need to be further determined to meet the expected utility of the capacity market mechanism. The capacity demand curve parameters to be decided in the market are the capacity demand parameter X C and the capacity price parameter X P . According to the capacity demand curve parameter optimization principle, to minimize the capacity market payment cost while meeting the system capacity adequacy requirement, the configuration point of the capacity demand curve parameter should be the point that makes the investment capacity incentive consistent with the capacity adequacy requirement, and the corresponding capacity market generates sufficient and not too high investment incentive signal strength. The present application adopts the parameter decision-making idea of mechanism utility closed-loop feedback to determine the capacity market boundary parameters, that is: first, simulate the capacity investment under the capacity market, then evaluate the mechanism utility and feedback correct the capacity market boundary parameters based on the evaluation results until the expected mechanism utility target is met. To incentivize the capacity investment amount consistent with the expected capacity investment, the capacity demand curve parameter decision-making method based on the closed-loop feedback of investment incentive utility needs to be applied to each submarket in the hierarchical capacity market to determine the capacity demand curve parameters of each submarket. Specifically, it includes the capacity investment simulation link and the parameter feedback correction link.

[0188] 1) Capacity investment simulation link

[0189] 1. Initial investment decision module

[0190] The initial investment willingness of investors is related to the previous year's income. For a power supplier its initial investment capacity in the yth year is calculated as shown in formula (27).

[0191]

[0192] wherein: represents the power producer The net profit of the previous year is the sum of the net profits of all the units belonging to the power producer as shown in equation (28)

[0193]

[0194] When the power producer has a loss in the previous year, i.e. the power producer tends to invest according to the set minimum initial investment decision amount . The value of the minimum initial investment decision amount may be 0, i.e. no investment.

[0195] When the power producer has a profit in the previous year, i.e. the power producer tends to participate in the capacity market auction according to the maximum investable capacity in order to obtain more economic support. The maximum investable capacity is determined by the allowed investment upper limit and the investment willingness upper limit . The allowed investment upper limit is the maximum capacity allowed for investment of the power producer limited by resources such as funds, sites, etc., and the investment willingness upper limit is the investment willingness of the power producer according to the current capacity scale and the profit in the previous year, as shown in equation (29).

[0196]

[0197] In the equation: is the investment rate; is the total capacity scale of the power producer in the yth year. The investment rate in equation (29) and the net profit in the previous year are coupled, and the relationship is as follows:

[0198]

[0199] In the equation: is the total fixed cost of the unit type invested by the power producer ; is the coupling coefficient, reflecting the investment risk preference of the power producer, the larger the coupling coefficient, the higher the acceptance of the unit to risk.

[0200] 2. Capacity market bidding decision module

[0201] Assuming that the capacity market in the yth year corresponds to the delivery year y+m, the power producer will estimate the total revenue according to the predicted load in the y+m year and the overall capacity situation of the network, combined with historical data prediction or by simulating the annual energy market. Considering that the power producer has limited information about the investment decision-making information of other market participants, it is assumed that other investors invest at a certain estimated investment rate. At the same time, considering unknown uncertainty factors and the risk aversion characteristics of the unit, it is considered that the power producer will make a conservative estimate of the load prediction, as shown in equation (31).

[0202]

[0203] In the formula: Pl y represents the load in the yth year; represents the predicted load in the y+m year; f is the actual growth rate of the load; g is the prediction error percentage, which is negative, and is related to the risk aversion degree of the investor.

[0204] The unit's bid in the capacity market is assumed to be reported according to the estimated cost gap. The cost gap refers to the cost that the unit is unable to cover in other markets and expects to recover in the capacity market. If the unit has fully recovered the cost, it is considered that the bid is 0. The capacity market bid of unit i in the yth year is as shown in equation (32).

[0205]

[0206] In the formula: is the estimated annual total cost of unit i in the y+m year; is the estimated market income of unit i in the y+m year other than the capacity market; is the unit capacity of unit i.

[0207] 3. Capacity market clearing module

[0208]

[0209] In the formula: Q is the capacity; F d (Q) represents the capacity demand curve, i.e., the corresponding capacity price under different capacity demand levels; is the winning capacity of unit i in the yth year capacity market; I C_market,y is the set of units participating in the yth year capacity market auction; is the upper limit of the winning capacity of unit i. According to the capacity market clearing price The annual revenue of unit i in the yth year capacity market is as shown in equation (36).

[0210]

[0211] 4. Final investment decision-making module

[0212] Final investment decision of power producers and initial investment decision It is assumed that the relationship is linear and positive within a certain range. When the net benefit rate is equal to a certain reference benefit rate value , it is considered that the power producer just obtains the expected profit, and the investment decision is not adjusted, i.e. the initial investment is made according to the initial investment amount; when the net benefit rate is greater than , the power producer obtains more than expected profit and increases the capacity investment according to the linear relationship. Considering the existence of fund, site and investment risk restrictions, a maximum allowed investment upper limit is set. When the net benefit rate is less than , the power producer will reduce the investment amount until the capacity investment is completely abandoned.

[0213] 2) Parameter feedback correction link

[0214] First, the capacity demand curve adjustment coefficient linearly corrects the curve parameters, as shown in equations (37)-(38).

[0215]

[0216] In the formula: and are the capacity adjustment coefficient and the price adjustment coefficient; and are the capacity demand parameter value and the capacity price parameter value based on the planning method, respectively determined by equations (39) and (40).

[0217]

[0218] In the formula: is the capacity demand parameter value; L max is the peak load of the capacity delivery year; IRM is the system reserve margin; e is the average equivalent availability rate of units in the region; is the capacity price parameter value determined according to the current planning method; C G is the total cost of the reference unit, which is the sum of the annualized investment cost, the annual operation and maintenance cost, and the annual variable cost; I E is the estimated annual income of the reference unit in the capacity delivery year in the energy market; I A is the estimated annual income of the reference unit in the capacity delivery year in the ancillary service market; Q g is the unit capacity.

[0219] The feedback correction strategy of the two adjustment coefficients, the capacity adjustment coefficient and the price adjustment coefficient is introduced below. After the final values of the adjustment coefficients are obtained, the capacity demand curve parameters are determined according to equations (37)-(38).

[0220] The feedback correction strategy of the adjustment coefficient is specifically as follows:

[0221] Step 1: setting and The initial value is 1;

[0222] Step 2: obtaining X according to formula (37)-(38) C and X P , determining the capacity demand curve;

[0223] Step 3: obtaining the system installed capacity in the target year based on the capacity investment simulation method in the capacity-energy market environment

[0224] Step 4: comparing and The difference, if the absolute value is less than a predetermined threshold, it is considered that the investment incentive utility target is met, and and is output, otherwise, iterative correction is needed, as shown in step 5.

[0225] Step 5: iterative updating and and returning to step 2, the updating strategy is as follows:

[0226] The adjustment coefficient should meet formula (41) to guarantee the capacity market investment incentive utility:

[0227]

[0228] In the formula: x refers to or The present application assumes that the two adjustment coefficients are updated according to the same iterative strategy; Q initial is the initial system capacity; Q I (x) is the simulated system investment increment, that is, the newly built capacity minus the decommissioned capacity. The updating strategy of the adjustment coefficient is shown in formula (42).

[0229]

[0230] In the formula: k is the iteration number; f'(x (k) ) is the derivative of f(x) with respect to x, that is, the derivative of Q I (x) with respect to x. f(x) is difficult to analytically represent, and an iterative method is used to calculate the approximate f'(x (k) ), as shown in formula (43).

[0231]

[0232] In the formula: is the change rate K at the initial iterationS the set value of the parameter.

[0233] Embodiment 12

[0234] A capacity market hierarchical design method for coordinating the utility incentive of the mechanism and the payment of the user, the main technical content of which is seen from Embodiment 11, further, (1) scenario construction: the number of layers of the hierarchical capacity market is set to 3, and a hierarchical capacity market containing three sub-markets is established. Compare the simulation results of the single capacity market (referred to as "single-layer market"), the hierarchical capacity market containing two sub-markets (referred to as "double-layer market") and the hierarchical capacity market containing three sub-markets (referred to as "three-layer market"), and analyze the influence of the number of layers on the utility of the capacity market mechanism.

[0235] (2) Analysis of market clearing results

[0236] 1) Sub-market division results: the division results of the three-layer market are shown in Table 1. As can be seen from the table, the wind power source is divided into sub-market 1, the A-class coal-fired power source is divided into sub-market 2, and the B-class and C-class coal-fired power sources are divided into sub-market 3. It shows that when the number of layers of the capacity market increases from two to three, the new energy is still divided into a sub-market due to the significant difference in the cost structure with the flexible power source, the cost recovery appeal of the A-class power source is less consistent than that of the B and C-class power sources, and it is divided into another sub-market, so as to weaken the power carpooling phenomenon caused by the same bidding of different types of coal-fired power sources.

[0237] Table 1: Sub-market division results of the three-layer market

[0238]

[0239] 2) Power revenue balance analysis: Table 2 shows the power revenue rate of each type of power source in the single-layer market, double-layer market and triple-layer market in the 3rd year. Since the capacity market organization conducts capacity auction for the next two years, Table 2 actually corresponds to the system status in the first year of simulation. At this time, no new capacity has been built, so the system status under the three capacity market modes is completely consistent. The results show that with the increase of the number of submarkets, the revenue rate of each type of power source gradually decreases. When the number of submarkets of the layered capacity market increases from two to three, the A-type power source is divided into a separate submarket, and its revenue rate decreases from 1.181 to 1.017, indicating that by avoiding A-type power sources and other coal-fired power sources bidding on the same platform, A-type power sources are prevented from hitchhiking, so that the capacity compensation they obtain can cover the cost and not be excessive. At the same time, the results show that the standard deviation of the revenue rate in the layered capacity market mode is lower than that in the single capacity market mode, indicating that by establishing multiple submarkets, the revenue of each type of power source can be more balanced, and the degree of power hitchhiking can be reduced. Table 3 shows the power revenue rate in the single-layer market, double-layer market and triple-layer market in the 16th year. The results show that in the single capacity market mode, due to the excessive capacity compensation obtained by new energy sources in the early stage, resulting in excessive investment, in the later stage of the simulation, new energy sources have difficulty in cost recovery. In the 16th year, the revenue rate of new energy sources decreases to 0.813, while flexible power sources hitchhike new energy sources in the opposite direction, resulting in excessively high revenue. In the layered capacity market mode, the situation of excessively high revenue rate caused by hitchhiking is weakened, and the difference between the revenue rates of each power source is reduced. The triple-layer market provides more reasonable capacity compensation for each type of power source than the double-layer market, enabling each type of power source to achieve cost recovery and not excessive profit, and the revenue rate of each power source is more balanced. The above results show that the establishment of a layered capacity market containing multiple submarkets can better meet the differentiated capacity compensation needs of power sources with differentiated cost structures, making the revenue of each type of power source more reasonable. Increasing the number of submarkets can further improve the balance of power revenue and reduce the degree of power hitchhiking.

[0240] Table 2 Power revenue rate under different capacity market modes in the 3rd year

[0241]

[0242] Table 3 Power revenue rate under different capacity market modes in the 16th year

[0243]

[0244] 3) Capacity market cost analysis: Table 4 shows the capacity market cost of single-layer market, double-layer market and triple-layer market. The results show that in the first 13 years, the triple-layer market further reduces the capacity market cost on the basis of the double-layer market. For example, in the first year, the operation cost of the triple-layer market is 42.88 billion yuan, which is 31.93% lower than that of the double-layer market and 56.40% lower than that of the single-layer market. In the 14th to 15th years, the single-layer market cannot achieve the full cost recovery of power supply and cannot meet the investment incentive expectations. The double-layer market and the triple-layer market can still meet the investment incentive target, but due to the high new energy penetration rate of the system at this time, the capacity market cost increases significantly to ensure system capacity adequacy. In the 15th year, the capacity market cost of the triple-layer market is slightly higher than that of the double-layer market, which is because in the early stage of simulation, the power supply in the triple-layer market has fewer opportunities to earn extra profits by hitchhiking, so the installed capacity increment in the triple-layer market is smaller than that in the double-layer market. By the later stage of simulation, the triple-layer market needs to increase capacity compensation to increase investment incentive, so as to encourage more capacity investment to ensure capacity adequacy.

[0245] Table 4 Capacity market cost under different capacity market modes

[0246]

[0247] At the same time, in the entire simulation process of this example, the power supply sub-market division result remains unchanged. If the sub-market division decision is updated in a timely manner according to the system capacity evolution, it is beneficial to produce more accurate investment incentive signals and improve the economy in the later stage of simulation. Overall, the example results show that increasing the number of sub-markets in the layered capacity market can further improve the power supply hitchhiking phenomenon and reduce the capacity payment cost. However, too many sub-markets will increase the difficulty of sub-market division decision calculation and the difficulty of organizing market clearing, so the number of sub-markets needs to be weighed when determining. In addition, the update period of the sub-market division decision can be further studied to further improve the effectiveness of the layered capacity market mode.

Claims

1. A capacity market hierarchical design method that coordinates mechanism utility incentives with user payments, characterized in that, The method comprises the following steps: 1) determining a total capacity demand according to a power system planning target and in combination with load prediction in a target year; 2) constructing a plurality of sub-capacity market platforms; 3) constructing a power source sub-market division decision model based on the total capacity demand; 4) solving the power source sub-market division decision model, classifying power sources with different cost structures into corresponding sub-capacity market platforms, and determining the capacity demand of each sub-capacity market platform; 5) determining key boundary parameters of each sub-capacity market platform based on the power sources and the capacity demand classified by each sub-capacity market platform through a capacity market parameter optimization method considering mechanism utility incentive targets; 6) feeding back the key boundary parameters of each sub-capacity market platform to each sub-capacity market platform to construct a hierarchical capacity market coordinating mechanism utility incentive and user payment.

2. The method of claim 1, wherein, The power sources include wind power sources, coal-fired power sources, and photovoltaic power sources.

3. The method of claim 1, wherein, The power source sub-market division decision model includes a double-layer optimization model.

4. The method of claim 3, wherein, When the power source sub-market division decision model is a double-layer optimization model, the objective function of the upper-layer optimization model is as follows: In the formula, is the total payment cost of each sub-capacity market platform; p, i are the indexes of sub-capacity market, power supply, respectively; is the clearing price, clearing capacity of the i-th power supply in the p-th sub-capacity market, respectively. The objective function of the lower-layer optimization model is as follows: wherein is the capacity offer for the i-th power source.

5. The method of claim 3, wherein, When the power source sub-market division decision model is a double-layer optimization model, the constraint conditions of the upper-layer optimization model include a capacity market demand constraint, a revenue equilibrium constraint, and a power source belonging sub-market unique constraint. The constraint conditions of the lower-layer optimization model include a capacity balance constraint and a winning capacity upper and lower limit constraint.

6. The method of claim 5, wherein, The capacity market demand constraint is as follows: where p is an index of sub-capacity markets; is the total capacity demand; is the capacity demand of the pth sub-capacity market. The revenue equilibrium constraint is as follows: where i is the index of the power supply; represents the average of all power supply yields; N1is the total number of power supplies; is the upper limit of yield balance degree; wherein the yield η of the i-th power source i As shown below: wherein, CT i respectively represent the total annual energy market revenue, total annual capacity market revenue, total annual cost of the i-th power source; is the initial capacity provided by the i-th power source in the target year; is the total annual energy market revenue predicted for the i-th power source in the target year; is the total cost predicted for the i-th power source in the target year; i p is the total annual energy market revenue predicted for the i-th power source in the target year; is the total cost predicted for the i-th power source in the target year; respectively represent the clearing price, clearing capacity of the i-th power source in the p-th sub-capacity market. The power source belonging sub-market unique constraint is as follows: In the formula, is a state variable representing that the ith power source belongs to the pth sub-capacity market, and if the ith power source belongs to the pth sub-capacity market, then On the contrary, then The capacity balance constraint is as follows: In the formula, is the dual multiplier corresponding to the capacity balance constraint; The winning capacity upper and lower limit constraint is as follows: In the formula, respectively, the upper and lower limits of the winning capacity of the ith power supply in the pth sub-capacity market, is the dual multiplier corresponding to the upper limit constraint of the winning capacity of the ith power supply in the pth sub-capacity market; respectively, the upper and lower limits of the capacity of the ith power supply participating in the sub-capacity market; is the dual multiplier of the unique constraint of the power supply belonging to the sub-market.

7. The method of claim 3, wherein the method further comprises: When the power source sub-market division decision model is a double-layer optimization model, the steps for solving the power source sub-market division decision model are as follows: 4.1) converting the double-layer optimization model into a single-layer MPEC model based on the KKT condition, i.e., adding the KKT condition of the lower-layer optimization model as a constraint to the upper-layer optimization model, as follows: where p, i are the indices of sub-capacity market, power source, respectively; is the dual multiplier corresponding to the capacity balance constraint; is the dual multiplier corresponding to the upper and lower bound of the winning capacity; are the upper and lower bound of the winning capacity of the ith power source in the pth sub-capacity market, respectively; is the clearing capacity of the ith power source in the pth sub-capacity market; is the capacity offer of the ith power source; 4.2) converting the single-layer MPEC model into a MILP model based on linearization processing, as follows: wherein is a 0-1 variable; is a positive number; q is the index of the linear segment point; is the clearing price of the ith power source in the pth sub-capacity market; is the capacity corresponding to the segment point q; is the distance between any two consecutive segment points; is a 0-1 variable; is a continuous variable, when , is the clearing price M q is a positive number; 4.3) solving the MILP model by using a solver, classifying power sources with different cost structures into corresponding sub-capacity market platforms, and determining the capacity demand of each sub-capacity market platform.

8. The method of claim 1, wherein the method is characterized by, In step 5), the steps for determining the key boundary parameters of each sub-capacity market platform are as follows: 5.1) constructing a capacity investment simulation model; 5.2) running the capacity investment simulation model based on the power sources and the capacity demand classified by each sub-capacity market platform, and evaluating the running result of the capacity investment simulation model through mechanism utility; 5.3) feeding back and correcting the key boundary parameters of each sub-capacity market platform based on the evaluated running result.

9. The method of claim 8, wherein, The capacity investment simulation model includes an initial investment decision module, a capacity market bidding decision module, a capacity market clearing module, and a final investment decision module. The initial investment decision module is as follows: In the formula, y is the year index; For e-commerce indexing; Indicates the first The initial investment capacity of a power generator in year y; Indicates the first Net profit of a power generator in year y-1; i is the index of the power source; For the first A collection of power supplies for individual power generators; v i,y-1 This represents the net profit of the i-th power source in year y-1. These are the upper and lower limits for permitted investment; This represents the upper limit of investment intentions; Investment rate; For the first The total capacity of a power generation company in year y; For the first The total fixed cost of the power supply type invested by each power generator; The coupling coefficient; The capacity market bidding decision module is as follows: where m is the delivery period; Pl y represents the load in year y; represents the predicted load in year y+m; f is the actual growth rate of load; g is the prediction error percentage; is the estimated total cost of the ith source in year y+m; is the estimated market revenue of the ith source in year y+m, excluding the capacity market; is the capacity of the ith source; is the capacity market price of the ith source in year y; The capacity market clearing module is as follows: where Q is the capacity; F d (Q) denotes the capacity demand curve; is the winning capacity of the ith generator in the yth year capacity market; I C_market,y is the set of generators participating in the yth year capacity market auction; is the upper bound of the winning capacity of the ith generator; is the capacity market clearing price; is the yth year capacity market annual revenue of the ith generator; θ denotes the capacity market utility; The final investment decision module is as follows: When the net yield is at the set benchmark yield value No adjustment is made to the investment decision; When the net yield is greater than a set benchmark yield value the investment decision sets a maximum allowed investment cap When the net benefit rate is less than the set reference benefit rate value the power producer reduces the amount of investment until completely giving up capacity investment; The running result of the mechanism utility evaluation capacity investment simulation model includes system installed capacity in the target year 10. The method of claim 9, wherein, The step of said feedback correcting the key boundary parameters of each sub-capacity market platform is shown as follows: 5.3.1) Setting the capacity adjustment factor and the price adjustment factor with initial values of 1, iteration number k = 0, and setting the target installed capacity 5.3.2) Let the iteration number k = k + 1, and calculate the capacity adjustment factor and the price adjustment factor based on the capacity demand parameter X C and the capacity price parameter X P as follows: wherein, is the capacity requirement parameter value; L max is the peak load in the capacity delivery year; IRM is the system reserve margin; e is the average equivalent availability of the in-region resources; is the capacity price parameter value; C G is the total cost of the reference resource; I E is the estimated annual energy market revenue of the reference resource in the capacity delivery year; I A is the estimated annual ancillary service market revenue of the reference resource in the capacity delivery year; Q g is the resource capacity; 5.3.3) Based on the capacity demand parameter X C and the capacity price parameter X P , determine the capacity demand curve F d (Q), run the capacity investment simulation model to obtain the system installed capacity 5.3.4) Calculate the system installed capacity of the target year and the difference between the target installed capacity and the difference between the target installed capacity and the difference between the target installed capacity If not, go to step 5.3.5) 5.3.5) Update the capacity adjustment factor and the price adjustment factor Go back to step 5.3.2); The capacity adjustment coefficient And the price adjustment coefficient The update strategy is as follows: where x (k) represents the capacity adjustment factor at the kth iteration or price adjustment factor x (k+1) represents the capacity adjustment factor at the k+1th iteration or price adjustment factor x (k-1) represents the capacity adjustment factor at the k-1th iteration or price adjustment factor f(x (k) ) represents the adjustment factor updating strategy; f'(x (k) ) represents the derivative of the updating strategy f(x (k) ) with respect to x (k) ; represents the system installed capacity in the target year; Q initial represents the initial system capacity; Q I (x (k) ), Q I (x (k-1) ) represent the system investment increment at the kth, k-1th iteration, respectively; K S (k) represents the change rate at the kth iteration represents the change rate at the initial iteration.

Images