Research on the capacity cost allocation and the electricity capacity price optimization method for a power system based on the BARY‐GA algorithm

Under the new power system, a high proportion of new energy is widely connected to the power grid, and it is necessary to increase investment in centralized and distributed energy storage, flexible resource regulation, and transmission and distribution grids, resulting in an increase in power system costs. The current electricity capacity price does not reflect the economic value of the added system adequately, and flexible capacity is needed to ensure the safety of the power grid under the new power system. In the context of the transformation of the new power system and investment function, it is increasingly difficult to recover the costs according to the traditional electricity price. It is necessary to establish a coordinated development mechanism for the electricity capacity price and the electricity price that recovers the fixed costs according to the investment function to ensure a reasonable return on investment in flexible resources and to provide effective investment signals. Therefore, this work first studied and proposed a mechanism for the formation of capacity and electricity prices based on the proportion of allowed income and fixed costs at different voltage levels and calculates the level of electricity capacity (demand) prices and electricity consumption prices. Then, using the BARY curve, genetic algorithm, and clustering of user load rates, a collaborative mechanism between the electricity capacity price and the electricity price was designed. With the goal of optimizing the electricity capacity price and considering constraints such as the flexibility and reliability of the new power system, the ratio of the capacity cost allocated to the electricity capacity price by voltage level users to the capacity cost allocated to the electricity price was calculated, and the optimal combination of electricity capacity price and electricity price for different load‐rate intervals was determined. Finally, the effectiveness of the model was verified through numerical simulations. On the basis of the investment characteristics of the new power system, suggestions for a capacity pricing mechanism under the new power system were proposed from the perspectives of transfer capacity function, transfer electricity function, grid reliability, and environmental and social benefits.


| INTRODUCTION
Since 2015, China's power system reform has entered an accelerated stage, and the development of the power industry is facing some problems that need to be solved through reform.One of them is that the price relationship has not been correctly determined, and the marketoriented pricing mechanism has not yet been fully formed.In the current situation of an unreasonable electricity price formation mechanism, establishing a grid electricity price formation mechanism that is suitable for the power generation process is the key point to rationalize the price relationship. 1The two-part grid electricity price can reasonably compensate for the fixed costs of power generation enterprises and effectively guide and encourage the power investment and technological innovation of power generation enterprises, which is in line with the development direction of China's power system reform and is also suitable for the current level of power market construction. 2n 2020, China issued the "Pricing Measures for Regional Power Grid Transmission Prices" (referred to as the "Pricing Measures"), which clearly stipulated that in addition to following the principles of improving the power grid efficiency, reasonably sharing costs, and regulating pricing behavior, the equation of regional power grid transmission prices should also promote electricity trading.The main manifestation is that the theory of two-part electricity pricing promoting electricity trading has not been fully understood and applied, and the electricity capacity pricing (fee) mechanism, such as the recovery of operation and maintenance fees through capacity and electricity fees, does not fully reflect the actual cost characteristics. 3In terms of functional research on power grid transmission prices, in 1992, the United States enacted the Energy Policy Act, which stipulated that power grid companies must provide transmission services. 4In 1996, the Federal Energy Regulatory Commission (FERC) issued decrees 888 and 889, 5,6 stipulating that the power generation and transmission must be functionally separated, and the power grid must be fair and open to all market players, while establishing open access transmission prices and ancillary service prices.Zhang et al. 7 believe that reasonable transmission prices are one of the important factors in promoting the optimal allocation of power resources, encouraging market members to actively participate in power trading, and even promoting the process of power reform.Jing et al. 8 explored the crossprovincial trading mechanism of China's unified large market and stated that the power grid is a platform and channel for electricity trading, and reasonable pricing can promote the opening of the power grid, thereby promoting electricity trading.Kurzidem and Andersson 9 stated that the transmission price is closely related to social welfare, and the management of a transmission price ceiling should be conducive to maximizing social welfare.
Regarding the research on the pricing mechanism of two-part electricity pricing, Ye 10 stated that for goods with a high proportion of fixed costs, while meeting the cost compensation requirements, two-part pricing has greater efficiency than average cost pricing and can achieve greater social benefits.Vogelsang 11 introduced a two-part electricity price into transmission service fees and reflected capacity costs in the form of fixed fees, grid fees, and auxiliary service fees in variable fees.Schlereth et al. 12 and Zhang and Ma 13 provided an electricity capacity price model calculated based on fixed costs and an electricity price model calculated based on variable costs, respectively, and compared it with a single electricity price to demonstrate the effectiveness of the two-part electricity price in mobilizing power generation, grid dispatch, and other aspects.Xiao and Zhang 14 optimized the two-part tariff scheme from two aspects expanding the scope of users applicable to the current two-part tariff and increasing the proportion of basic electricity charges.Using the control variates, they discussed the capacity tariff and kilowatt-hour tariff at the international level to increase the proportion of the capacity tariff to 30%-50% under the premise that the user electricity charges remain unchanged.Wang et al. 15 further allocated the power grid operation and maintenance fees based on the fixed cost and variable cost attributes using the efficiency value method and found that the proportion of operation and maintenance fees LIU ET AL.
| 169 (including labor costs) related to the transmission of electricity is 14%, while the proportion of operation and maintenance fees (including labor costs) unrelated to the transmission of electricity is 86%.
The research on the allocation method of capacity costs is different from the pricing method of the electricity transmission and distribution prices under natural monopoly systems.There are many pricing methods for electricity transmission and distribution prices in the spot electricity market, such as the stamp method, contract path method, peak load responsibility method, boundary flow method, megawatt kilometer method, and so forth. 16,17Regardless of the method used, the allocation method of the grid capacity costs among market entities is the core and key content.The transmission prices in the UK and US electricity markets mainly use single electricity capacity prices.Li and Xu 18 analyzed and found that the capacity cost allocation in the UK electricity market, the PJM electricity market in the United States, and the Texas electricity market all adopt the peak load responsibility method, which is to allocate the grid capacity cost according to the proportion of the highest electricity load of electricity users during the peak load period of the system.This paper revealed the mechanism through which the resulting transmission price signal reflects the degree of demand for power grid and transmission congestion of users in different regions during the peak period of power consumption.Li et al. 19 proposed a two-part transmission pricing scheme for incremental parts based on hierarchical transmission curves, determined the ratio between capacity and electricity charges, and used the peak load responsibility method to allocate the capacity costs.
It has been found that research on the mechanism of the transmission and distribution electricity prices and the allocation of capacity costs has achieved certain results, but from the actual implementation of electricity prices in various countries, there are still some problems, mainly manifested in two aspects: first, the proportion of the recovered electricity fees through electricity capacity prices of the total electricity fees is much smaller than the proportion of the fixed costs in the transmission and distribution costs, and the electricity capacity prices are at a relatively low level.The second reason is that the load-rate factor was not considered, resulting in electricity prices not accurately reflecting the power supply costs of users with different load rates.This leads to insufficient resource utilization in the transmission and distribution grid, and unfair allocation of the transmission and distribution costs among users.In view of this, this paper first calculates the load rate and simultaneity rate of all kinds of users based on the load characteristic data of user groups and calculates the proportion of the capacity cost in the current capacity price and electricity price based on the regional power sales data and the current transmission and distribution price level.Then, a collaborative allocation model for capacity and electricity prices based on the BARY-GA algorithm is constructed, and the relevant calculations for capacity cost allocation are carried out.The BARY curve principle is used to calculate the proportion of the capacity cost allocated by users of each voltage level to the capacity price and the capacity cost allocated to the electricity price.At the same time, in response to the constraints of grid flexibility construction, grid reliability, and grid sustainable development derived from the analysis of the typical new investment in the power grid brought by the current electricity capacity price and the current electricity price, the GA algorithm is used to calculate the electricity capacity price and electricity quantity's electricity price at all levels of load-rate conditions.Finally, the effectiveness of the model is verified through numerical simulation, and suggestions for a capacity pricing mechanism in a new power system are proposed.

| A CAPACITY COST ALLOCATION MODEL FOR THE COORDINATION OF THE CAPACITY PRICE AND THE ELECTRICITY PRICE
At present, the cost recovery of China's electricity capacity prices mainly relies on the basic electricity fees charged to large industrial users.In the coordination of the capacity cost allocation between the capacity pricing and the kilowatt-hour pricing, based on the BARY curve, this paper decomposes the capacity cost into two parts: one part is a fixed cost unrelated to the load-rate and utilization hours, which is transferred to the capacity pricing for recovery.The other part is related to the load-rate and utilization hours and is transferred to the electricity price.The BARY curve is an empirical curve describing the mutual relationship between the simultaneity rate and load rate.Generally speaking, the higher the load rate, the higher the rate.

| BARY curve principle
(1) Load rate, dispersion rate, and simultaneity rate The load rate refers to the proportion of the average power load of the maximum power load during a specific time period (day, month, and year) and is an important indicator for evaluating the stability of a power system's operation. 20The calculation equation for the load rate is where L ave is the average value of the load curve for a specific time period, and L max is the maximum value of the load curve for a specific time period.In a specific project, when the load-rate value is less than and infinitely close to 1, it indicates that the utilization rate of the power equipment is optimal.The dispersion factor is the reciprocal of the simultaneity rate.The calculation equation is where CF is the simultaneity rate, P max is the overall maximum load of a group of users over a certain period of time, P imax is the maximum load of user i during that period, and DF is the dispersion rate.
The load simultaneity rate depends on the shape of the user's daily load curve or daily load rate, the relative amplitude of the maximum load for each user, and the statistical cross correlation between the user randomness and the user load.The simultaneity rate has a significant impact on the user power supply costs, mainly on fixed costs. (

2) An empirical equation for the BARY Curve
The BARY curve is a curve that describes the relationship between the simultaneity rate and the load rate and is characterized by the empirical equation of the BARY curve. 21The empirical equation for the BARY curve is as follows: where CF is the simultaneity rate, LF is the load rate, and α is an unknown parameter to be estimated that is less than zero.CF and LF can be calculated from the collected sample data, and α represents the estimated values that can be obtained using a linear regression model using Eviews.The BARY curve showing the relationship between the simultaneity rate and the load rate is shown in Figure 1.
(3) Coupling relationship between the user load rate, load simultaneity rate, and system load rate Due to the fact that the load curve of the power system is obtained by overlaying the load curves of different users, and the system load rate is calculated based on the system load, the user load rate is not directly related to the system load rate or does not have a definite relationship but rather forms the basis for calculating the system load rate through the superposition of load curves. 22ccording to the definition of the system load rate, the superposition of load curves for different users can obtain the system load curve.At the same time, since the average load of the system is equal to the sum of the average load values of each user, the relationship between the user load rate, load simultaneity rate, and system load rate can be obtained.The load simultaneity rate of the power system is equal to the ratio of the average load rate of all users in the power system to the overall system load rate, as shown in Equation (5).
where CF represents the load simultaneity rate of the power system, P m represents the maximum load of the system, P mi represents the maximum load of user i, P represents the average load of the system, LF represents the system load rate, and LF ave represents the average load rate of all users in the power system.From this, the coupling relationship between the user load rate and the system load rate is obtained.This coupling effect is manifested as follows: the system load rate can only determine the capacity cost and cannot determine the user electricity cost or the final user electricity price.The user load rate does not directly affect the capacity cost but can only determine the final electricity price.It can also be seen that this coupling effect is achieved through the system load simultaneity rate.It is precisely because of the existence of system load simultaneity rate that the same user load rate may form different system load rates, resulting in different effects of the system load rate and user load rate on cost and price.
When allocating capacity costs between electricity capacity prices and kilowatt-hour electricity prices, the first consideration should be given to the load characteristics of users with different electricity consumption characteristics, and the relationship curve between the simultaneity rate and load rate, namely, the BARY experience curve, should be determined.On the basis of the analysis of the load characteristics of users at different voltage levels, this article first calculates the load rate and simultaneity rate of each voltage level user group: where LF sys represents the system load rate of a certain voltage level over a period of time, L sys,ave represents the average load rate of the system at that voltage level during this period, and L sys,max represents the maximum load of the system at that voltage level during this period.CF sys represents the load simultaneity rate of the system at this voltage level during this period, while L user,max,i represents the maximum load of the ith user at this voltage level during this period.
On the basis of the calculation results of the load rate and simultaneity rate for a certain voltage level, we determine the BARY curve for that voltage level.On the basis of the load rate and simultaneity rate of each month over a period of time, we obtain multiple sets of load-rate simultaneity rate data for this voltage level, then fit the BARY curve to obtain the α value, and determine whether the solution result is reasonable.On the basis of the calculation results of the load rate and the simultaneity rate for a certain voltage level, we determine the BARY curve for that voltage level.On the basis of the load rate and simultaneity rate of each month over a period of time, we obtain multiple sets of load-rate simultaneity rate data for this voltage level, then fit the BARY curve to obtain the α value, and determine whether the solution result is reasonable.The parameter α calculation equation is as follows: where α is the BARY curve parameter that needs to be obtained in this study, LF sys represents the system load rate of a certain voltage level over a period of time, and CF sys represents the load simultaneity rate of the system at that voltage level during that period.
On the basis of the calculation results of parameter α above, the capacity cost allocation proportion coefficient in the electricity capacity price and the electricity price can be determined.On the basis of this, the capacity cost allocation proportion coefficient can be calculated for each load-rate interval.
where e i represents the proportion coefficient of the capacity cost allocation of the electricity capacity price, CF i represents the load simultaneity rate of the ith class of users, α is the unknown parameter to be estimated that is less than zero, and LF i represents the load rate of this class of users.
As shown in Figure 2, if the electricity consumption characteristics of a certain type of power user are known, they can be tangent based on their load rate, simultaneity rate, and BARY curve to obtain the intercept point A, where the ordinate of point A is the value of e i .
F I G U R E 2 Example of the proportional coefficient of the capacity cost allocation of the capacity tariff.

| The capacity cost allocation model of the BARY curve
(1) Capacity cost allocated to capacity pricing: (10)   where C total represents the total capacity cost that needs to be recovered from the electricity bill under the new power system, C cap represents the capacity cost recovered through the electricity capacity price, and e i is the proportion coefficient of the capacity cost allocation of the electricity capacity price.(2) Capacity cost allocated to electricity pricing: where C ele represents the capacity cost recovered through electricity pricing and the load simultaneity rate of the CF i system.e i is the proportion coefficient of the capacity cost allocation of the electricity capacity price.
According to this allocation method, the total cost of the allocated system capacity is calculated as (3) An improved BARY curve capacity cost allocation model.
Due to the total concurrent rate CF i of the user group being less than 1, using this method for capacity cost allocation cannot recover all the capacity costs.On the basis of this, this article adopts an improved BARY curve capacity cost allocation model, and the main improvement method is to modify the proportion of the capacity cost allocation in capacity and electricity bills.The proportion coefficient of the improved system capacity cost in the capacity and electricity bills is as follows: where e cap represents the allocation proportion coefficient of the capacity cost of the voltage level allocated to the capacity tariff, e ele represents the allocation proportion coefficient of the capacity cost of the voltage level allocated to the electricity tariff, and E i represents the proportion of the capacity cost allocated to the capacity tariff and the capacity cost allocated to the electricity tariff based on the above capacity cost allocation model: where C′ cap represents the capacity cost that should be allocated to the electricity capacity bill based on the improved capacity cost allocation model; C′ ele represents the capacity cost that should be allocated to the electricity bill based on the improved capacity cost allocation model.After such modifications, the sum of the capacity cost allocation coefficient e cap in the electricity capacity bill and the allocation coefficient e ele in the energy electricity bill is only 1.At this time, the allocation proportion coefficient can meet the requirement that all system capacity costs are allocated without duplication, which is reasonable.

| Interest demands of different subjects
From the perspective of power users, their pursuit is the lowest electricity bill.On the one hand, users can obtain lower electricity bills by adjusting their electricity consumption behavior to increase electricity consumption during low periods or to have interruptions during peak periods.On the other hand, to improve the load rate of users and improve the efficiency of system operation, in the context of the construction of new power systems, the price difference between users with higher load rates and users with lower load rates should be further increased to achieve a reasonable response of electricity prices to user electricity consumption levels.
From the perspective of power grid companies, their pursuit is the fair allocation and recovery of transmission and distribution costs, as well as the maximization of electricity sales profits for power grid enterprises.Since the market-oriented reform of electricity, the cost accounting of the power transmission and distribution for power grid enterprises is independently supervised and audited.Therefore, the equation of the power transmission and distribution prices for power grid enterprises must consider whether they can maximize the recovery of the power transmission and distribution costs with minimal risk, and on this basis, obtain maximum benefits, so as to achieve sustainable development for power grid enterprises themselves.

| Objective function
On the basis of the collaborative capacity cost allocation model of capacity pricing and electricity pricing in the previous section, capacity pricing is calculated from the perspective of the user load rate.Assuming that the existing electricity price level can meet the current demand for cost recovery in the power system, the objective optimization function of capacity pricing based on the user load rate in the new power system is where p ¯is the average electricity price function for all users after optimizing the electricity capacity price and responding to the load; I ele represents the electricity rate charged to all users; I cap represents the total amount of the electricity capacity fees charged to all users; Q is the total electricity consumption of the user, which is the total amount of electricity purchased by the power grid company after deducting line losses and internal power plant power generation; S is the total capacity of the user transformer charged based on the transformer capacity; l is the total maximum demand of users charged for electricity capacity based on maximum demand; β is the proportion of the electricity capacity fee charged based on the transformer capacity in the overall electricity capacity fee.

| Constraint condition
Under this optimization objective function, we consider incorporating various constraints of the new power system: (1) Constraints on the sustainable development of the new power system grid If implementing a load-rate-based electricity price and capacity price collaborative electricity price scheme, it is necessary to ensure the sustainable development of the power grid and to achieve fair cost sharing and reasonable distribution of benefits.Therefore, the following constraints need to be met: ( ) s,q low s,q 0  ( where P P P , , high s,q mid s,q low s,q represent the adjustment part of the electricity capacity price of the power grid company for users with high, medium, and low monthly load rates (there are slight differences in the actual load rate levels of each voltage level, which is an overview, and the same applies to the following text), respectively; R represents the profit of the power grid company after optimizing and adjusting the electricity price; R 0 represents the original profit of the power grid company before the electricity price optimization adjustment.This equation indicates that the optimized electricity price mechanism needs to ensure the reasonable growth of profits for the power grid company and maintain the sustainable development of the power grid.
(2) Constraints on the flexibility construction of the new power system grid Under the new power system, frequent fluctuations in the high proportion of renewable energy make grid scheduling more complex.To adapt to the large-scale integration of new energy, the flexibility of the power system is an important indicator for the construction of new power systems.Therefore, we need to meet the following constraints: ( ) where I represents the electricity capacity price income of the power grid company after the optimization and adjustment of electricity prices; η OFD is the ratio of the electricity capacity price revenue used for power flexibility construction; Y OFD is the investment amount for the flexibility construction of the power grid; T OFD is the time requirement for grid flexibility, which refers to the duration of the upward increase in net load predicted power; P OFD is the flexible power demand, which refers to the maximum fluctuation of the net load power within the scheduling time window; R C refers to the maximum fluctuation rate of the net load power provided by the flexible power supply, which is the climbing ability.
(3) Constraints on the reliability construction of the new power system grid With the construction of new power systems, the integration of a large amount of renewable energy generation has changed the original operation and control methods of the power system, increased the uncertainty of the power system's operation and control, and posed a threat to power security.As a result, the investment in security and reliability issues in the power grid has gradually increased, and it needs to be recovered in the electricity capacity price.Therefore, we need to meet the following constraints: ( ) where I represents the electricity capacity price income of the power grid company after the optimization and adjustment of electricity prices; η SAR is the ratio of the electricity capacity price revenue used for grid reliability construction; Y SAR represents the investment in the construction of power grid flexibility.

| Solution to the optimization model
It can be seen that the optimization model for electricity capacity price calculation based on the BARY curve includes the electricity price adjustment part P P P , , high s,q mid s,q low s,q of the power grid company for high, medium, and low load-rate users, respectively.The optimization objective of capacity pricing is a multivariate nonlinear function, and there are also nonlinear constraint functions in the constraint conditions; so, the solution to the model is the solution to a multivariate nonlinear optimization model: s,q low s,q 0 high s,q mid s,q low s,q OFD OFD OFD OFD C high s,q mid s,q low s,q SAR SAR    (19)   Considering that general nonlinear optimization solving functions may not meet the requirements and may result in local optima, this section considers using the genetic algorithm (GA) for solving.This algorithm is a method of searching for the optimal solution by simulating natural evolution processes.When solving more complex combinatorial optimization problems, GA can usually obtain better optimization results faster than some conventional optimization algorithms. 23The calculation and solution steps of the electricity capacity price based on the GA are shown in Figure 3.
The calculation of the load rate and capacity price based on the GA abstracts the solving process of the nonlinear optimization model into the process of heredity, mutation, crossover, replication, and so forth, of individual genes, eliminates the inferior solution, and obtains the optimal solution in a similar way to natural selection.

| EXAMPLE ANALYSIS
This research is based on the 8760 h load data collected from 120 typical industrial and commercial users in a certain region of China in 2021 and calculates the capacity cost allocation ratio for users with different voltage levels.
First, based on the electricity sales situation in the region in 2021, we calculated the ratio of the electricity consumption and capacity consumption for each voltage level at the current electricity price level, as shown in Table 1.
F I G U R E 3 Optimization model of capacity pricing based on the genetic algorithm.power systems based on the load rate (1) Data visualization analysis Among the 120 power users with different voltage levels, there are 27 users with a 110 kV voltage level (five general industrial and commercial users and 22 large industrial users), and 21 users with a 35 kV voltage level (five general industrial and commercial users and 16 large industrial users).This example selected data from users with voltage levels of 110 and 35 kV for the capacity cost allocation calculation.We analyzed the monthly load rates of the above users in 2021, as shown in Figure 4.
As shown in Figure 4, it can be observed that users with 110 and 35 kV voltage levels generally have higher monthly load rates.The statistical results show that 78.01% of the above-scattered points have a load rate exceeding 60%, while only 3.73% have a load rate of less than 40%.Therefore, based on the load-rate clustering results, this work narrowed the load-rate range in the high and medium load-rate states, while expanding the load-rate range in the low load-rate range, to achieve precise differentiation between the high and medium load rates and to increase the coverage range of low loadrate intervals, so that the power grid can accurately distinguish power users with different load rates at this voltage level, to develop a reasonable electricity pricing mechanism.
(2) BARY curve determination We calculated the system load rate and user load simultaneity rate for each month of 2021 for the 110 and 35 kV user groups.Among them, the system load rate and the daily load simultaneity rate of the user group are shown in Table 2.
Then, based on the system load rate and user load simultaneity data of the user group, the α value was determined by substituting it into the BARY curve empirical equation mentioned earlier.After calculating and fitting based on MATLAB, the α value of the BARY curve for 110 kV voltage level users was determined to be −2.243.(3) Determination of the allocation ratio coefficient After determining the BARY curves for users at the 110 and 35 kV voltage levels, this article calculated the capacity cost allocation ratio based on the BARY curve for low monthly load-rate users with load rates within the range of (0%, 55.01%), mid-monthly load rates within the range of (55.01%, 74.7%), and high monthly load rates with load rates within the range of (74.7%, 100%), based on the division of different load rates for users.The calculation results are shown in Figure 5.
As shown in Figure 5, for users at the 110 and 35 kV voltage levels, their capacity costs should be more allocated to the electricity capacity price.The optimized electricity capacity price to electricity price allocation ratio coefficient obtained from the final calculation compared to the existing capacity cost allocation ratio is shown in Table 3.
As shown in Table 3, the proportion of the capacity cost allocated to capacity and electricity bills was calculated based on the BARY curve for each load-rate interval in this work.First, the proportion coefficient of the capacity cost allocated to the capacity and electricity prices at the current capacity and electricity price levels was calculated to be 0.2666.This means that under the current capacity and electricity price levels, the ratio of the capacity cost allocated to the capacity and electricity prices to the capacity cost allocated to the electricity prices is 0.2666.Then, this article calculated the optimized allocation proportion of the 110 and 35 kV voltage level users in different load-rate intervals under different load-rate levels: (1) For low load-rate users with a load rate in the (0%, 55.01%) range, the optimized capacity cost allocation ratio is 0.3642.This result indicates that for users with a load rate in the (0%, 55.01%) range, the ratio of the capacity cost allocated to the capacity tariff to the capacity cost allocated to the electricity tariff is 0.3642:1.
(2) For users with medium load rates in the range of (55.01%, 74.7%), the optimized capacity cost allocation ratio is 0.8069.This result indicates that for users with load rates in the range of (55.01%, 74.7%), the ratio of the capacity cost allocated to the capacity tariff to the capacity cost allocated to the electricity tariff is 0.8069:1.The capacity cost accounting results, capacity cost sharing proportion coefficient, user load characteristic data samples, and power sales data of users under the load rate of each section calculated in the previous section were substituted into the capacity price and electricity price optimization calculation model, which was optimized based on the GA algorithm.The optimized user capacity price results are shown in Table 4.
Comparing the optimized transmission and distribution electricity price with the actual executed transmission and distribution electricity price, it can be concluded that after optimization, the load-rate range of high loadrate users was between 74.7% and 100%, and the electricity price in the electricity transmission and distribution price should be reduced to 0.0859 CNY/ kW h.The electricity capacity price charged based on maximum demand should be increased to 79.47 CNY/ kW month, and the electricity capacity price charged based on the transformer capacity should be increased to 52.98 CNY/kVA month.The load-rate range of users with a medium load rate was between 55.01% and 74.7%.The electricity price in the transmission and distribution electricity price should be reduced to 0.1075 CNY/kW h.The electricity capacity price charged based on maximum demand should be increased to 63.65 CNY/kW month, and the electricity capacity price charged based on the transformer capacity should be increased to 42.43 CNY/ kVA month.The load-rate range of low load-rate users was between 0% and 55.01%, and the electricity price in their transmission and distribution electricity price should be reduced to 0.1423 CNY/kW h.The electricity capacity price charged based on maximum demand should be reduced to 38.05 CNY/kW month, and the electricity capacity price charged based on the transformer capacity should be reduced to 25.37 CNY/ kVA month.
On the basis of the above calculation results, this research argues that, for users with voltage levels of 110 and 35 kV, the load-rate characteristics of large industrial users and general industrial and commercial users are relatively similar, and the power load scale and kilowatt-hour electricity fee level are relatively similar, which meets the conditions for electricity price consolidation.At present, the pricing of electricity capacity prices for users of this voltage level is low, while the electricity price is on the high side.Under the condition of ensuring the basic stability of the overall electricity price, the electricity capacity price of users with high and medium load rates should be increased, and their electricity price should be moderately reduced.For users with low load rates, the increase in electricity capacity prices and the decrease in electricity prices are relatively small, to avoid the impact of significant electricity price adjustments on users with low load rates.This electricity price optimization method can effectively achieve electricity price differentiation for users with different load rates, ensure the economic operation of the power grid, and achieve effective recovery and reasonable allocation of capacity costs.

| DISCUSSION
With the increasing investment in the power grid and the construction of new power systems, the existing capacity pricing mechanism is no longer able to meet the needs of fair allocation and reasonable recovery of capacity costs.
It is necessary to establish a capacity pricing mechanism that adapts to the new power system.At present, the implementation range of electricity capacity prices in China is too small.Among users who implement capacity pricing, a large proportion of users charge basic electricity fees based on the transformer capacity, which does not have the effect of encouraging users to reduce maximum demand.At present, the two-part electricity price structure cannot accurately reflect the cost structure of electricity production, with a low electricity capacity price and a high electricity price.
The issue of increased capacity costs and less significant electricity growth in the new power system will become more prominent.On the basis of the above analysis, this study proposes the following development directions for the formation mechanism of electricity capacity prices: (1) Adopting the approach of this article, general industrial and commercial users and large industrial users are included in the scope of the two-part electricity price collection.Through this study, it was found that increasing the scope of charging electricity capacity prices and expanding the user group responsible for grid capacity costs can not only achieve fair allocation and reasonable recovery of capacity costs, but also effectively alleviate the cost pressure faced by some users when capacity costs increase.(2) On the basis of the load rate, different electricity capacity prices and kilowatt-hour electricity prices are implemented for users with different electricity load characteristics, achieving the precise division of user groups and fair allocation of capacity costs among different user types.(3) It reduces the price difference between the electricity capacity price charged based on transformer capacity and the electricity capacity price charged based on maximum demand appropriately.Charging electricity capacity based on maximum demand is beneficial for effectively reducing the maximum load of users in the power grid.This study suggests that, in the future, according to the needs of new power systems, the price difference should be appropriately reduced to make more users tend to pay the electricity capacity price based on maximum demand to achieve the effect of reducing the maximum load of users and reducing the pressure on the power grid.(4) This study realized a reasonable allocation of new capacity costs from the perspective of the electricity capacity price and electricity price synergy.In response to the capacity cost allocation problem of typical new investment in the new power system of the power grid, the proportion of new investment costs that should be allocated to users of different voltage levels is first determined based on the "beneficiary" principle.Second, it is recommended to use a collaborative approach of electricity capacity price and electricity price, based on the characteristics of user load, to allocate them reasonably among various user groups.

T A B L E 1 4
The ratio of the electricity capacity fee to kilowatt-hour electricity fee under the current electricity price level.Scatter plot of the monthly load rate for users with voltage levels of 110 and 35 kV.
Daily simultaneous rate and load rate for users of 110 and 35 kV voltage levels in 2021 (unit: MW).Calculation results of the capacity cost allocation based on the BARY curve.Capacity cost allocation ratio.For high load-rate users with load rates in the (74.7%, 100%) range, the optimized capacity cost allocation ratio is 1.2604.This result indicates that for power users with voltage levels of 110 and 35 kV, if their load rates are in the (74.7%, 100%) range, the ratio of the capacity cost allocated to the capacity fee to the capacity cost allocated to the electricity fee is 1.2604:1.
T A B L E 3 Electricity price results after optimizing the capacity price and the electricity price.