Method for analyzing key influence parameters of power spot market clearing price

Abstract

This invention discloses a method for analyzing key influencing parameters of the clearing price in the electricity spot market. Based on electricity spot market rules and nodal pricing models, it establishes the operating parameters affecting the clearing price and constructs an influencing factor system including unit cost, unit operation, system balance, and cross-sectional power flow. Then, it further constructs a full-time clearing simulation model of the electricity spot market considering carbon trading costs, establishing the objective function and constraints of the clearing model. Finally, based on the electricity spot market clearing simulation model, it compares the impact of each operating parameter on the market clearing price through sensitivity analysis and correlation diagram construction, thereby assessing the key parameters affecting the electricity spot market. This method not only systematically identifies the operating parameters affecting the clearing price in the electricity spot market but also allows for simple and efficient calculation and comparison of the impact of each parameter on the clearing price, thus clarifying the key parameters affecting the spot market and improving the decision-making efficiency of market participants.

Description

Technical Field

[0001] This invention relates to the field of electricity spot market clearing, and in particular to a method for analyzing key influencing parameters of electricity prices in electricity spot market clearing. Background Technology

[0002] Existing electricity spot market clearing models integrate unit combination and economic dispatch models with safety constraints. The clearing price is closely related to factors such as grid parameters, unit parameters, and supply and demand parameters. Compared to medium- and long-term trading, electricity spot trading profoundly reflects the real electricity price in both time and space dimensions. On the one hand, real-time supply and demand relationships significantly influence spot prices. When electricity supply and demand are unbalanced during a given period, spot prices may face strong volatility risks, posing challenges to power system operation and electricity market risk management. On the other hand, the market clearing price is also affected by different nodes in the grid. Factors such as the operating efficiency and reliability of the transmission network profoundly impact the returns of market participants. In particular, among the many influencing parameters, subtle changes in key parameters can have a significant impact on the clearing price, requiring serious attention from market participants. However, existing research on electricity spot market clearing prices lacks a comprehensive and clear understanding of the overall impact of various factors on the clearing price, and most existing studies remain at the qualitative analysis level.

[0003] Therefore, based on the electricity spot market clearing model, the most critical operating parameters affecting the clearing price can be extracted to identify the patterns of nodal price fluctuations. This not only helps control the risk of price fluctuations within the market but also effectively improves the decision-making efficiency and level of market participants. Furthermore, by combining quantitative analysis of the parameter impact with market simulation models, the rules and clearing model can be validated. This provides theoretical analysis and technical support for the revision of rules and the improvement of the platform in my country's electricity spot market, promoting the sustainable development of the electricity market. Summary of the Invention

[0004] The purpose of this invention is to overcome the aforementioned shortcomings in the existing technology and provide a method for analyzing key influencing parameters of the clearing price in the electricity spot market. This method systematically sorts out the operating parameters that affect the clearing price in the electricity spot market, clarifies the degree of influence of each operating parameter on the clearing price through market clearing simulation models and quantitative methods, thereby identifying the key parameters affecting the spot market, improving the decision-making efficiency of market participants, providing theoretical analysis and technical support for the revision of rules and improvement of the platform in my country's electricity spot market, and promoting the sustainable development of the electricity market.

[0005] The technical solution adopted by this invention to solve the above problems is: a method for analyzing key influencing parameters of electricity spot market clearing prices, characterized by comprising the following steps:

[0006] 1) Based on the rules of the electricity spot market and the nodal price model, establish the operating parameters that affect the clearing price, and construct a system of influencing factors including unit cost, unit operation, system balance, and cross-sectional power flow.

[0007] 2) By sorting out the influencing parameters, a full-time clearing simulation model of the electricity spot market considering carbon trading costs is further constructed, and the objective function and constraints of the clearing model are established.

[0008] 3) Based on the electricity spot market clearing simulation model, the impact of each operating parameter on the market clearing price is compared by using sensitivity analysis and constructing correlation diagrams, thereby assessing the key parameters affecting the electricity spot market.

[0009] In step (1), the operating parameters affecting the clearing price are classified according to the nodal pricing model. They can be divided into four categories: unit cost, unit operation, system operation, and cross-sectional power flow. The specific classification criteria are as follows:

[0010] 1) Unit cost parameters mainly include unit operating costs and CO2 emissions. These parameters affect the clearing price through the bidding behavior of enterprises. Under the same system load, the higher the marginal cost of the unit, the higher the market clearing price. In addition to the basic unit operating costs, after being included in the carbon emission market, carbon emission allowances can also be traded and circulated among power generation enterprises. The CO2 emissions of enterprises further affect the calculation of unit costs. Therefore, power generation enterprises need to consider the combination of electricity trading and carbon trading to maximize efficiency.

[0011] 2) Unit operation parameters mainly include the upper and lower limits of unit output and the ramp rate. These parameters affect the actual available output of the unit and thus the nodal price in the clearing calculation. When the output of low-bid units is restricted, the system cannot determine the clearing order according to the order of unit prices. Therefore, high-bid units with a relatively wide output range may be called up to meet the system balance, thereby causing the nodal price to rise.

[0012] 3) System operation parameters mainly include system forecast load and system reserve demand; when the system load increases, the power system needs to call up units with higher bids to meet the supply and demand balance; in addition, high system load will also affect the degree of line congestion, thereby increasing the congestion cost of nodal electricity price; system reserve demand, as the boundary condition for market clearing, affects nodal electricity price by affecting the system's supply and demand balance constraints.

[0013] 4) Cross-sectional power flow parameters mainly include line transmission capacity, which affects the possible congestion of the system. When the line transmission capacity is limited, the system may not be able to call the generators at the relevant nodes in order of price, and the generators with higher bids may be called to meet the supply and demand balance. Therefore, generators at key nodes can take advantage of line congestion to exert local market power.

[0014] The classification of various parameters affecting operation is shown in Table 1.

[0015] Table 1 Key Operating Parameters Affecting Spot Market Clearing Electricity Prices

[0016]

[0017]

[0018] In step (2), the objective function and constraints of the full-time clearing simulation model of the electricity spot market considering carbon emission trading are determined, and their specific expressions are as follows:

[0019] 1) Objective function

[0020] In this method, the total power generation cost of the unit consists of carbon trading costs and operating costs. The expression for carbon trading costs is as follows:

[0021] ρ co2i (t)=α(ε i P i (t)-κP i (t)) (2)

[0022] In the formula, α represents the trading price of each ton of CO2 in the carbon market, and ε i κ represents the carbon emission factor of unit i, P represents the carbon allowance benchmark in the market, and κ represents the carbon emission factor of unit i. i (t) represents the power generation of unit i in time period t;

[0023] The expression for the unit operating cost is as follows:

[0024] ρ Gi (t)=a i (P i (t)) 2 +b i P i (t)+c i (2)

[0025] In the formula, a i b i c i These are the parameters of the cost function for unit i;

[0026] By integrating the carbon emission trading costs and operating costs of the unit, the overall cost curve of the unit is expressed as follows:

[0027] C i (P i (t))=a i (P i (t)) 2 +[b i +α(ε i -k)]P i (t)+c i (3)

[0028] The overall cost curve is a quadratic function. In the actual electricity spot market, multiple bidding segments are usually allowed. The unit bidding curve is an increasing segmented step curve. At most m bidding segments can be submitted in each trading clearing period (i.e., 1 ≤ number of bidding segments ≤ m). Each segment of the output range must be connected end to end, and the start and end points of the output are the unit's minimum stable technical output (MW) and the unit's rated active power (MW), respectively. Therefore, the original overall cost curve needs to be linearized.

[0029] After linearly segmenting the original cost curve of a certain unit, the slopes from each power point to the next power point are k1, k2, k3, ..., k m By treating each slope as the quoted value and each power distribution point as the beginning and end of the quoted output range, a stepped pricing curve based on unit operating costs is formed. Because the operating cost function is convex, k1 < k2 < k3 < ... < k m This meets the requirement that the price curve is monotonically non-decreasing;

[0030] After considering the multiple bids from the generating units, the objective function of the clearing model can be ultimately expressed as follows:

[0031]

[0032]

[0033] In the formula, and These represent the quoted price and the winning bid capacity for the j-th segment of the i-th generating unit, respectively, with m being the number of segment bids for the generating units;

[0034] 2) Constraints

[0035] The clearing simulation model mainly includes five constraints: system balance, system reserve, unit output, unit ramp-up, and cross-sectional power flow, as detailed below:

[0036] a) System equilibrium constraints

[0037]

[0038] In the formula, λ(t) represents the shadow price of the system equilibrium constraint. d(t) represents the total output of all units in the system during time period t, and d(t) represents the system load during time period t. Since this invention uses the calculation method based on the optimal DC power flow, network loss factors are not considered here.

[0039] b) System standby constraints

[0040]

[0041]

[0042] In the formula, and v (t) represent the shadow prices of the upper and lower limits of the system's reserve capacity constraints, respectively. and P i These are the upper and lower limits of the output of unit i, respectively. and R (t) represents the system's increased and decreased spin-up reserves during time period t, respectively;

[0043] c) Unit output constraints

[0044]

[0045] t i (t): P i (t)≥ P i (10)

[0046] In the formula, and t i (t) represent the shadow prices of the upper and lower limits of the unit's output, respectively;

[0047] d) Unit climbing and landslide constraints

[0048]

[0049]

[0050] In the formula, and d i (t) represents the shadow prices of the unit's ramp-up and landslide constraints, respectively, Δ i This represents the maximum increase or decrease in output that unit i can achieve in each time period;

[0051] e) Cross-sectional power flow constraints

[0052]

[0053]

[0054] In the formula, and m l (t) represents the shadow price of the current and its downstream constraint at the cross-section, P l (t) represents the tidal power of section l during time period t. and P l This indicates the upper and lower limits of the power flow at section l.

[0055] In step (3), the finite difference method is used to calculate the sensitivity of the clearing price; after calculating the change value of the average price corresponding to the change of a single operating parameter, the sensitivity coefficient of the price to each parameter can be obtained to quantitatively measure the relative change between the two.

[0056] Because the dimensions of different parameters are inconsistent, in order to make a unified comparison of the data, it is necessary to normalize the changes in the parameters. The normalization formula is as follows:

[0057]

[0058] In the formula, Δx is the normalized change, x after It is the value after the parameter changes, x original The parameter represents the original ground state value;

[0059] The corresponding sensitivity analysis can be calculated using the following formula in the difference scheme:

[0060]

[0061] In the formula, F k Let be the sensitivity coefficient of electricity price under different conditions, Δk be the normalized change of the influencing parameters, and ΔA be the normalized change of the average node electricity price.

[0062] To identify the key operating parameters affecting the clearing electricity price, a correlation diagram between the influencing parameters and the clearing electricity price is established for comparison. The specific process is as follows:

[0063] a) Divide the entire day into peak load, off-peak load, and intermediate load periods according to load levels, and further calculate the sensitivity of the average nodal price of the entire network to each influencing parameter, i.e., the magnitude of the relative change.

[0064] b) Based on the ground-state example, the trial-and-error method is used to find the upper and lower limits of each influencing factor that makes the clearing model solvable and changes the clearing price; for a single operating influencing parameter, six different scenarios are generated by taking six different influencing parameters within the upper and lower limit ranges with the same step size; the forward difference method is used to calculate the six sensitivities of the clearing price to this parameter, as shown in the following expressions:

[0065]

[0066] c) Assign scores to the original sensitivity data and use a weighted average method to obtain the average score of the influence of each parameter. The sensitivity scoring criteria are shown in Table 2.

[0067] Table 2 Sensitivity Scoring Criteria

[0068] Sensitivity range 0 (0,0.5] (0.5,1] (1,3] (3,5] (5,7] Assigning α 0 1 2 3 4 5 Sensitivity range (7,9] (9,11] (11,13] (13,15] (15,17] (17,+∞) Assigning α 6 7 8 9 10 10

[0069] The mathematical expression for the average score assigned at time t is as follows:

[0070]

[0071] In the formula, F kn (t) represents the sensitivity in the nth scene, α(F) kn (t) represents the sensitivity value when F kn The corresponding score for (t);

[0072] The average value can be used to obtain the comprehensive score for different load levels during different periods, as shown in the following formula, for unified comparison, and finally to determine the key influencing parameters of the electricity spot market;

[0073]

[0074]

[0075]

[0076] In the formula, S H S M S L These represent the combined scores for peak load, mid-load, and trough load periods, respectively. H m M m L These represent the total number of periods during peak load, mid-load, and off-load periods, respectively.

[0077] Compared with existing technologies, this invention has the following advantages and effects: This method not only systematically sorts out the operating parameters that affect the clearing price in the electricity spot market, but also can simply and efficiently calculate and compare the degree of influence of each operating parameter on the clearing price, thereby clarifying the key parameters affecting the spot market, improving the decision-making efficiency of market participants, and providing theoretical analysis and technical support for the revision of rules and improvement of the platform in my country's electricity spot market, thus promoting the sustainable development of the electricity market. Attached Figure Description

[0078] Figure 1 This is a graph showing the linear processing of the original cost curve of this invention.

[0079] Figure 2This is a graph showing the correlation between the influencing parameters of this invention and the clearing electricity price.

[0080] Figure 3 This is a flowchart of the present invention. Detailed Implementation

[0081] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.

[0082] Example

[0083] See Figures 1 to 3 In this embodiment, a method for analyzing key influencing parameters of electricity spot market clearing prices includes the following steps:

[0084] 1. The IEEE-39 node system was used for verification analysis. The basic parameters of a certain electricity spot market were selected as follows: the operating cost coefficients of all generating units in the entire network were a = 0.21, b = 6.3, and c = 4.2. The physical parameters of the generating units are shown in Table 3, the day-ahead predicted load of the system is shown in Table 4, and the physical parameters of the lines are shown in Table 5. The positive and negative reserve demand was taken as 100MW. The carbon quota benchmark value was taken as 7.838 tons / 10,000 kWh for 100MW coal-fired units in Shanghai in 2020. The carbon unit transaction price was taken as 41.5 yuan / ton CO2 transaction price in the Shanghai carbon market in March 2021. In the example, the upper limit of unit output is the maximum declared output of each unit, the lower limit of unit output is the minimum technical output, and the clearing price is the weighted average nodal price of the entire network load side.

[0085] Table 3 Unit Physical Parameters

[0086]

[0087] Table 4 System Day-ahead Forecast Load

[0088]

[0089] Table 5 System Line Physical Parameters

[0090]

[0091] 2. Establish a system that considers the factors affecting the clearing price of the electricity spot market. Based on the specific data in this case, the ground state expression of the market parameters is selected as shown in Table 6 below.

[0092] Table 6 Factors Affecting the Clearing Price in the Electricity Spot Market

[0093]

[0094]

[0095] 3. Establish a full-time clearing simulation model for the electricity spot market that takes carbon emission trading into account, and define the objective function and constraints of the model.

[0096] The expressions for the carbon trading costs of each unit are as follows:

[0097] ρ co2i (t)=41.5(ε i P i (t)-783.8P i (t))

[0098] The expressions for the operating costs of each unit are as follows:

[0099] ρ Gi (t)=0.21(P i (t)) 2 +6.3P i (t)+4.2

[0100] By integrating the carbon emission trading costs and operating costs of the unit, the overall cost curve of the unit is expressed as follows:

[0101] C i (P i (t))=0.21(P i (t)) 2 +6.3+41.5(ε i -783.8)·10 -3 ]P i (t)+4.2

[0102] Based on the generator set comprehensive operating cost function, it is linearized to obtain the unit's four-segment price list as shown in Table 7.

[0103] Table 7 Generator Set Price Parameters

[0104] Unit: Price Quoted (RMB / MW) Capacity (MW)

[0105] unit Quotation capacity Quotation capacity Quotation capacity Quotation capacity 1 137.34 416 224.7 624 312.06 832 399.42 1040 2 87.57 258 141.75 387 195.93 516 250.32 646 3 97.65 290 158.55 435 219.45 580 280.35 725 4 105.63 260 142.8 390 197.4 520 252.42 652 5 88.2 204 113.4 306 156.24 408 202.23 508 6 70.56 276 151.2 414 209.16 552 266.49 687 7 93.24 232 128.1 348 176.82 464 225.54 580 8 79.38 226 124.95 339 172.41 452 219.66 564 9 77.49 346 187.95 519 260.61 692 333.27 865 10 115.29 440 237.3 660 329.7 880 422.1 1100

[0106] After considering the multi-segment pricing of the generating units, the scheduling pricing objective function can be expressed as follows:

[0107]

[0108]

[0109] In the formula, and These represent the quoted price and the winning bid for the j-th segment of the i-th generating unit, respectively. The number of unit quotation segments is m = 5.

[0110] The constraints of the clearing simulation model include system balance, system reserve, unit output, unit ramp-up, and cross-sectional power flow, and their specific expressions are as follows:

[0111] a) System equilibrium constraints

[0112]

[0113] In the formula, λ(t) represents the shadow price of the system equilibrium constraint. Let d(t) represent the total power output of all units in the system during time period t, and d(t) represent the system load during time period t. Since this invention uses a calculation method based on optimal DC power flow, network loss factors are not considered here.

[0114] b) System standby constraints

[0115]

[0116]

[0117] Where, and v (t) represent the shadow prices of the upper and lower limits of the system's reserve capacity constraints, respectively. and P i These are the upper and lower limits of the output of unit i, respectively. and R (t) represents the system's upward and downward rotational reserve adjustments during time period t.

[0118] c) Unit output constraints

[0119]

[0120] t i (t): P i (t)≥ P i

[0121] Where, and t i (t) represent the shadow prices of the upper and lower limits of the unit's output, respectively.

[0122] d) Unit climbing and landslide constraints

[0123]

[0124] d i (t): P i (t)-p i (t-1)≥-Δ i

[0125] Where, and d i (t) represents the shadow prices of the unit's ramp-up and landslide constraints, respectively, Δ i This represents the maximum increase or decrease in output that unit i can achieve in each time period.

[0126] e) Cross-sectional power flow constraints

[0127]

[0128] m l (t): P l (t)≥ P l

[0129] Where, and m l (t) represents the shadow price of the current and its downstream constraint at the cross-section, P l (t) represents the tidal power of section l during time period t. and P l This indicates the upper and lower limits of the power flow at section l.

[0130] 4. This invention employs the finite difference method to calculate the sensitivity of the clearing price. After calculating the change in the average price corresponding to a change in a single operating parameter, the sensitivity coefficient of the price to each parameter can be obtained to quantitatively measure the relative change between the two.

[0131] Because the dimensions of different parameters are inconsistent, in order to make a unified comparison of the data, it is necessary to normalize the changes in the parameters. The normalization formula is as follows:

[0132]

[0133] In the formula, Δx is the normalized change, x after It is the value after the parameter changes, x original The parameter represents the original ground state value.

[0134] The corresponding sensitivity analysis can be calculated using the following formula in the difference scheme:

[0135]

[0136] In the formula, F k Let be the sensitivity coefficient of electricity price under different conditions, Δk be the normalized change of the influencing parameters, and ΔA be the normalized change of the average node electricity price.

[0137] The entire day is divided into peak load periods (including periods 11, 16, 17, 19, and 20), off-peak load periods (including periods 4-7), and intermediate load periods (the remaining periods) based on load levels. The sensitivity of the average nodal price across the entire network to various influencing parameters, i.e., the magnitude of their relative changes, is then calculated. The magnitude of these relative changes, to a certain extent, measures the impact of various parameters on the clearing price.

[0138] This embodiment only shows the sensitivity analysis results of the more obvious average nodal electricity price on the parameters of unit operating cost, CO2 emissions, unit output limit, system predicted load, and line transmission capacity. The analysis methods for the impact of other influencing factors on electricity price are similar and will not be repeated.

[0139] Based on the ground-state example, the trial-and-error method is used to find the upper and lower limits of each influencing factor that makes the clearing model solvable and changes the clearing price. For unit operating costs and CO2 emissions, six different parameter scenarios are generated using ground-state parameters within the range of 90%–110% with a 4% step size. The forward difference method is used to calculate six sets of sensitivities of the price to these parameters. For the upper limit of unit output, six different scenarios are generated using ground-state parameters within the range of 94%–99% with a 1% step size. The six sets of sensitivities of the price to the upper limit of unit output parameters are calculated. For the predicted system load, six different load scenarios are generated using ground-state load within the range of 95%–105% with a 2% step size. The six sets of sensitivities of the price to the predicted system load parameters are calculated. For line transmission capacity, six different scenarios are generated using ground-state parameters within the range of 90%–105% with a 3% step size. The six sets of sensitivities of the price to the line transmission capacity parameters are calculated. The raw sensitivity data were scored according to the scoring criteria shown in Table 8, and the average scores were obtained for comparison. The comparison results are shown in Table 9. It is clear that the key influencing parameters differ under different load levels and time periods.

[0140] Table 8 Sensitivity Scoring Criteria

[0141] Sensitivity range 0 (0,0.5] (0.5,1] (1,3] (3,5] (5,7] Assignment 0 1 2 3 4 5 Sensitivity range (7,9] (9,11] (11,13] (13,15] (15,17] (17,+∞) Assignment 6 7 8 9 10 10

[0142] Table 9. Sensitivity scores assigned to some influencing parameters based on the clearing average electricity price.

[0143] Average score Operating costs CO2 emissions upper limit of unit output System predicted load Line transmission capacity Peak load period 3.20 2.97 4.30 4.90 3.20 Mid-load period 2.13 1.08 1.00 2.00 0.77 Low load period 2.21 1.08 0 1.54 0

[0144] Repeat the above research steps to calculate and statistically analyze the sensitivity of the clearing electricity price to all influencing parameters. The final correlation diagram between the influencing parameters and the clearing electricity price is shown below. Figure 2 As shown.

[0145] A horizontal comparison of the correlation diagrams shows that during peak load periods, the system's predicted load and maximum system output have the most significant impact on nodal prices; line transmission capacity and operating costs have the second most significant impact on clearing prices. During mid-load periods, system predicted load and operating costs are the most critical influencing factors. During off-peak periods, operating costs have a crucial impact on clearing prices.

[0146] A longitudinal comparison of the correlation diagrams shows that the system's predicted load, line transmission capacity, and upper limit of unit output have a significantly greater impact on nodal prices during peak load periods than during off-peak periods. Operating costs and CO2 emissions have a slightly greater impact on nodal prices during peak load periods than during off-peak periods. Conversely, the lower limit of unit output has a greater impact on clearing prices during off-peak periods than at other times.

[0147] Therefore, peak load periods, as critical times, should be given high priority by market participants, as they essentially reflect system congestion and supply-demand conditions. When system demand far exceeds supply, congestion is usually more severe. Thus, as a key influencing factor, inaccurate load forecasts will have a significant impact on clearing prices; power generators' strategies of not reporting full capacity and using high bids will also significantly increase nodal prices; and increasing transmission capacity through expansion and other means can effectively alleviate congestion during peak periods, thereby improving the overall economic security of the power grid.

[0148] Finally, in this embodiment, the lower limit of unit output, unit ramp-up rate, and system reserve requirements have the least impact on nodal electricity prices.

[0149] Any content not described in detail in this specification is prior art known to those skilled in the art.

[0150] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the scope of protection of the present invention. Any modifications and refinements made by those skilled in the art without departing from the concept and scope of the present invention shall fall within the scope of protection of the present invention.

Claims

1. A method for analyzing key influencing parameters of electricity spot market clearing prices, characterized in that, Includes the following steps: (1) Based on the rules of the electricity spot market and the nodal price model, establish the operating parameters that affect the clearing price and construct a system of influencing factors including unit cost, unit operation, system balance and cross-sectional power flow. (2) By sorting out the influencing parameters, we further construct a full-time clearing simulation model of the electricity spot market that takes into account carbon trading costs, and establish the objective function and constraints of the clearing model. (3) Based on the electricity spot market clearing simulation model, the impact of each operating parameter on the market clearing price is compared by using sensitivity analysis and constructing a correlation diagram, thereby assessing the key parameters affecting the electricity spot market; In step (1), the operating parameters affecting the clearing price are classified according to the nodal pricing model into unit cost category, unit operation category, system operation category and cross-sectional power flow category. In step (2), the objective function and constraints of the full-time clearing simulation model of the electricity spot market considering carbon emission trading are determined, and their specific expressions are as follows: 1) Objective function The total power generation cost of the unit consists of carbon trading costs and operating costs; the expression for carbon trading costs is as follows: r co2i (t)=α(ε i P i (t)-κP i (t)) (1) In the formula, α represents the trading price of each ton of CO2 in the carbon market, and ε i κ represents the carbon emission factor of unit i, P represents the carbon allowance benchmark in the market, and κ represents the carbon emission factor of unit i. i (t) represents the power generation of unit i in time period t; The expression for the unit operating cost is as follows: ρ Gi (t)=a i (P i (t)) 2 +b i P i (t)+c i (2) In the formula, a i b i c i These are the parameters of the cost function for unit i; By integrating the carbon emission trading costs and operating costs of the unit, the overall cost curve of the unit is expressed as follows: C i (P i (t))=a i (P i (t)) 2 +[b i +a(e i -k)]P i (t)+c i (3) The overall cost curve is a quadratic function; in the actual electricity spot market, multiple bidding segments are usually allowed; the unit bidding curve is an increasing segmented step curve, and a maximum of m bidding segments can be submitted in each trading clearing period. It is required that the output range of each segment be connected end to end, and the start and end points of the output are the minimum stable technical output of the unit and the rated active power of the unit, respectively; the original overall cost curve is linearized. After linearly segmenting the original cost curve of a certain unit, the slopes from each power point to the next power point are k1, k2, k3, ..., k m Each slope is used as the bid value, and each power distribution point is used as the beginning and end of the bid output range, forming a stepped bidding curve based on the unit operating cost; because the operating cost function is a convex function, therefore k1 < k2 < k3 < ... < k m This meets the requirement that the price curve is monotonically non-decreasing; After considering the multiple bids from the generating units, the objective function of the clearing model is finally expressed as follows: Where, and These represent the quoted price and the winning bid capacity for the j-th segment of the i-th generating unit, respectively, with m being the number of segment bids for the generating units; 2) Constraints The clearing simulation model includes five constraints: system balance, system reserve, unit output, unit ramp-up, and cross-sectional power flow, as detailed below: a) System equilibrium constraints In the formula, λ(t) represents the shadow price of the system equilibrium constraint. d(t) represents the total output of all units in the system during time period t, and d(t) represents the system load during time period t; network loss factors are not considered here. b) System standby constraints Where, and v (t) represent the shadow prices of the upper and lower limits of the system's reserve capacity constraints, respectively. and These are the upper and lower limits of the output of unit i, respectively. and R (t) represents the system's increased and decreased spin-up reserves during time period t, respectively; c) Unit output constraints Where, and These are the shadow prices of the upper and lower limits of the unit's output, respectively. d) Unit climbing and landslide constraints Where, and These are the shadow prices for unit ramp-up and landslide constraints, respectively, Δ i This represents the maximum increase or decrease in output that unit i can achieve in each time period; e) Cross-sectional power flow constraints Where, and P represents the shadow price of the current and the downstream current constraint on the cross-section. l (t) represents the tidal power of section l during time period t. and Indicates the upper and lower limits of the power flow at section l; In step (3), the finite difference method is used to calculate the sensitivity of the clearing price; after calculating the change value of the average price corresponding to the change of a single operating parameter, the sensitivity coefficient of the price to each parameter is obtained to quantitatively measure the relative change between the two. Because the dimensions of different parameters are inconsistent, in order to make a unified comparison of the data, it is necessary to normalize the changes in the parameters. The normalization formula is as follows: In the formula, Δx is the normalized change, x after It is the value after the parameter changes, x original The parameter represents the original ground state value; The corresponding sensitivity analysis is calculated using the following formula in the difference scheme: In the formula, F k Δk is the sensitivity coefficient of electricity price under different conditions, Δk is the normalized change of the influencing parameter, and ΔA is the normalized change of the average node electricity price. To identify the key operating parameters affecting the clearing electricity price, a correlation diagram between the influencing parameters and the clearing electricity price is established for comparison. The specific process is as follows: a) Divide the entire day into peak load, off-peak load, and intermediate load periods according to load levels, and further calculate the sensitivity of the average nodal price of the entire network to each influencing parameter, i.e., the magnitude of the relative change. b) Based on the ground-state example, the trial-and-error method is used to find the upper and lower limits of each influencing factor that makes the clearing model solvable and changes the clearing price; for a single operating influencing parameter, six different scenarios are generated by taking six different influencing parameters within the upper and lower limit ranges with the same step size; the forward difference method is used to calculate the six sensitivities of the clearing price to this parameter, as shown in the following expressions: c) Assign scores to the original sensitivity data, and use a weighted average method to obtain the average score of the influence of each parameter. The mathematical expression for the average score at time t is as follows: In the formula, F kn (t) represents the sensitivity in the nth scene, α(F) kn (t) represents the sensitivity value when F kn The corresponding score for (t); The average value is taken to obtain the comprehensive score for different load levels during different periods, as shown in the following formula, for unified comparison, and finally to determine the key influencing parameters of the electricity spot market; In the formula, S H S M S L These represent the combined scores for peak load, mid-load, and trough load periods, respectively. H m M m L These represent the total number of periods during peak load, mid-load, and off-load periods, respectively.

Images