A subject capacity allocation method participating in a main energy and demand response multi-market

Abstract

The application discloses a kind of subject capacity allocation methods participating in main energy and demand response multi-market, by analyzing the time-varying quantity price characteristics of large industrial users and virtual power plant, capacity allocation model is constructed to participate in main energy market and demand side response market simultaneously, to realize market subject benefit maximization and system cost minimization.Specific steps include: market subject time-varying quantity price characteristics analysis, capacity allocation model is constructed, main problem solving, subproblem solving, decision check and adjustment, and cyclic iteration.Using Gale generation algorithm and cplex iterative solution, the scheduling strategy of each market subject is optimized, to ensure the coordination and optimization effect of overall system.Through the method, users can make optimal decisions under different market conditions, improve the demand response effect, promote the stable operation of power system and efficient use of renewable energy.The application improves the income of users, and also provides important support for the further development of power market.

Description

Technical Field

[0001] This invention relates to the field of power energy distribution technology, and in particular to a method for allocating capacity among entities participating in multiple markets for primary energy and demand response. Background Technology

[0002] In modern electricity markets, power systems are gradually moving towards open and competitive marketization. With the increasing proportion of renewable energy, the volatility and uncertainty of power systems are also intensifying. To address these issues, demand response (DR), as an effective regulatory tool, has received widespread attention and application. Demand response encourages users to adjust their electricity consumption behavior based on price signals or incentive mechanisms, playing a crucial role in improving power system flexibility, peak shaving and valley filling, and promoting the integration of renewable energy.

[0003] However, in existing demand response mechanisms, users often face the challenge of allocating capacity between the primary energy market and the demand response market. The primary energy market mainly guides users to adjust their electricity consumption through price signals from the electricity spot market, while the demand response market guides user participation through incentives provided by power companies or grid operators. Because the incentive mechanisms and operating rules of these two markets differ, users need to weigh the two to maximize benefits and optimize response effectiveness.

[0004] Current research and practice mainly focus on optimization strategies for single markets, with relatively little research on joint optimization problems involving both the main energy market and the demand response market. In particular, many challenges remain regarding how to effectively allocate user capacity resources, coordinate the demands and incentives of the two markets, and thus optimize overall system efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide a capacity allocation method for participants in multiple markets for primary energy and demand response, aiming to solve the aforementioned problems. By constructing a multi-market capacity allocation model, combined with analysis of the time-varying price characteristics of market participants and a hierarchical optimization strategy, the method achieves the goals of maximizing market participant revenue and minimizing system costs. This method not only helps users make optimal decisions under different market conditions but also promotes the stable operation of the power system and the efficient utilization of renewable energy.

[0006] To achieve the above objectives, the present invention is implemented according to the following technical solution:

[0007] This invention includes the following steps:

[0008] S1: Market entity time-varying price characteristics analysis: Analyze large industrial users and virtual power plants to understand their electricity consumption patterns, load characteristics, incentive policies and participation costs. Electricity users can maximize their total profit by optimizing electricity costs and response efficiency under different electricity price levels and incentive prices at different times.

[0009] S2: Construct a capacity allocation model: Establish a capacity allocation model that simultaneously participates in the main energy market and the demand-side response market to maximize the benefits of market participants and minimize system costs;

[0010] S3: Solving the main problem: Under the consideration of grid demand, achieve demand response market capacity allocation that minimizes peak shaving costs, obtain market prices and specific dispatch schemes, and use them as inputs for subproblems;

[0011] S4: Subproblem Solving: Each market participant, based on the clearing result of the main problem and the current spot electricity price, and in conjunction with the availability of internal resources, solves its profit maximization problem to determine the optimal action plan for the market participant under the current market conditions.

[0012] S5: Decision Check and Adjustment: Check the optimal decisions of each market participant, determine whether they will lead to a violation of the aggregate demand constraint in the main problem, or whether there are opportunities to further reduce costs. If necessary, generate new constraints or decision variables and feed them back to the main problem for re-solving.

[0013] S6: Iterative Loop: Solve the market optimization and scheduling strategies of each entity through the column generation algorithm and cplex iteration, and continue to iterate until the goal of global optimization is achieved.

[0014] The beneficial effects of this invention are:

[0015] This invention presents a capacity allocation method for multiple participants in the primary energy and demand response markets. It introduces demand response trading into the energy market's trading rules and discusses market trading decision-making methods in conjunction with time-varying market signals and the characteristics of different participants. Compared to existing technologies, this invention fully considers the unique characteristics of my country's power structure, scarce flexible resources, and significantly higher load-side regulation costs compared to power-side costs. Under a joint mode, the clearing price effectively incentivizes demand response users to participate. It also specifically considers the decision-making differences among different participants and the interactions and potential impacts among multiple participants. Unlike traditional technologies that focus on the impact of demand response on the primary energy market, this invention addresses the reverse impact of primary energy market capacity allocation results on the demand response market. Simulation examples demonstrate the clearing results and the revenue of each participant, providing a reference for future demand response implementation. Attached Figure Description

[0016] Figure 1This is a flowchart of the model solution process of this invention;

[0017] Figure 2 This is a daily load curve of the system;

[0018] Figure 3 It is a load curve diagram for each electricity user;

[0019] Figure 4 This is a chart showing the clearing price curves before and after the main energy market demand response;

[0020] Figure 5 This is a bar chart showing the bidding results in the main energy market;

[0021] Figure 6 This is a bar chart showing the bidding results in the demand response market. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions herein are used to explain the present invention, but are not intended to limit the present invention.

[0023] This invention includes time-varying price characteristic analysis of market participants and a capacity allocation model for their participation in multiple markets of main energy and demand-side response.

[0024] First, we conduct a time-varying price characteristic analysis on market participants, whose main participants include large industrial users and virtual power plants.

[0025] For large industrial users, their demand response capability mainly depends on their electricity consumption patterns and load characteristics. Different incentive policies and participation costs will affect their decision-making behavior in the market. By optimizing electricity costs and response benefits, electricity users can maximize their total profit under different electricity price levels and incentive prices at different times. Specifically, as shown in formulas (1)-(4): where formula (1) represents the pursuit of profit maximization by large industrial users participating in the main energy and demand response market, and formulas (2)-(4) represent the energy balance relationship and upper and lower limit constraints of large industrial users in multiple markets.

[0026]

[0027]

[0028] In the formula: U LD,j,t The user's productive utility is represented by electricity consumption P. LD,j,t Regarding the unit electricity output value v LD,j A linearly increasing function; These refer to the electricity volumes won in the main energy market and the demand response market, respectively. These are the clearing prices in the main energy market and the demand response market, respectively.

[0029] Based on the solution results of the above model, further analysis of B... LD,j Taking the first derivative, as shown in equation (5), we can obtain the price π from the demand-side response market of large industrial users. LD,j,t As shown in equation (6).

[0030]

[0031] π LD,j,t =v LD,j (6)

[0032] A virtual power plant is an aggregation of various flexible resources. It can participate in day-ahead energy market bidding and sell its adjustable capacity in the demand response market to obtain response compensation. The virtual power plant achieves autonomous, coordinated, and optimized control based on market clearing results. The aggregation units of a virtual power plant include gas turbines, electric vehicles, and photovoltaic units, which are uniformly controlled and managed by the virtual power plant operator. The following are model descriptions of each unit:

[0033] (1) Gas turbine:

[0034]

[0035] F i,t (μ i.t )=μ i.t (1-μ i,t-1 )S i (8)

[0036] Output upper and lower limits:

[0037]

[0038] Upper and lower limits of climbing:

[0039]

[0040] In the formula: P GT,i,t The output of the i-th gas turbine during time period t; a i b i c i Its coefficients are the quadratic function coefficients of the power generation cost; S i For the start-up and shutdown costs of unit i, μ i.t To characterize the 0-1 variables representing the start-up and shutdown status of unit i during time period t, u is the variable when the unit is in the start-up state. i,t =1, u when the unit is in a stopped state i,t =0;

[0041] (2) Photovoltaic output:

[0042]

[0043] (3) Electric vehicles

[0044] Electric vehicles require regular battery replacements; battery degradation costs:

[0045]

[0046] Where: N EV For the number of electric vehicles; The purchase cost of the m-th electric vehicle; I m S represents the number of battery charge / discharge cycles during the entire lifespan of the m-th electric vehicle; EV,m d m , Let m be the battery capacity, depth of discharge, and discharge efficiency of the m-th electric vehicle. l m,t Let m be the discharge power and travel distance of the m-th electric vehicle during time period t; This represents the power consumed per unit distance traveled by the m-th electric vehicle.

[0047] Battery capacity constraints:

[0048]

[0049] In the formula: S EV,m,t Let be the energy stored in the m-th electric vehicle during time period t; The upper and lower limits of its stored capacity;

[0050] Charging power constraints:

[0051]

[0052] In the formula: Let m be the charging power of the m-th electric vehicle during time period t; The upper limit of its charging power;

[0053] Discharge power constraint:

[0054]

[0055] In the formula: Let be the discharge power of the m-th electric vehicle during time period t; This is the upper limit of its discharge power;

[0056]

[0057] Electric vehicle battery capacity constraints:

[0058]

[0059] In the formula: Charging power

[0060] Virtual power plants typically operate during peak price periods in their demand response phase. At this time, to maximize profits, their internal power generation equipment is often already at full capacity. Therefore, virtual power plants can only participate in the response by reducing load, discharging energy storage devices, and discharging electric vehicles. These response methods are all internal control behaviors of the virtual power plant. From the perspective of the power grid, a virtual power plant behaves as an entity with flexible load characteristics. Its decision-making model is shown in equations (18)-(21). Equation (18) describes the goal of maximizing profits when participating in the main energy market and the demand response market. Equations (19)-(21) describe the price characteristics, cost characteristics, and energy balance relationships of the virtual power plant in multiple markets.

[0061]

[0062]

[0063] C VPP,t =C GT,t +C EV,t (20)

[0064]

[0065] In the formula: The revenue of virtual power plants in the demand response market and the main energy market are respectively equal to the product of the corresponding deployment volume and the market price. These are, respectively, the electricity volume won in the demand response market, and the electricity volume sold and purchased in the spot market; C VPP,t Adjusting operating costs for various resources, including gas turbine power generation costs and electric vehicle battery depreciation costs; π VPP,m,t The bids for their participation in demand response are for the charging and discharging power of electric vehicles, gas turbines, and photovoltaic power output. P L,t These are electricity sales, electricity purchases, and existing loads.

[0066] Secondly, we will study a capacity allocation model for large industrial users and virtual power plants participating in multiple markets, including main energy and demand-side response.

[0067] (1) Main energy market

[0068] Based on power generation company volume and price information, renewable energy forecasts, load forecasts, and projected load gaps, the day-ahead dispatch plan is determined, resulting in time-period electricity prices. The objective is to minimize total electricity costs. The capacity allocation model is as follows:

[0069] 1) Objective function

[0070]

[0071] In the formula: T is the total number of time periods; N G The number of units; S x The startup cost of unit x; u x,t It is a 0-1 variable characterizing the start-up and shutdown status of unit x during time period t; C G,x,t Let P be the active power of unit x during time period t. G,x,t The corresponding operating costs.

[0072] 2) Constraints

[0073] Members participating in demand response also participate in the clearing of the main energy market. Considering all market participants, the system constraints include:

[0074]

[0075] Equations (24)-(26) represent the system power balance constraint, generator constraint, and line power flow constraint, respectively. l,x G is the power flow transfer distribution factor of the node where unit x is located relative to line l; l,k N is the power flow transfer distribution factor of node k on line l; K N LD N VPP These represent the number of system nodes, the number of large industrial users, and the number of virtual power plants, respectively; D k,t Let be the bus load power of node k during time period t.

[0076] (2) Demand Response Market

[0077] The load reduction amount for each time period is determined based on grid demand. The winning bid volume for each participant is determined with the goal of minimizing demand response cost, and the volume is cleared at the marginal price. The capacity allocation model is as follows:

[0078] 1) Objective function

[0079]

[0080] 2) Constraints

[0081] The response quantities of all members participating in demand response should meet the grid demand. The power balance constraints at each time point are:

[0082]

[0083] In the formula: P DR,t Let t be the load gap in the power grid under network security constraints.

[0084] Daily load reduction limit:

[0085]

[0086] Considering that the cost of the response subsidy is ultimately shared by market users based on their electricity consumption, the clearing price λ in the demand response market... DR,t The following conditions must be met:

[0087]

[0088] In the formula: Q D To meet the demand response to the total electricity consumption of market-based users in the current month, λ lim This sets a price ceiling for market-based users.

[0089] Based on the time relationship between the day-ahead spot market and the demand response market, capacity allocation in the primary energy market is performed first. This allocation is the starting point and key to the entire process, determining the market's basic framework and operating rules. Using a generative algorithm, the clearing process of the demand response market and the decision-making problems of individual demand response users can be decomposed into a two-tiered optimization structure containing a main problem and sub-problems. The goal of the main problem is to achieve demand response market capacity allocation that minimizes peak-shaving costs while taking into account grid demand. Solving this main problem yields market prices and specific dispatch plans, which serve as inputs for subsequent sub-problems. In the sub-problem stage, each market participant, based on the clearing results of the main problem and the day-ahead spot price, combined with the availability of its internal resources, individually solves its revenue maximization problem. Market participants need to find the optimal production or consumption decision given the price and conditions. Through this step, each market participant can clarify its optimal course of action under current market conditions. To ensure the coordination and optimization effect of the overall system, the optimal decisions of each participant need to be examined to determine whether these decisions would violate the aggregate demand constraint in the main problem or whether there are opportunities for further cost reduction. If the solution to a subproblem results in a violation of the total demand constraint, or if potential for further cost reduction is discovered, new constraints or decision variables need to be generated and fed back into the main problem. The flowchart for solving the entire model is as follows: Figure 1 As shown.

[0090] The simulation example is based on an improved IEEE 30-bus system, which includes 6 generating units. User parameters are shown in Table 1. The virtual power plant is located at node 14, and the parameters for each unit are shown in Table 2. The daily load curve of the system is shown below. Figure 2 As shown, this curve integrates load data from both market participants and those not involved in demand response transactions.

[0091] Table 1 User parameters

[0092]

[0093] Table 2 Parameter settings for each unit

[0094]

[0095]

[0096] The electricity demand and production and operating costs of typical industrial users in a certain region were selected, and the unit electricity output value was calculated based on annual profit and electricity consumption, thereby obtaining the compensation price in the demand response market. To adapt to the system's capacity requirements, the electricity demand of users was adjusted proportionally. The load curves of each user and virtual power plant are shown below. Figure 3 As shown. The system capacity safety margin is set at 335MW, and this is used as the trigger condition for peak-shaving demand response. Based on system load forecasting and capacity gaps, day-ahead energy market clearing is carried out, and the corresponding electricity price changes are as follows. Figure 4 As shown in the figure. The capacity allocation results of the main energy market and the demand response market are respectively as follows. Figure 5 , Figure 6 As shown in the figure, the electricity demand is mainly met by conventional thermal power. The virtual power plant operates strategically, purchasing electricity during periods of low electricity prices and selling it during periods of high prices. The amount of electricity sold and its market placement depend on the relative levels of the electricity price and the incentive price, utilizing price fluctuations to optimize revenue and achieve economic optimization in energy management.

[0097] This invention proposes a hierarchical optimization scheme based on coordinated decomposition, which considers solving the market optimization and scheduling strategies of each entity through column-and-constraint generation (C&CG) algorithm and cplex iterative solution.

[0098] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A method for allocating capacity among multiple entities participating in primary energy and demand response markets, characterized in that, Includes the following steps: S1: Market entity time-varying price characteristics analysis: Analyze large industrial users and virtual power plants to understand their electricity consumption patterns, load characteristics, incentive policies and participation costs. Electricity users can maximize their total profit by optimizing electricity costs and response efficiency under different electricity price levels and incentive prices at different times. S2: Construct a capacity allocation model: Establish a capacity allocation model that simultaneously participates in the main energy market and the demand-side response market to maximize the benefits of market participants and minimize system costs; S3: Solving the main problem: Under the consideration of grid demand, achieve demand response market capacity allocation that minimizes peak shaving costs, obtain market prices and specific dispatch schemes, and use them as inputs for subproblems; S4: Subproblem Solving: Each market participant, based on the clearing result of the main problem and the current spot electricity price, and in conjunction with the availability of internal resources, solves its profit maximization problem to determine the optimal action plan for the market participant under the current market conditions. S5: Decision Check and Adjustment: Check the optimal decisions of each market participant, determine whether they will lead to a violation of the aggregate demand constraint in the main problem, or whether there are opportunities to further reduce costs. If necessary, generate new constraints or decision variables and feed them back to the main problem for re-solving. S6: Iterative Loop: The market optimization and scheduling strategies of each entity are solved iteratively through the Column-and-Constraint Generation (C&CG) algorithm and cplex, and the iteration continues until the goal of global optimization is achieved.

2. The method for allocating the capacity of participating entities in multiple markets for primary energy and demand response, as described in claim 1, is characterized in that: The specific steps S1 are shown in formulas (1)-(4): where formula (1) represents the pursuit of profit maximization by large industrial users participating in the main energy and demand response market, and formulas (2)-(4) represent the energy balance relationship and upper and lower limits of large industrial users in multiple markets. (1) (2) (3) (4) In the formula: The user's productive utility is expressed as electricity consumption. Regarding unit electricity output value A linearly increasing function; , These refer to the electricity volumes won in the main energy market and the demand response market, respectively. , These are the clearing prices in the main energy market and the demand response market, respectively. Based on the solution results, further... Taking the first derivative, as shown in equation (5), yields the price quote from the demand-side response market for large industrial users. As shown in equation (6); (5) (6) Models for each unit: gas turbine: (7) (8) Output upper and lower limits: (9) Upper and lower limits of climbing: (10) In the formula: The output of the i-th gas turbine during time period t; , , Its power generation cost quadratic function coefficients; The start-up and shutdown costs for unit i. To characterize the 0-1 variables representing the start-up and shutdown status of unit i during time period t, when the unit is in the start-up state... When the unit is in a shutdown state ; Photovoltaic output: (11) Electric vehicles: Electric vehicles require regular battery replacements; battery degradation costs: (12) In the formula: For the number of electric vehicles; Let m be the purchase cost of the m-th electric vehicle; The number of battery charge / discharge cycles during the entire lifespan of the m-th electric vehicle; , , Let m be the battery capacity, depth of discharge, and discharge efficiency of the m-th electric vehicle. , Let m be the discharge power and travel distance of the m-th electric vehicle during time period t; The power consumed per unit distance traveled by the m-th electric vehicle; Battery capacity constraints: (13) In the formula: Let be the energy stored in the m-th electric vehicle during time period t; , The upper and lower limits of its stored capacity; Charging power constraints: (14) In the formula: Let m be the charging power of the m-th electric vehicle during time period t; The upper limit of its charging power; Discharge power constraint: (15) In the formula: Let be the discharge power of the m-th electric vehicle during time period t; This is the upper limit of its discharge power; (16) Electric vehicle battery capacity constraints: (17) In the formula: Charging power The decision-making model of the virtual power plant is shown in formulas (18)-(21); formula (18) describes the goal of the virtual power plant to maximize profits when participating in the main energy market and the demand response market; formulas (19)-(21) describe the price characteristics, cost characteristics and energy balance relationship of the virtual power plant in multiple markets. (18) (19) (20) (21) In the formula: , The revenue of virtual power plants in the demand response market and the main energy market are respectively equal to the product of the corresponding deployment volume and the market price. , , These are, respectively, the electricity volume won in the demand response market, and the electricity volume sold and purchased in the spot market; Adjust costs for the operation of various resources, including the cost of gas turbine power generation and the cost of electric vehicle battery depreciation; The bids for their participation in demand response are for the charging and discharging power of electric vehicles, gas turbines, and photovoltaic power output. , , These are electricity sales, electricity purchases, and existing loads.

3. The method for allocating the capacity of participating entities in multiple markets for primary energy and demand response, as described in claim 2, is characterized in that: The main energy market capacity allocation model for step S2 is as follows: Objective function: （22） （23） In the formula: T is the total number of time periods; The number of generating units; The startup cost for unit x; It is a 0-1 variable that characterizes the start-up and shutdown status of unit x during time period t; The active power of unit x during time period t Corresponding operating costs; Constraints: Members participating in demand response also participate in the clearing of the main energy market. Considering all market participants, the system constraints include: （24） （25） （26） Equations (24)-(26) represent the system power balance constraint, the generator constraint, and the line power flow constraint, respectively. The power flow transfer distribution factor of the node where unit x is located relative to line l; Let be the power flow transfer distribution factor of node k on line l; , , These are the number of system nodes, the number of large industrial users, and the number of virtual power plants, respectively. Let be the bus load power of node k during time period t; The demand response market capacity allocation model is as follows: Objective function: (27) Constraints: The response quantities of all members participating in demand response should meet the grid demand. The power balance constraints at each time point are: (28) In the formula: The load gap in the power grid at time t is considered under network security constraints; Daily load reduction limit: (29) Considering that the cost of the response subsidy is ultimately shared by market users based on their electricity consumption, the clearing price of the demand response market... The following conditions must be met: (30) In the formula: The demand response is based on the total electricity consumption of market-based users in the current month. This sets a price ceiling for market-based users.

Images