Multi-energy virtual power plant cooperative peak regulation and standby auxiliary service market robust bidding method considering uncertainty and related device

Abstract

The invention relates to a multi-energy virtual power plant cooperative peak regulation and standby auxiliary service market robust bidding method considering uncertainty and a related device. The robust bidding method comprises the following steps: constructing an operation model of each system in a virtual power plant in a real-time operation stage of a multi-energy virtual power plant MEVPP system; an energy uncertainty variable is introduced, an uncertainty set is constructed, and a bidding model of a virtual power plant participating in peak regulation and a standby auxiliary service market in a day-ahead bidding stage of the MEVPP system is constructed; establishing an MEVPP two-stage robust optimization model by taking the operation model and the bidding model as constraint conditions; a combination category grammar Camp is adopted; solving the MEVPP two-stage robust optimization model through a CG algorithm, and obtaining a strategy that the MEVPP participates in market robust bidding and a real-time operation strategy. And various energy sources such as electric energy, heat energy, refrigeration energy and natural gas are further integrated, so that the system can flexibly participate in peak regulation and standby auxiliary service markets.

Description

Technical Field

[0001] The present invention relates to a method for virtual power plants to collaboratively participate in peak-shaving and standby ancillary service market bidding, specifically a robust bidding method and related devices for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service market considering uncertainty. Background Art

[0002] Generally speaking, the capacity of decentralized distributed energy is too small to meet the access conditions of the energy market. Virtual power plant (VPP) is an effective way to integrate various distributed energy, energy storage systems, flexible loads and other flexible resources to participate in the energy market using information and communication technology. With the development of integrated energy systems (IES), multi-energy virtual power plant (MEVPP) has emerged as a new concept. Compared with traditional VPP, MEVPP integrates multiple energy sources such as electricity, heat, cooling, natural gas, etc., and is suitable for participating in the energy market. In addition, in order to coordinate multiple energy sources and improve energy efficiency, the transformation of traditional demand response to integrated demand response is an unstoppable trend.

[0003] As the demand for electricity from users increases, the power system experiences peak loads in winter and summer, and the peak-to-valley difference continues to expand. Therefore, the peak-shaving and reserve pressure on the power grid continues to increase. MEVPP is a novel and flexible peak-shaving and reserve mode that can gather a large number of dispersed flexible resources. At present, relevant policies encourage third parties with energy storage equipment and demand-side resources to participate in providing auxiliary power services. In the future, VPP demonstration projects will be carried out, including industrial adjustable loads, building air-conditioning loads, big data center loads, energy storage and other resources. The goal is to achieve a demand-side response capacity of 3%-5% of the maximum load by 2025. Therefore, it is of great significance to design a collaborative bidding strategy for MEVPP under the peak-shaving market and the reserve market.

[0004] In order to cope with climate change, the installed capacity of renewable energy is increasing. But renewable energy is unstable. How to deal with the uncertainty of renewable energy is a key task for MEVPP to participate in the peak load market. In addition, compared with pure electric VPP, MEVPP is composed of different types of energy. Therefore, MEVPP must deal with multiple uncertainties from sources such as PV, power load, chilled water volume and outdoor temperature.

[0005] In order to solve this problem, CN118410982A provides a virtual power plant optimization economic dispatch method and system, which comprehensively considers renewable energy and demand-side loads to participate in the optimization dispatch of virtual power plants. However, this method does not take into account the challenges brought by the multiple uncertainties faced by VPPs during operation. CN114614463A proposes a reasonable trading strategy for virtual power plants to participate in the electricity market that takes into account market uncertainty and revenue risks. However, this method takes VPP as the research subject and does not consider the joint optimization problem of MEVPP.

[0006] The difficulty in considering uncertainty and optimizing the scheduling of MEVPP lies in the fact that different types of energy have their own physical characteristics and sources of uncertainty, such as the volatility of wind and solar energy, the charging and discharging efficiency of energy storage systems, and the elasticity of demand-side response. These uncertainties make it difficult for traditional optimization methods to adapt to complex MEVPP systems. In addition, there is a coupling relationship between multiple energy sources, such as the mutual conversion and influence between electricity, thermal energy and cooling energy, which increases the complexity of the system. How to maximize the economic benefits and energy efficiency of MEVPP while ensuring system stability is a very challenging task. On the other hand, in the scenario of MEVPP collaborative operation, the coordination and competition relationship between different virtual power plants has also become a major problem in optimizing scheduling. Each virtual power plant may have different resource structures and operating objectives. How to maximize the overall system benefits on the basis of ensuring fairness is a problem that needs to be studied in depth. At the same time, different MEVPPs may be subject to different market rules, policies and technical standards, which makes the design of a unified optimization model more complicated. Summary of the invention

[0007] The purpose of the present invention is to solve the deficiencies of the above-mentioned prior art, thereby providing a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service markets taking into account uncertainty.

[0008] The present invention also aims to provide a robust bidding device, system and medium for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty.

[0009] A robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and reserve ancillary service markets considering uncertainty includes the following steps:

[0010] Construct the operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system;

[0011] By introducing energy uncertainty variables and constructing uncertainty sets, a bidding model for virtual power plants to participate in the peak load regulation and reserve ancillary service markets in the day-ahead bidding phase of the MEVPP system is constructed.

[0012] Taking the operation model and bidding model as constraints, a two-stage robust optimization model of MEVPP is established;

[0013] The combinatorial category grammar C&CG algorithm is used to solve the two-stage robust optimization model of MEVPP, and the robust bidding strategy and real-time operation strategy of MEVPP in the market are obtained.

[0014] The MEVPP system includes an energy storage system, flexible loads, and combined cooling, heating and power units.

[0015] Construct the operation model of each system in the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system, specifically:

[0016] On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power load is a transferable load, which is expressed as:

[0017]

[0018] Where t is the time interval index; N T is the total number of time intervals; is the optimized electrical load; is the initial electrical load; is the electrical load delivered; The upper limit of the power for transferring load;

[0019] Flexible heat load is hot water load, and the user's ideal range of water temperature is expressed as:

[0020]

[0021] Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the heat energy required to set the water temperature;

[0022] The acceptable range of indoor temperature for users in flexible cooling load is:

[0023]

[0024] In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; is the cooling load, which is the cooling energy required to set the indoor temperature;

[0025] On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide unit;

[0026] The electricity generated by the gas turbine is:

[0027]

[0028] In the formula Electricity generated by gas turbines; and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; is the natural gas consumed by the gas turbine;

[0029] The lithium bromide unit model is:

[0030]

[0031] In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable;

[0032] The heat energy generated by the heat pump is expressed as:

[0033]

[0034] In the formula The heat energy generated by the heat pump; COP HP and are the coefficient of performance and rated output power of the heat pump respectively;

[0035] The cooling energy generated by a centrifugal chiller is shown below:

[0036]

[0037] In the formula The cooling energy generated by the centrifugal chiller; COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; is the amount of electricity consumed by the centrifugal chiller;

[0038] The power storage system model is as follows:

[0039]

[0040] In the formula, l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

[0041] The supply and demand balance and ancillary service market constraints in the operation model include:

[0042] Supply and demand balance constraints for electricity, heat and cooling:

[0043]

[0044] In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms;

[0045] Interaction power constraints between MEVPP and the grid:

[0046]

[0047] In the formula A cap on the amount of electricity that can be purchased from the grid;

[0048] Peak shaving constraints:

[0049]

[0050] In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-load-and-valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding capacity;

[0051] Alternative constraints:

[0052]

[0053] In the formula and They represent the upper reserve and lower reserve capacities at time T respectively; the total reserve capacity reported by the virtual power plant in the reserve market within k time slots; and Respectively represent the upper and lower limits of the power grid.

[0054] By introducing energy uncertainty variables and constructing uncertainty sets, a bidding model for virtual power plants to participate in the peak load regulation and reserve ancillary service markets in the day-ahead bidding phase of the MEVPP system is constructed. Specifically:

[0055] During the MEVPP day-ahead bidding phase, the uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are:

[0056]

[0057] Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power, respectively. To predict the nominal value, To predict the maximum deviation, the band and is a binary variable, using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model;

[0058] Introducing uncertainty variables into the MEVPP day-ahead bidding model:

[0059]

[0060] The objective function of the MEVPP two-stage robust optimization model is:

[0061]

[0062] in for energy costs; is the compensation cost for users to adjust load; The operating cost of energy storage; It is the income from the sale of energy; For peak load benefits; For reserve income; and is the calorific value and price of natural gas; Grid means the electricity price comes from the grid; c l is the operating cost coefficient of energy storage;

[0063] Considering only the compensation cost of the electric load and ignoring the compensation cost of the cooling load and the heating load, the compensation cost of the user's load regulation is expressed as:

[0064]

[0065] In the formula Compensation prices for regulating electricity load; is an auxiliary variable;

[0066] Energy revenue from selling energy is expressed as:

[0067]

[0068] In the formula and Respectively represent the prices of cooling, heating and electricity; and They represent the amount of electricity traded for cooling and heat energy respectively;

[0069] Peak load benefits and reserve income Respectively expressed as:

[0070]

[0071] in and They are the peak-shaving price and valley-filling price in the peak-shaving market; and They are the upper standby price and the lower standby price in the standby market respectively.

[0072] The combined category grammar C&CG algorithm is used to solve the two-stage robust optimization model of MEVPP, and the robust bidding strategy and real-time operation strategy of MEVPP in the market are obtained, which are as follows:

[0073] Rewrite the MEVPP two-stage robust optimization model into matrix form:

[0074]

[0075] Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, λ, v and π are dual variables; the uncertainty set is:

[0076]

[0077] The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem.

[0078] The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem, specifically:

[0079] The duality theory is used to transform the inner minimum problem of the MEVPP two-stage robust optimization model rewritten in matrix form into a dual problem and then merged with the mid-extreme problem, which is expressed as:

[0080]

[0081] Assume v = 0, UB v →+∞,LB v →-∞, initialize x (0) ;

[0082] Decompose formula A3 into a sub-problem A4 and a main problem A5:

[0083]

[0084] Update v = v + 1 and solve the subproblem;

[0085] According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), and update the upper bound:

[0086] UB v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (z + ) (v) -(Δu max ) T (z - ) (v) A6

[0087] Based on the obtained u (k) (k=1,…,v), solve the main problem;

[0088] According to the solution of the main problem, we can get x (v) , update the lower bound:

[0089] LB v =η (v) A7

[0090] If UB v -LB v ≤ε, then output x (v) ,y(v) and u (v) .

[0091] A robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and reserve ancillary service markets considering uncertainty, comprising:

[0092] An operation model building module, wherein the operation model building module is used to build an operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system;

[0093] A bidding model building module, wherein the bidding model building module is used to build a bidding model for the virtual power plant to participate in the peak load regulation and standby auxiliary service market in the day-ahead bidding stage of the MEVPP system by introducing energy uncertainty variables and building an uncertainty set;

[0094] An optimization model building module, wherein the optimization model building module is used to establish a two-stage robust optimization model of MEVPP with an operation model and a bidding model as constraints;

[0095] and a solution module, wherein the solution module is used to solve the MEVPP two-stage robust optimization model by using the combined category grammar C&CG algorithm to obtain the MEVPP participation market robust bidding strategy and real-time operation strategy.

[0096] Run the model building module specifically for:

[0097] On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power load is a transferable load, which is expressed as:

[0098]

[0099] Where t is the time interval index; N T is the total number of time intervals; is the optimized electrical load; is the initial electrical load; is the electrical load delivered; The upper limit of the power for transferring load;

[0100] Flexible heat load is hot water load, and the user's ideal range of water temperature is expressed as:

[0101]

[0102] Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the heat energy required to set the water temperature;

[0103] The acceptable range of indoor temperature for users in flexible cooling load is:

[0104]

[0105] In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; is the cooling load, which is the cooling energy required to set the indoor temperature;

[0106] On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide unit;

[0107] The electricity generated by the gas turbine is:

[0108]

[0109] In the formula Electricity generated by gas turbines; and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; is the natural gas consumed by the gas turbine;

[0110] The lithium bromide unit model is:

[0111]

[0112] In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable;

[0113] The heat energy generated by the heat pump is expressed as:

[0114]

[0115] In the formula The heat energy generated by the heat pump; COP HP and are the coefficient of performance and rated output power of the heat pump respectively;

[0116] The cooling energy generated by a centrifugal chiller is shown below:

[0117]

[0118] In the formula The cooling energy generated by the centrifugal chiller; COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; is the amount of electricity consumed by the centrifugal chiller;

[0119] The power storage system model is as follows:

[0120]

[0121]

[0122] Where l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; S l t is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

[0123] The supply and demand balance and ancillary service market constraints in the operation model include:

[0124] Supply and demand balance constraints for electricity, heat and cooling:

[0125]

[0126] In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms;

[0127] Interaction power constraints between MEVPP and the grid:

[0128]

[0129] In the formula A cap on the amount of electricity that can be purchased from the grid;

[0130] Peak shaving constraints:

[0131]

[0132] In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-load-and-valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding capacity;

[0133] Alternative constraints:

[0134]

[0135] In the formula and They represent the upper reserve and lower reserve capacities at time T respectively; the total reserve capacity reported by the virtual power plant in the reserve market within k time slots; and Respectively represent the upper and lower limits of the power grid.

[0136] The bidding model building module is specifically used for:

[0137] During the MEVPP day-ahead bidding phase, the uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are:

[0138]

[0139] Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power, respectively. To predict the nominal value, To predict the maximum deviation, the band and is a binary variable, using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model;

[0140] Introducing uncertainty variables into the MEVPP day-ahead bidding model:

[0141]

[0142] The optimization model building module is specifically used to:

[0143] Rewrite the MEVPP two-stage robust optimization model into matrix form:

[0144]

[0145] Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, λ, v and π are dual variables; the uncertainty set is:

[0146]

[0147] The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem.

[0148] The Optimization Model building block is also used to:

[0149] Using the duality theory, the inner minimum problem of the MEVPP two-stage robust optimization model rewritten in matrix form is transformed into a dual problem and then merged with the mid-extreme problem, which is expressed as:

[0150]

[0151] Assume v = 0, UB v →+∞,LB v →-∞, initialize x (0) ;

[0152] Decompose formula A3 into a sub-problem A4 and a main problem A5:

[0153]

[0154] Update v = v + 1 and solve the subproblem;

[0155] According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), and update the upper bound:

[0156] UB v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (z + ) (v) -(Δu max ) T (z - ) (v) A6

[0157] Based on the obtained u (k) (k=1,…,v), solve the main problem;

[0158] According to the solution of the main problem, we can get x (v) , update the lower bound:

[0159] LB v =η (v) A7

[0160] If UB v -LB v ≤ε, then output x (v) ,y (v) and u (v) .

[0161] A robust bidding system for multi-energy virtual power plants to collaboratively participate in peak load regulation and reserve ancillary service markets considering uncertainty, comprising: a computer-readable storage medium and a processor;

[0162] The computer-readable storage medium is used to store executable instructions;

[0163] The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the robust bidding method for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service markets taking into account uncertainty as described in the above scheme.

[0164] A computer-readable storage medium, characterized in that: the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the robust bidding method for multi-energy virtual power plants that consider uncertainty in the above-mentioned scheme to collaboratively participate in peak-shaving and standby ancillary service markets.

[0165] Compared with the prior art, the present invention takes into account the integration of multiple energy sources in virtual power plants and the bidding strategy for flexible and collaborative participation in the peak-shaving and standby ancillary service markets. In the current energy system, the capacity of decentralized distributed energy is usually small, and it is difficult to meet the access requirements of the energy market. The virtual power plant (VPP) integrates distributed energy, energy storage systems and flexible load resources through information and communication technology, and the multi-energy virtual power plant (MEVPP) proposed in the present invention further integrates multiple energy sources such as electricity, heat, cooling energy and natural gas, enabling it to flexibly participate in the peak-shaving and standby ancillary service markets.

[0166] In addition, the present invention takes into account the uncertainties faced by virtual power plants in the real-time operation stage. In view of the volatility of renewable energy and the physical characteristics of different energy types, MEVPP faces multiple uncertainty challenges. The present invention solves how to optimize the scheduling and bidding strategies under these uncertain conditions to ensure the stability and economic benefits of the system.

[0167] The present invention proposes a bidding model based on robust optimization to deal with the coupling problem between multiple energy sources in the MEVPP system, and achieves the optimal scheduling of the system by coordinating multiple energy sources such as electricity, heat, refrigeration and natural gas. DETAILED DESCRIPTION

[0168] In order to make the implementation and advantages of the technical method of the present invention clear, the implementation process is described in detail below.

[0169] In a first aspect, an embodiment of the present invention provides a robust bidding method for a multi-energy virtual power plant to collaboratively participate in a peak load regulation and standby ancillary service market taking into account uncertainty, the method comprising the following steps:

[0170] Construct the operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system;

[0171] The MEVPP system includes energy storage system, flexible load, combined heating and cooling power unit, etc., which can be integrated into MEVPP.

[0172] On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power loads are mainly transferable loads.

[0173]

[0174] In the formula, is the optimized electrical load; is the initial electrical load: It is the electrical load that is transmitted. The flexible heat load is mainly the hot water load. The user has an ideal range for the water temperature.

[0175]

[0176] Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the thermal energy required to set the water temperature. The flexible cooling load is similar to the heat load. Users have an acceptable range for indoor temperature.

[0177]

[0178] In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; The cooling load is the cooling energy required to set the indoor temperature.

[0179] On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide device. The electricity generated by the gas turbine is:

[0180]

[0181] In the formula and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; It is the natural gas consumed by the gas turbine.

[0182] The lithium bromide unit model is obtained from (8)-(11).

[0183]

[0184] In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable.

[0185] The heat energy generated by the heat pump is expressed as:

[0186]

[0187] Where COP HP and are the coefficient of performance and rated output power of the heat pump respectively.

[0188] The cooling energy generated by a centrifugal chiller is shown below:

[0189]

[0190] In the formula, COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; It is the amount of electricity consumed by the centrifugal chiller.

[0191] The power storage system model is as follows:

[0192]

[0193]

[0194] In the formula, l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

[0195] The supply and demand balance constraints of electricity, heat and cooling are shown in equations (20) to (22):

[0196]

[0197] In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms.

[0198] The interaction power constraint between MEVPP and the grid is:

[0199]

[0200] Peak shaving constraints:

[0201]

[0202] In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-load-and-valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding ability.

[0203] Alternative constraints:

[0204]

[0205] In the formula and They represent the upper reserve capacity and lower reserve capacity at time T respectively; it is assumed that in the reserve market the virtual power plant reports the total capacity within each k time granularity period.

[0206] By introducing energy uncertainty variables and constructing uncertainty sets, a bidding model for virtual power plants to participate in the peak load regulation and reserve ancillary service markets in the day-ahead bidding phase of the MEVPP system is constructed.

[0207] The uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are:

[0208]

[0209] Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power respectively. To predict the nominal value, is the maximum prediction deviation. and is a binary variable. Using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model.

[0210] Introducing uncertainty variables into the MEVPP day-ahead bidding model:

[0211]

[0212] Taking the operation model and bidding model as constraints, a two-stage robust optimization model of MEVPP is established;

[0213] The objective function of the MEVPP two-stage robust optimization model is:

[0214]

[0215] in for energy costs; is the compensation cost for users to adjust load; The operating cost of energy storage; It is the income from the sale of energy; For peak load benefits; For reserve income; and is the calorific value and price of natural gas; grid means the electricity price comes from the grid; is the operating cost coefficient of energy storage.

[0216] Cooling and heating can meet the needs of users to adjust indoor temperature and hot water temperature. Therefore, the compensation cost of cooling load and heating load is not considered.

[0217]

[0218] In the formula Compensation prices for regulating electricity load; is an auxiliary variable. The energy income from selling energy consists of three parts, as shown in (31). It is worth noting that the energy cost of users should not increase.

[0219]

[0220] The benefits of participating in the peak-shaving market include peak-shaving and valley-filling benefits.

[0221]

[0222] in and They are the peak-shaving price and valley-filling price in the peak-shaving market; and They are the upper standby price and the lower standby price in the standby market respectively.

[0223] MEVPP two-stage robust optimization model

[0224]

[0225] The combined category grammar C&CG algorithm is used to solve the two-stage robust optimization model of MEVPP, and the robust bidding strategy and real-time operation strategy of MEVPP in the market are obtained;

[0226] The two-stage robust bidding model is rewritten into the following matrix form:

[0227]

[0228] Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, respectively, λ, v and π are dual variables. The uncertainty set is as follows:

[0229]

[0230] The C&CG algorithm is used to transform the two-stage robust problem into a main problem and a sub-problem to solve the robust problem.

[0231] First, the inner minimum problem is transformed into a dual problem using duality theory, and then it is combined with the mid-maximum problem.

[0232]

[0233] Then initialize the problem, set v = 0, UB v →+∞,LB v →-∞, initialize x (0) ;

[0234] Decompose formula A3 into a sub-problem A4 and a main problem A5 A4

[0235]

[0236] where x (v-1) is a known value and is optimized in the main problem; + and z - is an auxiliary variable; M is a given maximum value.

[0237] Update v = v + 1 to the given x (v-1) Solve subproblem A4.

[0238] According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), update the upper bound:

[0239] UB v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (z + ) (v) -(Δu max ) T (z - ) (v) A6

[0240] Based on the obtained u (k) (k=1,…,v), solve the main problem A6.

[0241] According to the solution of the main problem, we can get x (v) , update the lower bound:

[0242] LB v =η (v) A7

[0243] Finally, if UB v -LB v ≤ε, then output x (v) ,y (v) and u (v) ; Otherwise continue to update v=v+1 and solve the problem.

[0244] Example

[0245] Table 1 Day-ahead economic analysis of demand-side resources without considering uncertainty (RMB / day)

[0246]

[0247] In Case 1 of Table 1, without integrated demand response and energy storage systems, MEVPP cannot participate in the ancillary service market and has the lowest total revenue. Compared with Cases 2 and 3, although more load compensation and energy storage costs are spent, Case 4, which uses an integrated demand response and energy storage system at the same time, has the highest peak-shaving revenue. In addition, by coordinating integrated demand response and energy storage systems, energy costs can be reduced, ultimately increasing total revenue from the previous day.

[0248] Within the range of prediction deviation, a set of data is randomly selected as the actual value of the uncertainty variable. The peak-shaving penalty price is twice the peak-shaving compensation price. Sunny and cloudy days are similar. When the prediction deviation is small (0% and 5%), the actual total benefit of robust optimization is less than that of deterministic optimization. The reason is that robust optimization is conservative. Its day-ahead bidding ability is less than that of deterministic optimization, and the peak-shaving benefit is reduced. However, when the deviation increases, its peak-shaving penalty is less than that of deterministic optimization, so its actual total benefit is greater than that of deterministic optimization. In summary, due to the conservatism of robust optimization, its day-ahead bidding ability becomes smaller and its day-ahead total benefit decreases. However, in the actual operation stage, there are prediction deviations for photovoltaics and loads. Compared with deterministic optimization, robust optimization can obtain more benefits in the actual operation stage.

[0249] In terms of model convergence, the C&CG algorithm is used to solve the day-ahead robust bidding model. It takes 5 iterations on a sunny day and 4 iterations on a cloudy day to meet the convergence requirements. As the number of iterations increases, the lower bound first increases and then remains unchanged, which is a characteristic of the C&CG algorithm.

[0250] In addition, during the day-ahead bidding phase, as Γ increases, the benefits decrease. In addition, during the actual operation phase, when the forecast deviation is small (0% and 5%), the benefits decrease. When the deviation increases to 10% and 20%, the benefits begin to increase, then decrease, and finally level off. When the deviation increases to 30%, the benefits first increase and then level off. Therefore, the multi-energy virtual power plant operator (MEVPPO) can choose the benefits and risks by adjusting Γ.

[0251] In a second aspect, the present invention provides a robust bidding device for a multi-energy virtual power plant to collaboratively participate in a peak load regulation and standby ancillary service market taking into account uncertainty, comprising:

[0252] An operation model building module, wherein the operation model building module is used to build an operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system;

[0253] A bidding model building module, wherein the bidding model building module is used to build a bidding model for the virtual power plant to participate in the peak load regulation and standby auxiliary service market in the day-ahead bidding stage of the MEVPP system by introducing energy uncertainty variables and building an uncertainty set;

[0254] An optimization model building module, wherein the optimization model building module is used to establish a two-stage robust optimization model of MEVPP with an operation model and a bidding model as constraints;

[0255] and a solution module, wherein the solution module is used to solve the MEVPP two-stage robust optimization model by using the combined category grammar C&CG algorithm to obtain the MEVPP participation market robust bidding strategy and real-time operation strategy.

[0256] Run the model building module specifically for:

[0257] On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power load is a transferable load, which is expressed as:

[0258]

[0259] Where t is the time interval index; N T is the total number of time intervals; is the optimized electrical load; is the initial electrical load; is the electrical load delivered; The upper limit of the power for transferring load;

[0260] Flexible heat load is hot water load, and the user's ideal range of water temperature is expressed as:

[0261]

[0262] Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the heat energy required to set the water temperature;

[0263] The acceptable range of indoor temperature for users in flexible cooling load is:

[0264]

[0265] In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; is the cooling load, which is the cooling energy required to set the indoor temperature;

[0266] On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide unit;

[0267] The electricity generated by the gas turbine is:

[0268]

[0269] In the formula Electricity generated by gas turbines; and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; is the natural gas consumed by the gas turbine;

[0270] The lithium bromide unit model is:

[0271]

[0272] In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable;

[0273] The heat energy generated by the heat pump is expressed as:

[0274]

[0275] In the formula The heat energy generated by the heat pump; COP HP and are the coefficient of performance and rated output power of the heat pump respectively;

[0276] The cooling energy generated by a centrifugal chiller is shown below:

[0277]

[0278] In the formula The cooling energy generated by the centrifugal chiller; COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; is the amount of electricity consumed by the centrifugal chiller;

[0279] The power storage system model is as follows:

[0280]

[0281] In the formula, l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

[0282] The supply and demand balance and ancillary service market constraints in the operation model include:

[0283] Supply and demand balance constraints for electricity, heat and cooling:

[0284]

[0285]

[0286] In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms;

[0287] Interaction power constraints between MEVPP and the grid:

[0288]

[0289] In the formula A cap on the amount of electricity that can be purchased from the grid;

[0290] Peak shaving constraints:

[0291]

[0292] In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-load-and-valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding capacity;

[0293] Alternative constraints:

[0294]

[0295] In the formula and They represent the upper reserve and lower reserve capacities at time T respectively; the total reserve capacity reported by the virtual power plant in the reserve market within k time slots; and Respectively represent the upper and lower limits of the power grid.

[0296] The bidding model building module is specifically used for:

[0297] During the MEVPP day-ahead bidding phase, the uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are:

[0298]

[0299] Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power, respectively. To predict the nominal value, To predict the maximum deviation, the band and is a binary variable, using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model;

[0300] Introducing uncertainty variables into the MEVPP day-ahead bidding model:

[0301]

[0302] The optimization model building module is specifically used to:

[0303] Rewrite the MEVPP two-stage robust optimization model into matrix form:

[0304]

[0305] Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, λ, v and π are dual variables; the uncertainty set is:

[0306]

[0307] The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem.

[0308] The Optimization Model building block is also used to:

[0309] First, the inner minimum problem of the MEVPP two-stage robust optimization model rewritten in matrix form is transformed into a dual problem by using the duality theory and then merged with the middle extreme problem, which is expressed as:

[0310]

[0311] Then initialize the problem, set v = 0, UB v →+∞,LB v →-∞, initialize x (0) ;

[0312] Decompose formula A3 into a sub-problem A4 and a main problem A5:

[0313]

[0314] Update v = v + 1 and solve the subproblem;

[0315] According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), and update the upper bound:

[0316] UB v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (z + ) (v) -(Δu max ) T (z - ) (v) A6

[0317] Based on the obtained u (k) (k=1,…,v), solve the main problem;

[0318] According to the solution of the main problem, we can get x (v) , update the lower bound:

[0319] LB v =η (v) A7

[0320] If UB v -LBv ≤ε, then output x (v) ,y (v) and u (v) .

[0321] In a third aspect, the present invention provides a robust bidding system for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty, including: a computer-readable storage medium and a processor;

[0322] The computer-readable storage medium is used to store executable instructions;

[0323] The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the robust bidding method for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service markets taking into account uncertainty as described in the above scheme.

[0324] In the fourth aspect, the present application proposes a computer-readable storage medium, characterized in that: the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the robust bidding method for multi-energy virtual power plants that consider uncertainty in the above-mentioned scheme to collaboratively participate in peak-shaving and standby ancillary service markets.

[0325] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0326] The present application is described with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, and the combination of the process and / or box in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and / or one box or multiple boxes in the block diagram.

[0327] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0328] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0329] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets considering uncertainty, characterized by: The following steps are involved: Construct the operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system; By introducing energy uncertainty variables and constructing uncertainty sets, a bidding model for virtual power plants to participate in the peak load regulation and reserve ancillary service markets in the day-ahead bidding phase of the MEVPP system is constructed. Taking the operation model and bidding model as constraints, a two-stage robust optimization model of MEVPP is established; The combinatorial category grammar C&CG algorithm is used to solve the two-stage robust optimization model of MEVPP, and the robust bidding strategy and real-time operation strategy of MEVPP in the market are obtained.

2. According to claim 1, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: The MEVPP system includes an energy storage system, flexible loads, and combined cooling, heating and power units.

3. According to claim 1, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: Construct the operation model of each system in the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system, specifically: On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power load is a transferable load, which is expressed as: Where t is the time interval index; N T is the total number of time intervals; is the optimized electrical load; is the initial electrical load; is the electrical load delivered; The upper limit of the power for transferring load; Flexible heat load is hot water load, and the user's ideal range of water temperature is expressed as: Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the heat energy required to set the water temperature; The acceptable range of indoor temperature for users in flexible cooling load is: In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; is the cooling load, which is the cooling energy required to set the indoor temperature; On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide unit; The electricity generated by the gas turbine is: In the formula Electricity generated by gas turbines; and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; is the natural gas consumed by the gas turbine; The lithium bromide unit model is: In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable; The heat energy generated by the heat pump is expressed as: In the formula The heat energy generated by the heat pump; COP HP and are the coefficient of performance and rated output power of the heat pump respectively; The cooling energy generated by a centrifugal chiller is shown below: In the formula The cooling energy generated by the centrifugal chiller; COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; is the amount of electricity consumed by the centrifugal chiller; The power storage system model is as follows: In the formula, l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

4. A robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets considering uncertainty according to claim 1 or 3, characterized in that: The supply and demand balance and ancillary service market constraints in the operation model include: Supply and demand balance constraints for electricity, heat and cooling: In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms; Interaction power constraints between MEVPP and the grid: In the formula A cap on the amount of electricity that can be purchased from the grid; Peak shaving constraints: In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-shaving and valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding capacity; Alternative constraints: In the formula and They represent the upper reserve and lower reserve capacities at time T respectively; the total reserve capacity reported by the virtual power plant in the reserve market within k time slots; and Respectively represent the upper and lower limits of the power grid.

5. According to claim 1, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: By introducing energy uncertainty variables and constructing uncertainty sets, a bidding model for virtual power plants to participate in the peak load regulation and reserve ancillary service markets in the day-ahead bidding phase of the MEVPP system is constructed. Specifically: During the MEVPP day-ahead bidding phase, the uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are: Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power, respectively. To predict the nominal value, To predict the maximum deviation, the band and is a binary variable, using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model; Introducing uncertainty variables into the MEVPP day-ahead bidding model:

6. According to claim 1, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: The objective function of the MEVPP two-stage robust optimization model is: in for energy costs; is the compensation cost for users to adjust load; The operating cost of energy storage; It is the income from the sale of energy; For peak load benefits; For reserve income; and is the calorific value and price of natural gas; Grid means the electricity price comes from the grid; c l is the operating cost coefficient of energy storage; Considering only the compensation cost of the electric load and ignoring the compensation cost of the cooling load and the heating load, the compensation cost of the user's load regulation is expressed as: In the formula Compensation prices for regulating electricity load; is an auxiliary variable; Energy revenue from selling energy is expressed as: In the formula and Respectively represent the prices of cooling, heating and electricity; and They represent the amount of electricity traded for cooling and heat energy respectively; Peak load benefits and reserve income Respectively expressed as: in and They are the peak-shaving price and valley-filling price in the peak-shaving market; and They are the upper standby price and the lower standby price in the standby market respectively.

7. According to claim 1, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: The combined category grammar C&CG algorithm is used to solve the two-stage robust optimization model of MEVPP, and the robust bidding strategy and real-time operation strategy of MEVPP in the market are obtained, which are as follows: Rewrite the MEVPP two-stage robust optimization model into matrix form: Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, λ, v and π are dual variables; the uncertainty set is: The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem.

8. According to claim 7, a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby ancillary service markets taking into account uncertainty is characterized by: The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem, specifically: The duality theory is used to transform the inner minimum problem of the MEVPP two-stage robust optimization model rewritten in matrix form into a dual problem and then merged with the mid-extreme problem, which is expressed as: Assume v = 0, UB v →+∞,LB v →-∞, initialize x (0) ; Decompose formula A3 into a sub-problem A4 and a main problem A5: Update v = v + 1 and solve the subproblem; According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), and update the upper bound: Security Service v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (With + ) (v) -(Δu max ) T (With - ) (v) A6 Based on the obtained u (k) (k=1,…,v), solve the main problem; According to the solution of the main problem, we can get x (v) , update the lower bound: <h2 style=";text-align:left;direction:ltr">LB<h2 style=";text-align:left;direction:ltr"> v <h2 style=";text-align:left;direction:ltr"> =η<h2 style=";text-align:left;direction:ltr"> (v) <h2 style=";text-align:left;direction:ltr"> A7 If UB v -LB v ≤ε, then output x (v) ,y (v) and u (v) .

9. A robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby auxiliary service markets taking into account uncertainty, characterized by: include: An operation model building module, wherein the operation model building module is used to build an operation model of each system inside the virtual power plant during the real-time operation phase of the multi-energy virtual power plant MEVPP system; A bidding model building module, wherein the bidding model building module is used to build a bidding model for the virtual power plant to participate in the peak load regulation and standby auxiliary service market in the day-ahead bidding stage of the MEVPP system by introducing energy uncertainty variables and building an uncertainty set; An optimization model building module, wherein the optimization model building module is used to establish a two-stage robust optimization model of MEVPP with an operation model and a bidding model as constraints; and a solution module, wherein the solution module is used to solve the MEVPP two-stage robust optimization model by using the combined category grammar C&CG algorithm to obtain the MEVPP participation market robust bidding strategy and real-time operation strategy.

10. A robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and backup ancillary service markets considering uncertainty according to claim 9, characterized in that: Run the model building module specifically for: On the demand side, integrated demand response includes demand response for electricity, cooling and heat. Flexible power load is a transferable load, which is expressed as: Where t is the time interval index; N T is the total number of time intervals; is the optimized electrical load; is the initial electrical load; is the electrical load delivered; The upper limit of the power for transferring load; Flexible heat load is hot water load, and the user's ideal range of water temperature is expressed as: Where c w and ρ w are the specific heat capacity and density of water respectively; is the volume of cold water; is the initial water temperature; and They are the minimum and maximum water temperatures expected by the user: The heat load is the heat energy required to set the water temperature; The acceptable range of indoor temperature for users in flexible cooling load is: In the formula and are the minimum and maximum indoor temperatures expected by the user, respectively; is the outdoor temperature; R is the thermal resistance of the house; is the cooling load, which is the cooling energy required to set the indoor temperature; On the supply side, the combined cooling, heating and power unit consists of a gas turbine and a lithium bromide unit; The electricity generated by the gas turbine is: In the formula Electricity generated by gas turbines; and are the rated capacity of the gas turbine and the natural gas-to-electricity efficiency, respectively; is the natural gas consumed by the gas turbine; The lithium bromide unit model is: In the formula, and are the heat and cooling energy generated by the lithium bromide device, respectively; and The conversion efficiency of flue gas waste heat into thermal and cooling energy; is the conversion efficiency of natural gas to flue gas waste heat; is the rated heating and cooling output; is a binary variable; The heat energy generated by the heat pump is expressed as: In the formula The heat energy generated by the heat pump; COP HP and are the coefficient of performance and rated output power of the heat pump respectively; The cooling energy generated by a centrifugal chiller is shown below: In the formula The cooling energy generated by the centrifugal chiller; COP HP , They are the coefficient of performance and rated output of the centrifugal chiller respectively; is the amount of electricity consumed by the centrifugal chiller; The power storage system model is as follows: In the formula, l∈{e,h,c} is the type of energy storage device, representing electric storage, thermal storage and ice storage respectively; is the energy storage level; α l is the energy loss, and For charging and discharging power and efficiency; and are the minimum and maximum charge states, respectively; is the rated energy storage capacity; P l max is the maximum charge and discharge power; and are the energy storage levels at the start time and end time respectively.

11. A robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and backup ancillary service markets considering uncertainty according to claim 9 or 10, characterized in that: The supply and demand balance and ancillary service market constraints in the operation model include: Supply and demand balance constraints for electricity, heat and cooling: In the formula, is the output power of PV; is the electricity purchased from the grid; n room is the number of rooms; Interaction power constraints between MEVPP and the grid: In the formula A cap on the amount of electricity that can be purchased from the grid; Peak shaving constraints: In the formula is the base power purchased from the grid; is the lower bound of bidding ability; and is a binary variable; and Indicates peak-shaving and valley-filling power; T p and T v are peak and trough periods respectively; and is the upper bound of bidding capacity; Alternative constraints: In the formula and They represent the upper reserve and lower reserve capacities at time T respectively; the total reserve capacity reported by the virtual power plant in the reserve market within k time slots; and Respectively represent the upper and lower limits of the power grid.

12. According to claim 9, a robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby auxiliary service markets taking into account uncertainty is characterized by: The bidding model building module is specifically used for: During the MEVPP day-ahead bidding phase, the uncertainty sets of photovoltaic power generation and the uncertainty sets of electric load, thermal load, and cooling load are: Among them, the uncertainty variable and are the actual values ​​of initial power load, chilled water volume, outdoor temperature and PV output power, respectively. To predict the nominal value, To predict the maximum deviation, the band and is a binary variable, using uncertainty budget parameter Γ i To adjust the conservativeness of the robust bidding model; Introducing uncertainty variables into the MEVPP day-ahead bidding model:

13. According to claim 9, a robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and standby auxiliary service markets taking into account uncertainty is characterized by: The optimization model building module is specifically used for: Rewrite the MEVPP two-stage robust optimization model into matrix form: Where x is the binary variable in the first stage, u and y are the uncertainty variable and continuous variable in the second stage, respectively, c is the coefficient column vector corresponding to the objective function, E, F, K, I u are the coefficient matrix and constant column vector of the corresponding constraints, λ, v and π are dual variables; the uncertainty set is: The C&CG algorithm is used to transform the MEVPP two-stage robust optimization model into a main problem and a sub-problem to solve the robust problem.

14. A robust bidding device for multi-energy virtual power plants to collaboratively participate in peak load regulation and backup ancillary service markets taking into account uncertainty according to claim 13, characterized in that: The Optimization Model Building Blocks are also used to: The duality theory is used to transform the inner minimum problem of the MEVPP two-stage robust optimization model rewritten in matrix form into a dual problem and then merged with the mid-extreme problem, which is expressed as: Assume v = 0, UB v →+∞,LB v →-∞, initialize x (0) ; Decompose formula A3 into a sub-problem A4 and a main problem A5: Update v = v + 1 and solve the subproblem; According to the solution of the sub-problem, we get u (v) , that is, we get u (k) (k=1,…,v), and update the upper bound: Security Service v =(d-Fx (v-1) ) T λ (v) +(u 0 ) T π (v) +(Δu max ) T (With + ) (v) -(Δu max ) T (With - ) (v) A6 Based on the obtained u (k) (k=1,…,v), solve the main problem; According to the solution of the main problem, we can get x (v) , update the lower bound: <h2 style=";text-align:left;direction:ltr">LB<h2 style=";text-align:left;direction:ltr"> v <h2 style=";text-align:left;direction:ltr"> =η<h2 style=";text-align:left;direction:ltr"> (v) <h2 style=";text-align:left;direction:ltr"> A7 If UB v -LB v ≤ε, then output x (v) ,y (v) and u (v) .

15. A robust bidding system for multi-energy virtual power plants to participate in peak load regulation and backup ancillary service markets considering uncertainty, characterized in that: include: A computer readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the robust bidding method for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service markets taking into account uncertainty as described in any one of claims 1-8.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a robust bidding method for multi-energy virtual power plants to collaboratively participate in peak-shaving and standby ancillary service markets taking into account uncertainties as described in any one of claims 1-8.

Images