Robust optimization scheduling method for multi-energy coupling micro-energy grid based on optimal risk interval

Abstract

The present invention discloses a robust optimization scheduling method for a multi-energy coupling micro-energy network based on an optimal risk interval. The method first models the energy flow relationship and equipment characteristics of the multi-energy coupling network, including power-to-gas equipment (P2G), electric refrigerators, fuel and gas units, wind power generation equipment, etc.; secondly, a value risk assessment strategy is proposed, which can obtain the optimal wind abandonment and load shedding confidence intervals through joint optimization with the scheduling strategy, solving the problem of the uncertainty set boundary that cannot be accurately estimated in the existing robust optimization problem. Finally, a two-stage robust optimization scheduling model is established with the optimal wind abandonment, load shedding risk interval and economy as the objective function. Solving the model with the CPLEX solver can obtain the optimal energy flow relationship, optimal equipment output and optimal risk interval of the multi-energy equipment in the system, and can be well applied in power systems with widespread penetration of uncertain new energy.

Description

Technical Field

[0001] The present invention belongs to the field of power system optimization operation, and in particular relates to a robust optimization scheduling method for a multi-energy coupling micro energy network based on an optimal risk interval. Background Art

[0002] The improvement of comprehensive energy efficiency is the primary concern in the construction of the energy and power industries. Decarbonization requires a major transformation of the energy system, including energy efficiency measures, a shift to low-carbon or zero-carbon energy carriers at the final consumption point (i.e., electricity, hydrogen or bioenergy), and clean energy supply. This provides a variety of options for electricity supply, including nuclear power, carbon capture and storage (CCS), and renewable energy (or a combination of the two). Wind and solar are globally abundant variable renewable energy (VRE), and their costs have fallen sharply in recent years, so that in many regions, they are the cheapest option for power generation. The widespread use of VRE and the electrification of large parts of the economy (such as transportation and heating) are considered to be technically and economically feasible decarbonization pathways. With the increasing penetration of VRE resources and widespread electrification, the planning and operation of future energy systems requires more integrated / hybrid / coupled energy systems, characterized by multi-energy flow systems. Multi-energy flow systems are defined as energy systems with strong coordination in the operation and planning of multiple energy carriers and / or sectors to provide reliable, cost-effective energy services with minimal impact on the environment.

[0003] Multi-energy flow systems are flexible in design and can improve system performance and efficiency. This is often achieved by implementing energy hubs where multiple energy carriers (such as electricity and gas) at the input port can be converted, conditioned and transformed to provide certain required energy services at the output port, such as electricity, heating, cooling and / or other forms of energy such as compressed air, hydrogen and fuels. For example, heating services can be provided by renewable electricity or, if renewable electricity is in short supply, previously stored heat can provide heating services. In addition, MCESs can also effectively store energy. For example, if there is excess VRE in the grid, MCESs can provide the opportunity to store this energy for later use by converting electricity to another form of energy, such as electricity to fuel. MCESs can also overcome energy network limitations. For example, if the grid is constrained, the gas grid can be used. This flexibility of MCES is at the core of its potential social benefits, namely cost-effective energy services with minimal environmental impact.

[0004] At present, in the field of multi-energy flow system optimization and dispatch, most of the focus is on economy, that is, wind curtailment and load shedding are directly embedded in the optimization model as optimization targets. This simple processing can improve the energy efficiency of the system and improve the economy of the system. However, among the three elements of the actual power system, safety is more concerned, that is, wind curtailment and load shedding will bring huge safety challenges to wind turbines and network resilience to some extent. Summary of the invention

[0005] The purpose of the present invention is to provide a robust optimization scheduling method for a multi-energy coupled micro energy grid based on an optimal risk interval in view of the deficiencies in the prior art.

[0006] The object of the present invention is to achieve the following technical solution: a robust optimization scheduling method for multi-energy coupling micro-energy grid based on optimal risk interval, comprising the following steps:

[0007] 1) Model the energy flow relationship and equipment characteristics of the multi-energy coupling network, including P2G equipment, electric refrigeration machines, oil and gas units, wind power generation equipment, etc. Consider the multi-energy flow coupling relationship between the three networks of electricity, gas and cooling, make full use of electric refrigeration and power-to-gas equipment for the interaction of the three networks, and combine it to deal with the uncertain power supply characteristics of wind turbines, so as to establish a mathematical model that can reflect multi-energy synergy.

[0008] 2) The present invention proposes a value-risk assessment strategy, which obtains the optimal wind curtailment and load shedding confidence intervals by jointly optimizing with the scheduling strategy, thereby solving the problem of the inability to accurately estimate the uncertainty set boundaries of parameters in existing robust optimization problems, and realizing the ability to adaptively obtain a scheduling plan with operating risk within the optimal confidence interval.

[0009] 3) Taking the optimality of wind curtailment, load shedding risk interval and economy as the objective function, a two-stage robust optimization dispatch model is established, in which various operating costs constitute the first-stage objective function, and the minimum wind curtailment and load shedding risk interval is the second-stage objective function. Solving the model with the CPLEX solver can obtain the optimal energy flow relationship, optimal equipment output and optimal risk interval of multi-energy equipment in the system, which can be well applied in power systems with uncertain and widespread penetration of new energy.

[0010] Furthermore, the specific contents of step 1) are:

[0011] 1.1) Oil and gas generator sets for power generation

[0012] Formula (1) characterizes the upper and lower bounds of the power generation of the oil and gas units, (2)-(3) characterize the adjustment range of the generator set under the dynamic ramp constraint, (4)-(6) are the signs and constraints of the start and stop and operation of the unit, that is, the unit cannot be started and shut down in a very short time. Formula (7) evaluates whether the ramp capacity of the unit is within the optimized reserve range.

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Where: γ is the combination of oil and gas units, T on,i , T off,i It is the factory key parameter related to the startup and shutdown of the unit. on,i , H off,i Set the maximum fault tolerance rate for the unit adjustment status.

[0021] 1.2) Power flow balance in multi-energy networks

[0022] Equations (8)-(11) describe the DC power flow relationship, that is, the relationship between node voltage, phase angle and tie line power. Equation (12) ensures that the power generation limit of the wind turbine does not exceed the available wind power capacity at the current moment, where Aw q is an uncertain parameter. Equation (13) describes the dynamic power balance of the entire network. Equations (14)-(16) show that the power flow exceeds the limit to a certain extent in extreme cases, but at the same time it will cause a great penalty on the operating cost.

[0023]

[0024] pf hk,t =(θ h,t -θ k,t ) / x hk (9)

[0025]

[0026] θ j,t =0,j∈REF (11)

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] Where: W h is the associated index of all wind turbines, Expresses the clean wind energy that wind turbine q can absorb at time t, It is the power supply required by users in each node building in the multi-energy flow network in real time. h 、PLF h It is the direction indicator of the interaction between each power tie line in the DC power flow, pf hk,t is the power load on the tie line and is a bidirectional variable. When it is positive, it means the power flows out of the node, and when it is negative, it means the power flows into the node. h,t ,θ k,t It is the phase angle state of the key contact points h and k in the relevant grid. REF identifies the state and contour of the power balance point in the multi-energy flow network. It is a load pool in the entire network that has real-time demand for electricity.

[0033] 1.3) Characteristic modeling of gas network and cooling network in multi-energy flow network

[0034] Formula (17) is a four-stage gas consumption equation for gas turbines, that is, four different parameters are used to express the start-up and shutdown of gas turbines and the demand for natural gas for power generation. Formulas (18)-(21) describe the dynamic balance of the natural gas network in a manner similar to the characteristics of the entire power flow network in a multi-energy flow network. In the present invention, the sgn(·) function is designed to further express the binary flow direction of natural gas. Furthermore, the pressure condition of the key nodes of natural gas in the entire network is judged by formula (22). Similar to the power flow, since the subsequent optimization objectives are divided into two stages, the solution of the second stage will affect the scheduling plan of the first stage. When the scenario screened out in the second stage is an extreme scenario, the original power balance constraint of the first stage may exceed the limit. Therefore, the present invention designs the over-limit working condition of the natural gas network as constraints (23)-(25).

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044] Where: α 1,i , α 2,i , α 3,i , α 4,i Corresponding to the key calculation parameters of the four-stage gas generator operating conditions mentioned above, GPE m , GPF m Similar to the direction of the tie line in the electric power network, it can accurately express the distribution of pipelines in the natural gas network. The meaning of gf and pf is similar. and The meaning of expression is similar. κ explains the relationship between the air pressure at both ends of the cold pipe and the cold air capacity of the pipe. m,t ,pr n,t It is the air pressure distribution state in the key nodes of the refrigeration network.

[0045] Furthermore, the specific contents of step 2) are:

[0046] In the field of power system optimization operation, the concept of conditional value-at-risk (CVAR) is proposed to reflect the possible impact of uncertain factors on system operation, such as wind curtailment and load shedding, both of which will cause extremely severe safety problems to the power system. In this invention, in order to ensure the safe and economic operation of the proposed multi-energy flow coupling system as much as possible, the concept of CVAR is introduced into the multi-energy flow coupling network. Its expression is:

[0047]

[0048]

[0049] in, They respectively express the risk of wind abandonment in the full multi-energy flow network and the risk of load loss in the full multi-energy flow coupled system.

[0050] In order to efficiently solve the optimal risk index CVAR, the idea of ​​integral linearization is applied in the present invention, and its specific process includes equations (28)-(33):

[0051]

[0052]

[0053]

[0054]

[0055]

[0056]

[0057] Among them, Δs is the optimal number of segments for the piecewise linearization of the optimal operating risk interval, and is a continuous auxiliary variable in a piecewise linear process.

[0058] Furthermore, the specific content of step 3) is:

[0059] In the first stage objective function, the lowest economic operating cost is considered, including the lowest operating energy consumption and resource consumption of fuel oil, gas, and refrigeration units, and the lowest natural gas production. The second stage objective function seeks the optimal operating risk interval under the worst wind power and load demand scenarios. The entire objective function structure is as follows (34):

[0060] min PC+GC+IL max min CVAR (34)

[0061]

[0062] Where: PC, GC, IL constitute the first stage cost content, including fuel oil, gas, refrigeration unit operating energy consumption and consumed resources, and CVAR provides the corresponding optimal operating risk interval. By solving the min-max-min three-level planning problem, the optimal energy flow relationship of multi-energy equipment in the system can be obtained while providing the optimal equipment output and the optimal risk interval.

[0063] The beneficial effects of the present invention are as follows: the present invention quantifies the impact of wind curtailment and load shedding on the power grid and proposes the concept of operating risk; the present invention can obtain the optimal energy flow relationship, optimal equipment output and optimal risk interval of multi-energy equipment in the system, while ensuring low-dimensional wind curtailment and load shedding, and provides the power grid with a safe operating status indicator, can be well applied in power systems with widespread penetration of uncertain new energy, and has great application potential for actual power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The overall structure diagram of the robust optimization scheduling method of multi-energy coupling micro-energy grid based on optimal risk interval;

[0065] Figure 2 Schematic diagram of the simulation state of the optimal operating risk range. DETAILED DESCRIPTION

[0066] The specific implementation modes of the present invention will be further described below in conjunction with the accompanying drawings.

[0067] like Figure 1 As shown, the present invention provides a robust optimization scheduling method for a multi-energy coupling micro-energy grid based on an optimal risk interval, comprising the following steps:

[0068] 1) Model the energy flow relationship and equipment characteristics of the multi-energy coupling network, including power-to-gas P2G equipment, electric refrigeration machines, oil and gas units, wind power generation equipment, etc. Consider the multi-energy flow coupling relationship between the three networks of electricity, gas and cooling, make full use of electric refrigeration and power-to-gas equipment for the interaction of the three networks, and combine them to deal with the uncertain power supply characteristics of wind turbines, so as to establish a mathematical model that can reflect multi-energy synergy. Specifically include:

[0069] 1.1) Oil and gas generator sets for power generation

[0070] Characterize the upper and lower limits of power generation for oil-fired and gas-fired units:

[0071]

[0072] Where γ is the set of oil and gas units, t represents the time; I i,t represents the startup and operation status of unit i at time t, represents the pre-dispatching output of unit i at time t; P i min represents the lower limit of the dispatch processing of unit i, P i max Represents the upper limit of the dispatch output of unit i.

[0073] Under dynamic climbing constraints, the adjustment range of the generator set is:

[0074]

[0075]

[0076] in, represents the upward ramp rate of unit i at time t, represents the downward climbing rate of unit i at time t; u i,t Indicates that at time t, unit i is in the startup command, v i,tIt means that at time t, unit i is in shutdown command.

[0077] The unit cannot be started or shut down in a very short time. The signs and restrictions for starting, stopping and running the unit are as follows:

[0078]

[0079]

[0080]

[0081] Among them, H on,i Indicates the maximum fault tolerance setting for the startup of unit i in adjustment state, H off,i Indicates the maximum fault tolerance setting for shutdown of unit i in adjustment state; NT indicates the fastest adjustment time parameter of the unit; T on,i Indicates the factory key parameters related to the startup of unit i, T off,i Indicates the factory key parameters related to the shutdown of unit i.

[0082] Evaluate whether the ramp capacity of the unit is within the optimized reserve range:

[0083]

[0084] Where P i,t Represents the actual dispatch output of unit i at time t.

[0085] 1.2) Power flow balance in multi-energy networks

[0086] The DC power flow relationship, that is, the relationship between node voltage, phase angle and tie line power:

[0087]

[0088] pf hk,t =(θ h,t -θ k,t ) / x hk (9)

[0089]

[0090] θ j,t =0,j∈REF (11)

[0091] Among them, h represents the node in the system network, γ h represents the set of units connected to node h, W h represents the set of wind farms connected to node h, PLE represents the clean wind energy that can be absorbed by wind turbine q at time t. hIndicates the outflow mark of the interaction between each power connection point h in the DC power flow, PLF h Indicates the inward flow mark between each power connection point h in the DC power flow; hk represents the connection line connecting the connection point h and another connection point k; the power load pf on the connection line hk at time t hk,t is a bidirectional variable, which means outflow from the node when positive and inflow from the node when negative; represents the real-time power supply demanded by users in the h-node building in the multi-energy flow network at time t; x hk represents the impedance parameter of hk, θ h,t represents the phase angle state of the key contact point h in the relevant grid at time t, θ k,t It represents the phase angle state of the key contact point k in the relevant grid at time t; It indicates the maximum power flow limit of the tie line hk; REF identifies the state and outline of the power balance point in the multi-energy flow network.

[0092] Ensure that the power generation limit of wind turbines does not exceed the available wind power capacity at the current moment,

[0093]

[0094] Among them, Aw q is the uncertain parameter of wind turbine generator q, ζ q,t Represents the wind power consumption parameter of wind turbine q at time t.

[0095] Dynamic power balance of the entire network:

[0096]

[0097] Among them, W h is the collection of all wind turbines, Represents the load pool with real-time demand for electric energy in the entire network.

[0098] In extreme cases, the power flow may exceed the limit to a certain extent, but at the same time it will cause a great penalty on the operating cost:

[0099]

[0100]

[0101]

[0102] In the formula, Indicates that the energy flow of node h at time t actually exceeds the limit; ΔFL e,max Indicates the maximum limit of energy flow exceeding the limit at time t.

[0103] 1.3) Characteristic modeling of gas network and cooling network in multi-energy flow network

[0104] Four different parameters are used to express the start and stop of the gas-fired unit and the demand for natural gas for power generation. The four-stage gas consumption equation of the gas-fired unit is as follows:

[0105]

[0106] Among them, VG i,t represents the gas production of gas generator set i at time t, α 1,i , α 2,i , α 3,i , α 4,i Corresponding to the key calculation parameters of the four-stage gas generator operating conditions, represents the power generation of gas generator set i at time t, and NGU represents the set of gas generator sets.

[0107] In the present invention, the sgn(·) step function is designed to further express the binary flow direction of natural gas. Similar to the characteristics of the entire power flow network in the multi-energy flow network, the dynamic balance of the natural gas network is as follows:

[0108]

[0109]

[0110]

[0111]

[0112] Where m represents a natural gas pipeline node, mn represents a natural gas pipeline connecting node m to another node n, IP represents the set of natural gas pipelines that can be connected; sp represents a natural gas well, GS m represents the set of natural gas wells connected to node m, SW sp,t Represents the gas production of the natural gas well sp at time t; similar to the direction of the tie line in the power network, GPE m represents the pipeline flowing into node m in the natural gas network, GPF m represents the pipeline flowing out of node m in the natural gas network; gf mn,t represents the gas flow rate of natural gas pipeline mn at time t, NGU m Represents the set of gas generators connected to node m; represents the demand of natural gas users connected to node m at time t; κ explains the relationship between the gas pressure at both ends of the cold pipe and the cold gas capacity of the pipe, pr m,t is the air pressure distribution state in the key node m in the refrigeration network at time t, pr n,t is the air pressure distribution state in the key node n in the refrigeration network at time t; represents the minimum cooling value at node m, Indicates the maximum cooling value at node m; Indicates the minimum flow limit of the condensing pipe mn, Indicates the maximum flow limit of the condensing pipe mn.

[0113] Furthermore, similar to the electric energy flow, the pressure conditions of the key nodes of the entire natural gas network are determined:

[0114]

[0115] in, represents the minimum value of natural gas pressure in natural gas well sp, Indicates the maximum value of natural gas pressure in natural gas well sp.

[0116] Since the subsequent optimization objectives are divided into two stages, the solution of the second stage will affect the scheduling plan of the first stage. When the scenario screened out in the second stage is an extreme scenario, the original power balance constraint of the first stage may exceed the limit. Therefore, the present invention designs the natural gas network over-limit working condition as the following constraint:

[0117]

[0118]

[0119]

[0120] In the formula, represents the actual over-limit of the natural gas flow at the natural gas pipeline node m at time t, represents the actual power supply of gas generator set i at time t, ΔFL g,max Indicates the maximum limit of natural gas flow exceeding the limit.

[0121] 2) The present invention proposes a value risk assessment strategy, which obtains the optimal wind curtailment and load shedding confidence interval by jointly optimizing with the dispatching strategy, solves the problem of being unable to accurately estimate the uncertainty set boundary of parameters in the existing robust optimization problem, and realizes the dispatching plan that can adaptively obtain the operation risk in the optimal confidence interval. Specifically including:

[0122] In the field of power system optimization operation, the concept of conditional value-at-risk (CVAR) is proposed to reflect the possible impact of uncertain factors on system operation, such as wind curtailment and load shedding, both of which will cause extremely severe safety problems to the power system. In the present invention, in order to ensure the safe and economical operation of the proposed multi-energy flow coupling system as much as possible, the concept of CVAR is introduced into the multi-energy flow coupling network. Figure 2 As shown in the figure, the optimal risk interval is divided into two shaded areas. The left shaded area is the risk of wind cutting, and the right shaded area is the risk of wind abandonment. The expression is:

[0123]

[0124]

[0125] Among them, w represents the wind farm or user building that needs to consider the risk value; is the wind curtailment risk of the wind farm or user building w in the full multi-energy flow network at time t, is the load loss risk of the wind farm or user building w in the full multi-energy flow coupling system at time t; It represents the wind farm abandonment value w that needs to consider the risk value at time t, represents the load shedding of building w that needs to consider the risk value at time t; ζ represents the day-ahead wind power forecast value, Aw represents the maximum predicted wind power available at the wind farm or user building w days ago. q,t represents the wind power that can be absorbed in the actual dispatch of wind turbine q at time t, and Pr(·) represents the probability density function obtained by wind power prediction.

[0126] In order to efficiently solve the optimal risk index CVAR, the present invention applies the idea of ​​integral linearization, and the specific process is as follows:

[0127]

[0128]

[0129]

[0130]

[0131]

[0132]

[0133] Where s represents the piecewise linear segment in the piecewise linear method; It represents the intermediate variable of wind farm w that needs to consider risk value in the piecewise linear process s at time t. It is used in the algorithm and has no actual physical meaning. It represents the intermediate variable of load shedding of user building w that needs to consider risk value in the piecewise linear process s at time t. It is used by the algorithm and has no actual physical meaning. Δs is the optimal number of segments for piecewise linearization of the optimal operation risk interval, and represents the continuous auxiliary variable of the inward cutting of the wind farm or user building w in the piecewise linear process s at time t, A continuous auxiliary variable representing the outward cutting of a wind farm or user building w in a piecewise linear process s at time t.

[0134] 3) Taking the optimality of wind curtailment, load shedding risk interval and economy as the objective function, a two-stage robust optimization dispatch model is established, in which various operating costs constitute the first-stage objective function, and the minimum wind curtailment and load shedding risk interval is the second-stage objective function. Solving the model with the CPLEX solver can obtain the optimal energy flow relationship, optimal equipment output and optimal risk interval of multi-energy equipment in the system, which can be well applied in power systems with uncertain and widespread penetration of new energy. Specifically including:

[0135] In the first stage objective function, the lowest economic operating cost is considered, including the lowest operating energy consumption and resource consumption of fuel oil, gas, and refrigeration units, and the lowest natural gas production. The second stage objective function seeks the optimal operating risk interval under the worst wind power and load demand scenarios. The entire objective function structure is as follows:

[0136] min PC+GC+IL max min CVAR (34)

[0137]

[0138]

[0139]

[0140]

[0141] In the formula, PC, GC, and IL constitute the contents of the first-stage cost, and CVAR provides the corresponding optimal risk interval for operation; PC is the operating energy consumption and resources consumed by the fuel unit, GC is the operating energy consumption and resources consumed by the gas unit, and IL is the operating energy consumption and resources consumed by the refrigeration unit. T represents the time of the entire scheduling cycle, and NCU represents the set of coal-fired units included in the objective function; represents the startup cost of coal-fired unit i included in the objective function, represents the shutdown cost of coal-fired unit i included in the objective function, represents the continuous operation cost of coal-fired unit i included in the objective function, represents the cost of electricity generation by coal-fired unit i included in the objective function, represents the ramping cost of coal-fired unit i included in the objective function; represents the pre-dispatching power generation cost of coal-fired unit i included in the objective function at time t; represents the upward rotation standby parameter of coal-fired unit i included in the objective function at time t, It represents the downward rotation standby parameter of coal-fired unit i included in the objective function at time t. ρ g represents the gas production cost of the natural gas well g included in the objective function, represents the startup cost of gas unit i included in the objective function, represents the shutdown cost of gas unit i included in the objective function, represents the operating cost of gas unit i included in the objective function, represents the cost of power generation consumption of gas unit i included in the objective function, represents the ramping cost of gas generator set i included in the objective function, represents the upward rotation standby parameter of gas unit i included in the objective function at time t, Represents the downward rotation reserve parameter of gas generator set i included in the objective function at time t. ρ a represents the power flow over-limit penalty factored into the objective function, W represents the actual over-limit value of power flow in power grid i included in the objective function at time t, H represents the set of over-limit tie lines in the power grid included in the objective function, and M represents the set of over-limit tie lines in the gas grid included in the objective function. se represents the set of abandoned wind farms included in the objective function, W re represents the set of load shedding user buildings included in the objective function, It represents the penalty cost of wind curtailment of wind farm or user building w included in the objective function.

[0142] By solving the min-max-min three-level planning problem of formula (34), the optimal energy flow relationship of multi-energy equipment in the system can be obtained while providing the optimal equipment output and the optimal risk range.

Claims

1. A robust optimization scheduling method for multi-energy coupling micro-energy grid based on optimal risk interval, characterized in that: The following steps are involved: Step 1: Model the energy flow relationship and equipment characteristics of the multi-energy coupling network, including power-to-gas equipment, electric refrigerators, oil and gas units, and wind power generation equipment; Considering the multi-energy flow coupling relationship among the three networks of electricity, gas and cooling, we make full use of electric refrigeration and power-to-gas equipment to interact with the three networks, and combine it to deal with the uncertain power supply characteristics of wind turbines, so as to establish a mathematical model that can reflect multi-energy synergy; Step 2: Design a value-risk assessment strategy, which obtains the optimal wind curtailment and load shedding confidence intervals through joint optimization with the dispatch strategy, and realizes the ability to adaptively obtain a dispatch plan with an operation risk in the optimal confidence interval; Step 3: Taking the optimality of wind curtailment, load shedding risk interval and economy as the objective function, a two-stage robust optimization scheduling model is established, in which various operating costs constitute the first-stage objective function, and the minimum wind curtailment and load shedding risk interval is the second-stage objective function; by solving the two-stage robust optimization scheduling model, the optimal energy flow relationship, optimal equipment output and optimal risk interval of multi-energy equipment in the system can be obtained; Among them, the entire wind abandonment, load shedding risk interval and economic optimization are the objective function, and the establishment and solution of the two-stage robust optimization scheduling model include: In the first stage objective function, the lowest economic operating cost is considered, including the lowest operating energy consumption and resource consumption of fuel oil, gas, and refrigeration units, and the lowest natural gas production; the second stage objective function seeks the optimal operating risk range under the worst wind power and load demand scenarios. The entire objective function structure is as follows (34): min PC+GC+IL max min CVAR(34) Where: PC, GC, IL constitute the content of the first stage cost, and CVAR provides the corresponding optimal risk interval; PC is the operating energy consumption and resources consumed by the fuel unit, GC is the operating energy consumption and resources consumed by the gas unit, IL is the operating energy consumption and resources consumed by the refrigeration unit; T represents the time of the entire scheduling cycle, and NCU represents the set of coal-fired units included in the objective function; represents the startup cost of coal-fired unit i included in the objective function, represents the shutdown cost of coal-fired unit i included in the objective function, represents the continuous operation cost of coal-fired unit i included in the objective function, represents the cost of electricity generation by coal-fired unit i included in the objective function, represents the ramping cost of coal-fired unit i included in the objective function; represents the pre-dispatching power generation cost of coal-fired unit i included in the objective function at time t; represents the upward rotation standby parameter of coal-fired unit i included in the objective function at time t, represents the downward rotation standby parameter of coal-fired unit i included in the objective function at time t; ρ g represents the gas production cost of the natural gas well g included in the objective function, represents the startup cost of gas unit i included in the objective function, represents the shutdown cost of gas unit i included in the objective function, represents the operating cost of gas unit i included in the objective function, represents the cost of power generation consumption of gas unit i included in the objective function, represents the ramping cost of gas generator set i included in the objective function, represents the upward rotation standby parameter of gas unit i included in the objective function at time t, represents the downward rotation standby parameter of gas generator set i included in the objective function at time t; ρ a represents the power flow over-limit penalty factored into the objective function, represents the actual over-limit value of power flow in power grid i included in the objective function at time t, H represents the set of over-limit tie lines in the power grid included in the objective function, and M represents the set of over-limit tie lines in the gas grid included in the objective function; W se represents the set of abandoned wind farms included in the objective function, W re represents the set of load shedding user buildings included in the objective function, represents the penalty cost of wind curtailment of wind farm or user building w included in the objective function; By solving the min-max-min three-level planning problem of formula (34), the optimal energy flow relationship of multi-energy equipment in the system is obtained while providing the optimal equipment output and the optimal risk range.

2. The robust optimization scheduling method for multi-energy coupling micro-energy grid based on optimal risk interval as claimed in claim 1 is characterized in that: In step 1, the energy flow relationship and device characteristics of the multi-energy coupling network are modeled, including: 2.1) Oil and gas generator sets for power generation Formula (1) characterizes the upper and lower bounds of the power generation of the oil and gas units. Formulas (2) to (3) characterize the adjustment range of the generator set under the dynamic ramp constraint. Formulas (4) to (6) are the signs and constraints of the start and stop and operation of the unit, that is, the unit cannot be started and shut down in a very short time. Formula (7) evaluates whether the ramp capacity of the unit is within the optimized reserve range. Where: γ is the set of oil and gas units, t represents the time; I i,t represents the startup and operation status of unit i at time t, represents the pre-dispatching output of unit i at time t; P i min represents the lower limit of the dispatch processing of unit i, P i max represents the upper limit of dispatch output of unit i; represents the upward ramp rate of unit i at time t, represents the downward climbing rate of unit i at time t; u i,t Indicates that at time t, unit i is in the startup command, v i,t Indicates that unit i is in shutdown command at time t; H on,i Indicates the maximum fault tolerance setting for the startup of unit i in adjustment state, H off,i Indicates the maximum fault tolerance setting for shutdown of unit i in adjustment state; NT indicates the fastest adjustment time parameter of the unit; T on,i Indicates the factory key parameters related to the startup of unit i, T off,i Indicates the factory key parameters related to the shutdown of unit i; P i,t represents the actual dispatch output of unit i at time t; 2.2) Power flow balance in multi-energy networks Equations (8) to (11) describe the DC power flow relationship, that is, the relationship between node voltage, phase angle and tie line power. Equation (12) ensures that the power generation limit of the wind turbine does not exceed the available wind power capacity at the current moment, where Aw q is an uncertain parameter. Equation (13) describes the dynamic power balance of the entire network. Equations (14)-(16) show that the power flow exceeds the limit to a certain extent in extreme cases, but at the same time, it will cause a penalty on the operating cost. pf hk,t =(θ h,t -θ k,t ) / x hk (9) i j,t =0,j∈REF (11) Where: h represents the node in the system network, γ h represents the set of units connected to node h, W h represents the set of wind farms connected to node h, PLE represents the clean wind energy that can be absorbed by wind turbine q at time t. h Indicates the outflow mark of the interaction between each power connection point h in the DC power flow, PLF h Indicates the inward flow mark between each power connection point h in the DC power flow; hk represents the connection line connecting the connection point h and another connection point k; the power load pf on the connection line hk at time t hk,t is a bidirectional variable, which means outflow from the node when positive and inflow from the node when negative; represents the real-time power supply demanded by users in the h-node building in the multi-energy flow network at time t; x hk represents the impedance parameter of hk, θ h,t represents the phase angle state of the key contact point h in the relevant grid at time t, θ k,t It represents the phase angle state of the key contact point k in the relevant grid at time t; represents the maximum power flow limit of the tie line hk; REF identifies the state and outline of the power balance point in the multi-energy flow network; Aw q is the uncertain parameter of wind turbine generator q, ζ q,t represents the wind power consumption parameter of wind turbine q at time t; Represents the load pool with real-time demand for electric energy in the entire network; Indicates that the energy flow of node h at time t actually exceeds the limit; ΔFL e,max It indicates the maximum limit of energy flow exceeding the limit at time t; 2.3) Characteristic modeling of gas network and cooling network in multi-energy flow network Formula (17) is a four-stage gas consumption equation for gas turbines, which uses four different parameters to express the start-up and shutdown of gas turbines and the demand for natural gas for power generation. Formulas (18)-(21) describe the dynamic balance of the natural gas network in a similar way to the characteristics of the entire power flow network in a multi-energy flow network. The sgn(·) function is designed to further express the binary flow direction of natural gas. Formula (22) is used to determine the pressure status of the key nodes of natural gas in the entire network. Since the subsequent optimization objectives are divided into two stages, the solution of the second stage will affect the scheduling plan of the first stage. When the scenario selected in the second stage is an extreme scenario, the original power balance constraint of the first stage may exceed the limit. Therefore, the over-limit working condition of the natural gas network is designed as the constraint of formulas (23)-(25); Where: VG i,t represents the gas production of gas generator set i at time t, α 1,i , α 2,i , α 3,i , α 4,i Corresponding to the key calculation parameters of the four-stage gas generator operating conditions, represents the power generated by gas unit i at time t, NGU represents the set of gas units; m represents the natural gas pipeline node, mn represents the natural gas pipeline connecting node m to another node n, IP represents the set of natural gas pipelines that can be connected; sp represents the natural gas well, GS m represents the set of natural gas wells connected to node m, SW sp,t Represents the gas production of the natural gas well sp at time t; similar to the direction of the tie line in the power network, GPE m represents the pipeline flowing into node m in the natural gas network, GPF m represents the pipeline flowing out of node m in the natural gas network; gf mn,t represents the gas flow rate of natural gas pipeline mn at time t, NGU m Represents the set of gas generators connected to node m; represents the demand of natural gas users connected to node m at time t; κ explains the relationship between the gas pressure at both ends of the cold pipe and the cold gas capacity of the pipe, pr m,t is the air pressure distribution state in the key node m in the refrigeration network at time t, pr n,t is the air pressure distribution state in the key node n in the refrigeration network at time t; represents the minimum cooling value at node m, Indicates the maximum cooling value at node m; Indicates the minimum flow limit of the condensing pipe mn, Indicates the maximum flow limit of the condensing pipe mn; represents the minimum value of natural gas pressure in natural gas well sp, Indicates the maximum value of natural gas pressure in natural gas well sp; represents the actual over-limit of the natural gas flow at the natural gas pipeline node m at time t, represents the actual power supply of gas generator set i at time t, ΔFL g,max Indicates the maximum limit of natural gas flow exceeding the limit.

3. The robust optimization scheduling method for multi-energy coupling micro-energy grid based on optimal risk interval as claimed in claim 1 is characterized in that: In step 2, the optimal value risk range is proposed, including: In a multi-energy flow coupled network, the expression of conditional risk value CVAR is: Among them, w represents the wind farm or user building that needs to consider the risk value; is the wind curtailment risk of the wind farm or user building w in the full multi-energy flow network at time t, is the load loss risk of the wind farm or user building w in the full multi-energy flow coupling system at time t; It represents the wind farm abandonment value w that needs to consider the risk value at time t, represents the load shedding of building w that needs to consider the risk value at time t; ζ represents the day-ahead wind power forecast value, Aw represents the maximum predicted wind power available at the wind farm or user building w days ago. q,t represents the wind power that can be absorbed in the actual dispatch of wind turbine q at time t, and Pr(·) represents the probability density function obtained by wind power prediction; In order to solve the optimal risk index CVAR, the idea of ​​integral linearization is applied, and the specific process includes equations (28)-(33): Where s represents the piecewise linear segment in the piecewise linear method; It represents the intermediate variable of wind curtailment of wind farm w that needs to consider risk value in the piecewise linear process s at time t; represents the intermediate variable of load shedding of user building w that needs to consider risk value in the piecewise linear process s at time t; Δs is the optimal number of segments for piecewise linearization of the optimal operation risk interval, and represents the continuous auxiliary variable of the inward cutting of the wind farm or user building w in the piecewise linear process s at time t, A continuous auxiliary variable representing the outward cutting of a wind farm or user building w in a piecewise linear process s at time t.

Images