CN113256045B - Park comprehensive energy system day-ahead economic dispatching method considering wind and light uncertainty - Google Patents

Abstract

The invention discloses a park comprehensive energy system day-ahead economic dispatching method considering wind and light uncertainty. Because the CCG method has better calculation efficiency in processing the robust optimization model, the method adopts the CCG method to solve. And finally, carrying out example simulation by using a commercial solver Gurobi to obtain a two-stage robust optimization scheduling strategy considering the combined thermoelectric demand response, verifying that the method can better treat the uncertainty in the system, and simultaneously can reduce the operation cost of the park and promote the consumption of new energy.

Description

Day-ahead economic dispatch method for integrated energy system in a park considering uncertainty of wind and solar power

Technical Field

The present invention belongs to the technical field of optimized operation of integrated energy systems, and in particular relates to a day-ahead economic dispatching method for an integrated energy system in a park taking into account the uncertainty of wind and solar power.

Background Art

The variance between actual wind power output and predicted output is much larger than the traditional load variance, and the traditional power system operation mode is not enough to ensure the reliability of the system. Due to the physical limitations of the power system (such as the ramp limit of conventional generators and transmission line capacity), wind power curtailment occurs frequently, resulting in low wind power utilization, which inhibits the enthusiasm for wind power investment in the long run. When large-scale renewable energy is connected to the power system, the strong randomness and intermittency of renewable energy make it difficult to accurately predict its output. In the planning and scheduling of the power system, the error of the predicted value of renewable energy output will have a non-negligible impact. In other words, the random fluctuations of renewable energy generation reduce the supply flexibility of the energy system. Therefore, the traditional deterministic optimization scheduling method cannot meet the requirements, and the uncertainty of renewable energy must be considered on the basis of the original deterministic optimization scheduling method.

Stochastic optimization and robust optimization are typical methods for solving optimization problems with uncertainty. Compared with stochastic optimization, robust optimization has the advantages of simple data acquisition, fast solution speed, and is suitable for solving large-scale uncertainty problems. It has been widely used in dealing with uncertainty problems. At present, many scholars have studied the application of robust optimization in integrated energy systems, but the existing literature does not fully consider the energy coupling relationship of various equipment in the system and the impact of load-side demand-side response on renewable energy uncertainty when using robust optimization to deal with wind and solar uncertainties in integrated energy systems.

A more comprehensive integrated energy system model can adaptively adjust the unit output to adapt to changes in renewable energy generation through the energy conversion relationship between various devices, thereby ensuring the safety of system operation. The combined thermal power demand response can make full use of the coupling relationship between various devices, further improve the system's ability to cope with the uncertainty of renewable energy, reduce wind and solar power abandonment, and increase the penetration of renewable energy. Therefore, based on existing research, further considering the two-stage adjustable robust optimization model of the park integrated energy system with combined thermal power demand response is of great significance to the optimal operation of the integrated energy system.

Summary of the invention

The technical problem to be solved by the present invention is to propose a day-ahead economic dispatch method for a comprehensive energy system in a park that takes into account the uncertainty of wind and solar power, establish a two-stage adjustable robust optimization model that takes into account the response of combined thermal power demand to deal with the uncertainty problem caused by wind and solar power output in the park, and convert the double-layer max-min problem in the model into a single-layer max problem through the duality theory, and use the CCG (column and constraint generation) method to solve it, which improves the solution speed and makes the dispatch result more in line with reality, further promoting the consumption of new energy.

To solve the above technical problems, the technical solution adopted by the present invention is: in order to consider the uncertainty of renewable energy, for the day-ahead economic dispatch problem of the park integrated energy system, a day-ahead economic dispatch method of the park integrated energy system considering the uncertainty of wind and solar is proposed, and a two-stage adjustable robust model considering the joint thermal power demand response is established. The day-ahead economic dispatch of the park integrated energy system is aimed at the basic scenario with wind power and photovoltaic output values as predicted values. When uncertainty occurs during operation, it can adaptively and safely redistribute the energy exchange between the generator set, the heating unit, the P2G equipment, the energy storage equipment and the park and the upper network. This optimization scheduling decision is completely in line with the idea of the two-stage adjustable robust model. In other words, the first stage determines the start and stop states of the gas turbine and the cogeneration unit under the basic scenario, and the second stage searches for the worst scenario that causes system insecurity when uncertainty occurs. The two-stage adjustable robust dispatch model proposed by the present invention assumes that the start and stop state of the unit is a first-stage variable, that is, when the system uncertainty variable changes within its fluctuation range, the start and stop state of the unit will remain unchanged. This is because the physical characteristics of most generator sets limit them from being able to quickly change the start and stop state of the unit under uncertain conditions.

A day-ahead economic dispatch method for a park integrated energy system considering the uncertainty of wind and solar power, comprising the following steps:

(1) Determine the specific composition of the multi-energy park, including the new energy forms introduced and the specific equipment composition;

(2) Establish models of various energy conversion equipment in the park;

(3) Establish a demand response model;

(4) Under the premise of meeting the system safety constraints, taking the minimum operating cost of the basic scenario as the objective function, a two-stage adjustable robust model considering the joint thermal power demand response is established;

(5) Obtain the abstract expression of the two-stage adjustable robust optimization model for the day-ahead economic dispatch of the park's integrated energy system;

(6) Establish the maximal and minimal sub-problems of worst-case scenario identification;

(7) Using CCG method to solve the two-stage adjustable robust model considering the combined heat and power demand response;

(8) The energy access, new energy output data, equipment parameters, and operating parameters of the park's integrated energy system are input, and the commercial solver Gurobi is used to solve the two-stage robust optimization model of the day-ahead economic dispatch of the park's integrated energy system considering the uncertainty of wind and solar power to obtain its dispatch strategy.

Furthermore, the specific composition of the park comprehensive energy system in step (1) is as follows:

(1) The new energy forms connected to the park's integrated energy system are: wind power and photovoltaic power generation;

(2) The energy conversion equipment introduced into the park’s comprehensive energy system includes: power-to-gas equipment, electric boilers, gas turbines, cogeneration units, and gas/heat storage equipment.

The models of the energy conversion equipment in the park described in step (2) are as follows;

(2.1) Power-to-gas equipment model

分别为电转气(P2G)设备制气功率、耗电功率和电转气效率，L^HANG为天然气低热值，取9.7kWh/m³；

为第m台P2G的最小、最大气功率。Where: t is the scheduling time; m is the power-to-gas equipment index;

They are respectively the gas production power, power consumption and power-to-gas efficiency of the power-to-gas (P2G) equipment, L ^HANG is the lower calorific value of natural gas, which is 9.7 kWh/m ³ ;

is the minimum and maximum gas power of the mth P2G.

(2.2) Electric boiler model

和

分别为第n台电锅炉在t时段的耗电功率和产热功率；

为第n台电锅炉的电热转换效率，

分为第n台电锅炉最小、最大制热功率；

为第n台电锅炉在t时段的启停状态(1代表开机，0代表停机)。Where: t is the dispatch time; n is the electric boiler index;

and

are the power consumption and heat generation of the nth electric boiler in period t respectively;

is the electric-to-heat conversion efficiency of the nth electric boiler,

It is divided into the minimum and maximum heating power of the nth electric boiler;

It is the start and stop status of the nth electric boiler in period t (1 represents start, 0 represents stop).

(2.3) Gas turbine model

和

分别表示燃气轮机的发电功率和耗气功率；F(·)表示燃气轮机能耗曲线；

和

分别表示燃气轮机的开机和关机所需天然气消耗量；a_q、b_q和c_q表示F(·)的燃气系数；L^HANG为天然气低热值，取9.7kWh/m³；

分为第q台燃气轮机最小、最大发电功率。

表示第q台燃气轮机在t时段的发电功率；

为第q台燃气轮机在t时段的启停状态(1代表开机，0代表停机)，

为第q台燃气轮机在t时段的启停状态；

为第q台燃气轮机的上爬坡率、下爬坡率，

为第q台燃气轮机在t-1时段内连续开机、停机时间，

为第q台燃气轮机在时段t内的最小开机、停机时间。Where: t is the scheduling time; q is the gas turbine index;

and

They represent the power generation and gas consumption of the gas turbine respectively; F(·) represents the energy consumption curve of the gas turbine;

and

They represent the natural gas consumption required for starting and shutting down the gas turbine respectively; a _q , b _q and c _q represent the gas coefficient of F(·); L ^HANG is the lower calorific value of natural gas, which is 9.7 kWh/m ³ ;

Divided into the minimum and maximum power generation capacity of the qth gas turbine.

represents the power generation of the qth gas turbine in period t;

is the start/stop status of the qth gas turbine in period t (1 represents start, 0 represents stop),

is the start and stop status of the qth gas turbine in period t;

is the ramp-up rate and ramp-down rate of the qth gas turbine,

is the continuous start-up and shutdown time of the qth gas turbine in the t-1 period,

is the minimum startup and shutdown time of the qth gas turbine in time period t.

(2.4) Combined heat and power unit model

和

分别表示热电联产机组(CHP)的产热功率和耗气功率；

为微燃机t时段的发电功率、发电效率，

为散热损失率，

和

分别为溴冷机的制热系数和烟气回收率；L^HANG为天然气低热值，取9.7kWh/m³；

分为第p台CHP机组最小、最大发电功率；

表示第p台热电联产机组在t时段的发电功率；

为第p台CHP机组在t时段的启停状态(1代表开机，0代表停机)，

为第P台热电联产机组在t时段的启停状态；

为第p台CHP机组的上爬坡率、下爬坡率；

为第p台CHP机组在t-1时段内连续开机、停机时间，

为第p台CHP机组在时段t内的最小开机、停机时间。Where: t is the dispatch time; p is the index of the cogeneration unit;

and

They represent the heat production power and gas consumption power of the combined heat and power (CHP) unit respectively;

is the power generation and efficiency of the micro-turbine during period t,

is the heat loss rate,

and

are the heating coefficient and flue gas recovery rate of the bromide refrigerator respectively; L ^HANG is the lower calorific value of natural gas, which is 9.7kWh/m ³ ;

It is divided into the minimum and maximum power generation of the pth CHP unit;

represents the power generation of the pth cogeneration unit in period t;

is the start/stop status of the pth CHP unit in period t (1 represents start, 0 represents stop),

is the start and stop status of the Pth cogeneration unit in period t;

is the ramp-up rate and ramp-down rate of the pth CHP unit;

is the continuous start-up and shutdown time of the pth CHP unit in the t-1 period,

is the minimum startup and shutdown time of the pth CHP unit in time period t.

(2.5) Energy storage device model

ESS(t)＝(1-σ ^S )·ESS(t-1)+ES ⁱⁿ (t)·η ⁱⁿ -ES ^out (t)/η ^out

为t时段储气、放气功率，GS^{in ,max}、GS^out,max分别为储气设备的最大储气、放气功率；C_t ^GS为t时段储气设备的储气量，

为t-1时段储气设备的储气量；C^GS,min、C^GS,max分别为储气设备的最小、最大储气容量，η^CGS、η^GS,in、η^GS,out为储气设备自耗率、储气效率、放气效率。

为t时段储热、放热功率，HS^in,max、HS^out,max分别为储热设备的最大储热、放热功率；

为t时段储热设备的储气量，

为t-1时段储热设备的储气量；C^HS,min、C^HS,max分别为储热设备的最小、最大储热容量，η^CHS、η^HS,in、η^HS,out为储热设备自耗率、储热效率、放热效率。Where: t is the scheduling time; ESS(t), ES ⁱⁿ (t), ES ^out (t) are the storage energy, storage energy power and released energy power of the energy storage device in period t respectively; ESS(t-1) represents the storage energy of the energy storage device in period t-1; σ ^S is the self-consumption rate of the energy storage system; η ⁱⁿ and η ^out are the energy storage and release efficiencies of the energy storage device respectively;

is the gas storage and release power in period t, GS ^{in ,max} and GS ^out,max are the maximum gas storage and release power of the gas storage equipment respectively; C _t ^GS is the gas storage capacity of the gas storage equipment in period t,

is the gas storage capacity of the gas storage equipment in period t-1; C ^GS,min and C ^GS,max are the minimum and maximum gas storage capacities of the gas storage equipment, respectively; η ^CGS ,η ^GS,in and η ^GS,out are the self-consumption rate, gas storage efficiency and gas release efficiency of the gas storage equipment.

is the heat storage and heat release power in period t, HS ^in,max and HS ^out,max are the maximum heat storage and heat release power of the heat storage equipment respectively;

is the gas storage capacity of the heat storage equipment during period t,

is the gas storage capacity of the heat storage equipment in period t-1; C ^HS,min and C ^HS,max are the minimum and maximum heat storage capacities of the heat storage equipment, respectively; η ^CHS , η ^HS,in and η ^HS,out are the self-consumption rate, heat storage efficiency and heat release efficiency of the heat storage equipment.

The demand response model in step (3) is as follows:

P_t ^DR、P_t ^DR，inte、P_t ^DR，shif、P_t ^LD，max为t时段电负荷预测值、需求侧响应电负荷、需求侧转移电负荷、系统允许的最大电负荷，P^inte,max为调度时间内最大可中断电负荷。

为t时段最大可中断和可转出的电负荷比例。

为正代表转出可平移负荷，反之，代表转入可平移负荷。Where: t is the scheduling time;

P _t ^DR , P _t ^{DR, inte} , P _t ^{DR, shift} , and P _t ^{LD, max} are the predicted electric load in period t, the response electric load on the demand side, the transfer electric load on the demand side, and the maximum electric load allowed by the system. P ^{inte, max} is the maximum interruptible electric load within the dispatching time.

It is the maximum ratio of the electric load that can be interrupted and transferred out during the period t.

A positive value indicates a transfer out of the translatable load, whereas a negative value indicates a transfer in of the translatable load.

和

分别为时间段t热负荷预测值、需求侧响应热负荷、需求侧可响应热负荷比例、系统允许的最大热负荷和考虑需求响应后的热负荷；H^DR,max为调度时间内最大可中断热负荷；N_t为整个调度时间段。Where: t is the scheduling time;

and

are respectively the heat load forecast value in time period t, the demand side response heat load, the demand side responsive heat load ratio, the maximum heat load allowed by the system and the heat load after considering demand response; H ^DR,max is the maximum interruptible heat load within the scheduling time; N _t is the entire scheduling time period.

The two-stage adjustable robust model considering the combined heat and power demand response in step (4) is specifically as follows:

(4.1) Objective function

分别为购电、气费用；

分别为第p台CHP的开机、关机费用；

为弃风惩罚费用，

为弃光惩罚费用；

为售电收益，C^E,cc为需求侧响应中断电负荷的补偿成本；

为购电功率，

为购气功率，

分别为第i台风机、第j组光伏电池在t时段的弃风、弃光功率；

为售电功率，

为需求侧中断电负荷。Δt为调度时间间隔，N_t为调度时段数。

分别为单位购电、购气、售电价格，

分别为单位弃风、弃光的惩罚价格，

为单位需求侧响应中断电负荷的补偿价格。

为第p台CHP在t时段的启停状态(1代表开机，0代表停机)，

分别为第p台CHP的开机、停机一次的费用。N^WT、N^PV、N^CHP分别为风机、光伏电池、电锅炉、CHP的数量。Where: t is the scheduling time;

They are the cost of purchasing electricity and gas respectively;

are the startup and shutdown costs of the pth CHP, respectively;

Penalty fees for wind curtailment,

Penalty fee for abandonment;

is the revenue from electricity sales, C ^E,cc is the compensation cost for the load interruption in response to power outage on the demand side;

For the purchased power,

is the gas purchasing power,

are the abandoned wind and solar power of the i-th wind turbine and the j-th photovoltaic cell group in period t, respectively;

For selling electricity,

is the power outage load on the demand side. Δt is the dispatching time interval, and _Nt is the number of dispatching periods.

They are the unit's electricity purchase, gas purchase, and electricity sales prices,

are the penalty prices for abandoning wind and solar power,

It is the compensation price for unit demand side response to power outage load.

is the start/stop status of the pth CHP in period t (1 represents start, 0 represents stop),

are the startup and shutdown costs of the pth CHP, respectively. N ^WT , N ^PV , and N ^CHP are the numbers of wind turbines, photovoltaic cells, electric boilers, and CHPs, respectively.

(4.2) Constraints

v_1t,v_2t,v_3t,v_4t≥0

v _1t ,v _2t ,v _3t ,v _4t ≥ 0

分别为风电、光伏出力与预测值的偏差，

为风电、光伏出力偏差值与预测值的比例；

为不确定合集中的0-1变量；Δ_i、Δ_j分别为风电、光伏出力不确定性预算值；

为第p台CHP机组上爬坡和下爬坡纠错能力；

为第q台燃气轮机机组上爬坡和下爬坡纠错能力；λ(·)为约束条件对应的对偶变量。Where: t is the dispatch time; (·) ^u is the adjusted variable corresponding to the real-time change of wind and solar power output; v _1,t and v _2,t are the slack variables of power balance constraint; v _3,t and v _4,t are the slack variables of heat balance constraint; Ω ^WT and Ω ^PV are the uncertain sets of wind power and photovoltaic output respectively;

are the deviations of wind power and photovoltaic output from the predicted values,

is the ratio of wind power and photovoltaic output deviation to the predicted value;

is a 0-1 variable in the uncertainty set; Δ _i and Δ _j are the uncertainty budget values of wind power and photovoltaic power output respectively;

The up-ramp and down-ramp error correction capability of the pth CHP unit;

is the up-ramp and down-ramp error correction capability of the qth gas turbine unit; λ(·) is the dual variable corresponding to the constraint condition.

The abstract expression of the two-stage adjustable robust optimization model for the day-ahead economic dispatch of the park integrated energy system in step (5) is as follows:

Where: x represents the start and stop status of the units related to CHP and gas turbine, y and z represent the basic scenario and the dispatching output of the remaining units in the system adjusted according to the wind and solar output conversion, respectively, u is the uncertain variable related to the uncertainty of wind power and photovoltaic output, c _b , c _g , A, B, b, f, h, C, D, E, F, and G can be obtained through the objective function and constraints in 5.

The maximum and minimum sub-problems of the worst scenario identification in step (6) are as follows:

The CCG method is used in step (7) to solve the two-stage adjustable robust model considering the combined heat and power demand response as follows:

(7.1) Main problem

Ax+By≤b

(7.2) Sub-problem

(7.3) CCG solution steps

Step 1: Set the iteration counter s=0, and set the maximum value ε ^RO of the safety violation allowed by the system.

Step 2: Solve the main problem. If there is a solution, obtain the system unit start and stop status x and unit output arrangement y, and proceed to step 3; otherwise, stop the iteration and output no solution.

Step 3: Based on the x and y obtained in step 2, solve the maximum and minimum sub-problems to find the wind and photovoltaic output values under the worst scenario that leads to the maximum possible violation of safety regulations.

向主问题中增加以下所示的CCG约束，返回步骤2。Step 4: If the maximum possible violation of safety regulations solved in step 3 is less than ε ^RO , then x and y are the final optimization solutions and the iteration stops; otherwise, let s = s + 1, and according to the wind power and photovoltaic output values under the worst scenario solved in step 3

Add the following CCG constraints to the main problem and return to step 2.

f ^T v _s ≤ ε ^RO

The park comprehensive energy system data described in step (8) also includes the specific composition of the park comprehensive energy system, energy prices, equipment parameters and values of each energy conversion equipment, demand response ratio, new energy output fluctuations under basic scenarios and worst-case scenarios, maximum violation of safety regulations, and predicted values of new energy output and load.

Compared with the prior art, the present invention has the following beneficial effects:

(1) Robust optimization is used to solve optimization problems with uncertainty, and a two-stage adjustable robust optimization model is established, so that only simple data can be obtained in the process of scheduling the park's integrated energy system to obtain results close to reality, and the solution speed is fast. The improved two-stage adjustable robust optimization model fully considers the energy coupling relationship of various equipment in the system and the impact of the load-side demand-side response on the uncertainty of renewable energy while considering the uncertainty of wind and solar power in the integrated energy system. The day-ahead economic scheduling of the park's integrated energy system is based on the basic scenario with wind power and photovoltaic output values as predicted values. When uncertainty occurs during operation, it can adaptively and safely reallocate the energy exchange between generators, heating units, P2G equipment, energy storage equipment, and the park and the upper-level network. The two-stage robust optimization ensures that the system can meet safety constraints under any circumstances, minimizes the operating cost of the basic scenario, and meets the system safety and economy requirements.

(2) After considering the combined heat and power demand response, the role of gas energy storage, thermal energy storage and electric heat load demand response can be fully utilized. The combined heat and power demand response can make full use of the coupling relationship between various devices, further improve the system's ability to cope with the uncertainty of renewable energy, reduce wind and solar abandonment, and increase the penetration of renewable energy. A more comprehensive integrated energy system model can adaptively adjust the unit output to adapt to changes in renewable energy generation through the energy conversion relationship between various devices, ensure the safety of system operation, promote the consumption of renewable energy, improve the robustness of the system, and improve the economic benefits of the park.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of the steps of the method of the present invention;

Figure 2 is a detailed diagram of the integrated energy system of the park;

FIG3 is a flow chart of the CCG method solution;

Figure 4 shows the predicted values of wind turbine, photovoltaic output, electrical load and thermal load of the park's multi-energy integrated system on a typical winter day.

DETAILED DESCRIPTION

In order to fully illustrate the technical solution disclosed by the present invention, the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

The present invention discloses a method for day-ahead economic dispatch of a park integrated energy system taking into account the uncertainty of wind and solar power. The specific implementation steps are shown in FIG1 . The technical solution of the present invention includes the following steps:

Step 1: Determine the specific composition of the park's integrated energy system, including the new energy forms introduced and the specific equipment composition.

(1.1) The new energy forms connected to the park's integrated energy system are: wind power and photovoltaic power generation;

(1.2) The energy conversion equipment introduced into the park’s comprehensive energy system includes: power-to-gas equipment, electric boilers, gas turbines, cogeneration units, and gas/heat storage equipment.

Step 2: Establish models of various energy conversion equipment in the park.

(2.1) Power-to-gas equipment model

P2G technology can realize the conversion of electric energy into natural gas. Natural gas can be transmitted in large capacity, for a long time and over long distances through natural gas pipelines, providing strong technical support for the consumption of renewable energy, and can realize the large-scale and long-distance space-time transfer of wind power. At the same time, P2G responds quickly and has strong application prospects. The relationship between the gas production power and power consumption of P2G equipment and the gas production power limit are as follows:

分别为电转气(P2G)设备制气功率、耗电功率和电转气效率，L^HANG为天然气低热值，取9.7kWh/m³；

为第m台P2G的最小、最大气功率。Where: t is the scheduling time; m is the power-to-gas equipment index;

They are respectively the gas production power, power consumption and power-to-gas efficiency of the power-to-gas (P2G) equipment, L ^HANG is the lower calorific value of natural gas, which is 9.7 kWh/m ³ ;

is the minimum and maximum gas power of the mth P2G.

(2.2) Electric boiler model

The introduction of electric boilers can break the rigid constraints of the electric-heat coupling of CHP units and change the traditional "heat-to-electricity" scheduling method. Electric boilers can coordinate the peaks and valleys of electric-heat loads. The relationship between heating capacity and power consumption and the heating capacity constraints are as follows:

和

分别为第n台电锅炉在t时段的耗电功率和产热功率；

为第n台电锅炉的电热转换效率，

分为第n台电锅炉最小、最大制热功率；

为第n台电锅炉在t时段的启停状态(1代表开机，0代表停机)。Where: t is the dispatch time; n is the electric boiler index;

and

are the power consumption and heat generation of the nth electric boiler in period t respectively;

is the electric-to-heat conversion efficiency of the nth electric boiler,

It is divided into the minimum and maximum heating power of the nth electric boiler;

It is the start and stop status of the nth electric boiler in period t (1 represents start, 0 represents stop).

(2.3) Gas turbine model

The gas turbine converts the chemical energy in natural gas into electrical energy. The relationship between the natural gas power consumption and the power generation, power generation limit, ramp constraint and minimum on/off time constraint are as follows:

和

分别表示燃气轮机的发电功率和耗气功率；F(·)表示燃气轮机能耗曲线；

和

分别表示燃气轮机的开机和关机所需天然气消耗量；a_q、b_q和c_q表示F(·)的燃气系数；L^HANG为天然气低热值，取9.7kWh/m³；

分为第q台燃气轮机最小、最大发电功率。

表示第q台燃气轮机在t时段的发电功率；

为第q台燃气轮机在t时段的启停状态(1代表开机，0代表停机)，

为第q台燃气轮机在t时段的启停状态；

为第q台燃气轮机的上爬坡率、下爬坡率，

为第q台燃气轮机在t-1时段内连续开机、停机时间，

为第q台燃气轮机在时段t内的最小开机、停机时间。Where: t is the scheduling time; q is the gas turbine index;

and

They represent the power generation and gas consumption of the gas turbine respectively; F(·) represents the energy consumption curve of the gas turbine;

and

They represent the natural gas consumption required for starting and shutting down the gas turbine respectively; a _q , b _q and c _q represent the gas coefficient of F(·); L ^HANG is the lower calorific value of natural gas, which is 9.7 kWh/m ³ ;

Divided into the minimum and maximum power generation capacity of the qth gas turbine.

represents the power generation of the qth gas turbine in period t;

is the start/stop status of the qth gas turbine in period t (1 represents start, 0 represents stop),

is the start and stop status of the qth gas turbine in period t;

is the ramp-up rate and ramp-down rate of the qth gas turbine,

is the continuous start-up and shutdown time of the qth gas turbine in the t-1 period,

is the minimum startup and shutdown time of the qth gas turbine in time period t.

(2.4) Combined heat and power unit model

Ignoring the impact of external environmental changes on power generation and fuel combustion efficiency, the thermoelectric relationship, gas-electricity relationship, relationship between power generation, power generation limit, ramp constraint and minimum on/off time constraint are as follows:

和

分别表示热电联产机组(CHP)的产热功率和耗气功率；

为微燃机t时段的发电功率、发电效率，

为散热损失率，

和

分别为溴冷机的制热系数和烟气回收率；L^HANG为天然气低热值，取9.7kWh/m³；

分为第p台CHP机组最小、最大发电功率；

表示第p台热电联产机组在t时段的发电功率；

为第p台CHP机组在t时段的启停状态(1代表开机，0代表停机)，

为第P台热电联产机组在t时段的启停状态；

为第p台CHP机组的上爬坡率、下爬坡率；

为第p台CHP机组在t-1时段内连续开机、停机时间，

为第p台CHP机组在时段t内的最小开机、停机时间。Where: t is the dispatch time; p is the index of the cogeneration unit;

and

They represent the heat production power and gas consumption power of the combined heat and power (CHP) unit respectively;

is the power generation and efficiency of the micro-turbine during period t,

is the heat loss rate,

and

are the heating coefficient and flue gas recovery rate of the bromide refrigerator respectively; L ^HANG is the lower calorific value of natural gas, which is 9.7kWh/m ³ ;

It is divided into the minimum and maximum power generation of the pth CHP unit;

represents the power generation of the pth cogeneration unit in period t;

is the start/stop status of the pth CHP unit in period t (1 represents start, 0 represents stop),

is the start and stop status of the Pth cogeneration unit in period t;

is the ramp-up rate and ramp-down rate of the pth CHP unit;

is the continuous start-up and shutdown time of the pth CHP unit in the t-1 period,

is the minimum startup and shutdown time of the pth CHP unit in time period t.

(2.5) Energy storage device model

The relationship between the energy storage equipment model, the gas storage/discharge, heat storage/discharge power constraints of the gas storage and heat storage equipment at time t, and the gas storage and heat storage capacity of the gas storage and heat storage equipment at time t and the gas storage and heat storage capacity at time t-1 and the gas storage/discharge, heat storage/discharge power at time t is as follows:

ESS(t)＝(1-σ ^S )·ESS(t-1)+ES ⁱⁿ (t)·η ⁱⁿ -ES ^out (t)/η ^out

为t时段储气、放气功率，GS^{in ,max}、GS^out,max分别为储气设备的最大储气、放气功率；

为t时段储气设备的储气量，

为t-1时段储气设备的储气量；C^GS,min、C^GS,max分别为储气设备的最小、最大储气容量，η^CGS、η^GS,in、η^GS,out为储气设备自耗率、储气效率、放气效率。

为t时段储热、放热功率，HS^in,max、HS^out,max分别为储热设备的最大储热、放热功率；

为t时段储热设备的储气量，

为t-1时段储热设备的储气量；C^HS,min、C^HS,max分别为储热设备的最小、最大储热容量，η^CHS、η^HS,in、η^HS,out为储热设备自耗率、储热效率、放热效率。Where: t is the scheduling time; ESS(t), ES ⁱⁿ (t), ES ^out (t) are the storage energy, storage energy power and released energy power of the energy storage device in period t respectively; ESS(t-1) represents the storage energy of the energy storage device in period t-1; σ ^S is the self-consumption rate of the energy storage system; η ⁱⁿ and η ^out are the energy storage and release efficiencies of the energy storage device respectively;

is the gas storage and gas release power in period t, GS ^{in ,max} and GS ^out,max are the maximum gas storage and gas release power of the gas storage equipment respectively;

is the gas storage capacity of the gas storage equipment during period t,

is the gas storage capacity of the gas storage equipment in period t-1; C ^GS,min and C ^GS,max are the minimum and maximum gas storage capacities of the gas storage equipment, respectively; η ^CGS ,η ^GS,in and η ^GS,out are the self-consumption rate, gas storage efficiency and gas release efficiency of the gas storage equipment.

is the heat storage and heat release power in period t, HS ^in,max and HS ^out,max are the maximum heat storage and heat release power of the heat storage equipment respectively;

is the gas storage capacity of the heat storage equipment during period t,

is the gas storage capacity of the heat storage equipment in period t-1; C ^HS,min and C ^HS,max are the minimum and maximum heat storage capacities of the heat storage equipment, respectively; η ^CHS , η ^HS,in and η ^HS,out are the self-consumption rate, heat storage efficiency and heat release efficiency of the heat storage equipment.

Step 3: Build a demand response model.

The responsive electric load on the demand side is divided into shiftable electric load and interruptible electric load. For the interruptible load, the interruption cost is considered. The sum of the responsive electric load on the demand side and the electric load demand in the current period must be less than the maximum electric load allowed in the current period. The relationship between the electric load prediction value and the electric load after demand response is:

P_t ^DR、P_t ^DR，inte、P_t ^DR，shif、P_t ^LD，max为t时段电负荷预测值、需求侧响应电负荷、需求侧转移电负荷、系统允许的最大电负荷，P^inte,max为调度时间内最大可中断电负荷。

为t时段最大可中断和可转出的电负荷比例。P_t ^DR,shif为正代表转出可平移负荷，反之，代表转入可平移负荷。Where: t is the scheduling time;

P _t ^DR , P _t ^{DR, inte} , P _t ^{DR, shift} , and P _t ^{LD, max} are the predicted electric load in period t, the response electric load on the demand side, the transfer electric load on the demand side, and the maximum electric load allowed by the system. P ^{inte, max} is the maximum interruptible electric load within the dispatching time.

It is the ratio of the maximum interruptible and transferable loads in period t. A positive value of _{P t} ^DR,shif represents a transferable load, and a negative value represents a transferable load.

The demand side responsive heat load constraints are as follows:

和

分别为时间段t热负荷预测值、需求侧响应热负荷、需求侧可响应热负荷比例、系统允许的最大热负荷和考虑需求响应后的电负荷；H^DR,max为调度时间内最大可中断热负荷；N_t为整个调度时间段。Where: t is the scheduling time;

and

are respectively the heat load forecast value in time period t, the demand side response heat load, the demand side responsive heat load ratio, the maximum heat load allowed by the system and the electric load after considering demand response; H ^DR,max is the maximum interruptible heat load within the scheduling time; N _t is the entire scheduling time period.

Step 4: Under the premise of meeting the system safety constraints, taking the minimum operating cost of the basic scenario as the objective function, a two-stage adjustable robust model considering the joint thermal power demand response is established.

(1) Objective function

The goal of the two-stage robust optimization model proposed in this invention is to minimize the operating cost of the basic scenario under the premise of satisfying the system safety constraints. Since load loss is not allowed in the basic scenario, the objective function does not contain load loss related variables. Its objective function and related equality constraints are as follows:

分别为购电、气费用；

分别为第p台CHP的开机、关机费用；

为弃风惩罚费用，

为弃光惩罚费用；

为售电收益，C^E,cc为需求侧响应中断电负荷的补偿成本；

为购电功率，

为购气功率，

分别为第i台风机、第j组光伏电池在t时段的弃风、弃光功率；P_t ^out为售电功率，P_t ^DR,inte为需求侧中断电负荷。Δt为调度时间间隔，N_t为调度时段数。

分别为单位购电、购气、售电价格，φ_t ^WT、φ_t ^PV分别为单位弃风、弃光的惩罚价格，

为单位需求侧响应中断电负荷的补偿价格。

为第p台CHP在t时段的启停状态(1代表开机，0代表停机)，

分别为第p台CHP的开机、停机一次的费用。N^WT、N^PV、N^CHP分别为风机、光伏电池、电锅炉、CHP的数量。Where: t is the scheduling time;

They are the cost of purchasing electricity and gas respectively;

are the startup and shutdown costs of the pth CHP, respectively;

Penalty fees for wind curtailment,

Penalty fee for abandonment;

is the revenue from electricity sales, C ^E,cc is the compensation cost for the load interruption in response to power outage on the demand side;

For the purchased power,

is the gas purchasing power,

are the abandoned wind and solar power of the i-th wind turbine and the j-th photovoltaic cell in period t, respectively; _Ptout is the electricity sales power, _PtDR ^,inte is the power outage load on the demand side, Δt is the scheduling time interval, ^and _Nt is the number of scheduling periods.

are the unit electricity purchase, gas purchase, and electricity sales prices, respectively; φ _t ^WT and φ _t ^PV are the penalty prices for unit wind and solar abandonment, respectively.

It is the compensation price for unit demand side response to power outage load.

is the start/stop status of the pth CHP in period t (1 represents start, 0 represents stop),

are the startup and shutdown costs of the pth CHP, respectively. N ^WT , N ^PV , and N ^CHP are the numbers of wind turbines, photovoltaic cells, electric boilers, and CHPs, respectively.

(2) Constraints

To ensure the safe operation of the system, this section uses a two-layer max-min model to identify the worst scenario that causes the most unsafe system operation, that is, the maximum violation of the safety regulations (security violation). The maximum violation of the system's safety regulations in the worst scenario must be less than the preset value ε ^RO . The setting of the ε ^RO value is related to the predetermined system safety level to ensure the safe operation of the park's integrated energy system. The maximum value of the violation of the system safety regulations caused by the worst scenario under uncertainty, the uncertainty set of wind power output and photovoltaic output, the system energy balance constraint, the slack variable is always greater than zero constraint, the unit output constraint and the unit error correction climbing constraint under the uncertainty of renewable energy output, the energy storage constraint, the power exchange constraint with the upper network, the demand side response constraint, and the wind and solar abandonment constraint are as follows:

v _1t ,v _2t ,v _3t ,v _4t ≥ 0

P _t ^u,out ≤P ^out,min ,P _t ^u,out ≤P ^out,max :(λ _{9_3,t} ,λ _{9_4,t} )

分别为弃风、弃光、需求响应后电负荷、P2G消耗电功率、电锅炉消耗电功率、园区售电功率、CHP机组出力、燃气轮机出力根据不同风电出力

和光伏出力

调整后的实时值；(·)^u为在风光出力实时变化下对应的调整后变量；v_1,t和v_2,t为电力平衡约束松弛变量；v_3,t和v_4,t为热量平衡约束松弛变量；Ω^WT、Ω^PV分别为风电、光伏出力不确定合集；

分别为风电、光伏出力与预测值的偏差，

为风电、光伏出力偏差值与预测值的比例；

为不确定合集中的0-1变量；Δ_i、Δ_j分别为风电、光伏出力不确定性预算值；

为第p台CHP机组上爬坡和下爬坡纠错能力；

为第q台燃气轮机机组上爬坡和下爬坡纠错能力；λ(·)为约束条件对应的对偶变量。Where: t is the scheduling time;

They are wind curtailment, solar curtailment, power load after demand response, P2G power consumption, electric boiler power consumption, park power sales, CHP unit output, gas turbine output according to different wind power output

and photovoltaic output

The adjusted real-time value; (·) ^u is the adjusted variable corresponding to the real-time change of wind and solar power output; v _1,t and v _2,t are the slack variables of power balance constraint; v _3,t and v _4,t are the slack variables of heat balance constraint; Ω ^WT and Ω ^PV are the uncertain sets of wind power and photovoltaic output respectively;

are the deviations of wind power and photovoltaic output from the predicted values,

is the ratio of wind power and photovoltaic output deviation to the predicted value;

is a 0-1 variable in the uncertainty set; Δ _i and Δ _j are the uncertainty budget values of wind power and photovoltaic power output respectively;

The up-ramp and down-ramp error correction capability of the pth CHP unit;

is the up-ramp and down-ramp error correction capability of the qth gas turbine unit; λ(·) is the dual variable corresponding to the constraint condition.

Step 5: Obtain the abstract expression of the two-stage adjustable robust optimization model for the day-ahead economic dispatch of the park's integrated energy system.

The proposed two-stage robust optimization model involves three systems: electricity, gas, and heat. It involves many constraints and contains uncertain parameters. It is also a nonlinear mixed integer programming problem. For the convenience of discussion, the two-stage robust optimization scheduling model proposed in this chapter can adopt the abstract form of the robust optimization model as follows:

s.t.Ax+By≤b

Where: x represents the start and stop status of the units related to CHP and gas turbine, y and z represent the basic scenario and the dispatching output of the remaining units in the system adjusted according to the wind and solar output conversion, respectively, u is the uncertain variable related to the uncertainty of wind power and photovoltaic output, c _b , c _g , A, B, b, C, D, E, F, and G can be obtained through the objective function and constraints in 5.

Step 6: Establish the maximal and minimal sub-problems for worst-case scenario identification.

The subproblem is to identify the worst-case scenario. Through the double-layer max-min problem shown in step 4, we can find the scenario that causes the system to violate the maximum safety regulations, that is, determine the specific value of the uncertainty in the worst-case scenario. Then, through the duality theory, we can transform the double-layer max-min problem into a single-layer bilinear maximum optimization subproblem as shown below.

Step 7: Use the CCG (column and constraint generation) method to solve the two-stage adjustable robust model considering the combined heat and power demand response.

(7.1) Main problem

Ax+By≤b

(7.2) Sub-problem

(7.3) CCG solution steps

Step (7.3.1): Set the iteration counter s=0, and set the maximum value ε ^RO of the safety violation allowed by the system.

Step (7.3.2): Solve the main problem. If there is a solution, obtain the system unit start and stop status x and the unit output arrangement y, and proceed to step (7.3.3); otherwise, stop the iteration and output that there is no solution.

Step (7.3.3): Based on the x and y obtained in step (7.3.2), solve the maximum and minimum sub-problems to find the wind and photovoltaic output values under the worst scenario that leads to the maximum possible violation of the safety regulations.

向主问题中增加以下所示的CCG约束，返回步骤(7.3.2)。Step (7.3.4): If the maximum possible violation of safety regulations solved in step (7.3.3) is less than ε ^RO , then x and y are the final optimization solutions and the iteration is stopped; otherwise, let s = s + 1, and according to the wind power and photovoltaic output values under the worst scenario solved in step (7.3.3)

Add the following CCG constraints to the main problem and return to step (7.3.2).

f ^T v _s ≤ ε ^RO

Step 8 inputs the park comprehensive energy system data, including the specific composition of the park comprehensive energy system, energy prices, equipment parameters and values of each energy conversion equipment, demand response ratio, new energy output fluctuations under basic scenarios and worst-case scenarios, maximum violation of safety regulations, and predicted values of new energy output and load, etc. The commercial solver Gurobi is used to solve the two-stage adjustable robust optimization operation model of the park comprehensive energy system to obtain the robust optimization scheduling results.

The effects of the present invention are described in detail below through specific embodiments.

(1) Example introduction.

和烟气回收率

分别为0.9、1.2，燃气轮机、CHP、电锅炉的一次开机和停机成本分别为：3.5、1.94、2.74元。假设CHP、燃气轮机初始状态为停运状态，储气设备初始储气量为10m³，储热设备初始储热量为100kW·h，储气、储热设备自耗率为0.01。园区多能源综合系统冬季典型日的风机、光伏出力、电负荷和热负荷的预测值如图4所示。As shown in Figure 2, the specific composition of the park's comprehensive energy system. The simulation selected a multi-energy complementary park system model that includes wind, light, gas, storage, and considers power-to-gas and power-to-heat technologies. The system includes a gas turbine, a wind turbine, an electric boiler, a P2G device, a heat storage device, a gas storage device, two CHPs, and a group of photovoltaic cells. The heating coefficient of the bromide refrigerator

and flue gas recovery rate

They are 0.9 and 1.2 respectively. The startup and shutdown costs of gas turbine, CHP and electric boiler are 3.5, 1.94 and 2.74 yuan respectively. Assume that the initial state of CHP and gas turbine is shutdown, the initial gas storage capacity of gas storage equipment is ^10m3 , the initial heat storage capacity of heat storage equipment is 100kW·h, and the self-consumption rate of gas storage and heat storage equipment is 0.01. The predicted values of wind turbine, photovoltaic output, electric load and heat load on a typical day in winter for the park's multi-energy integrated system are shown in Figure 4.

(2) Description of implementation scenario

In order to verify the effectiveness of the proposed two-stage robust adjustable optimization model considering the combined heat and power demand response, seven scheduling operation modes shown in Table 1 are set.

Table 1 7 scheduling operation modes

The wind and photovoltaic forecast outputs are adjusted to 2 times the original, and at the same time, the uncertainty of wind and photovoltaic output is considered to obtain new operation plans 1-7. The threshold for violating the safety regulations is set to 0, that is, the system is not allowed to lose load or overload in any scenario. It is assumed that the uncertainty budget of wind power output and photovoltaic output is 24.

The electric load is increased to 2.4 times of the original value. At the same time, the uncertainty of wind and solar power output is taken into account to obtain a new operation plan 8-14, and the violation value of safety regulations is set to 0.

(3) Analysis of the results of the implementation examples.

Table 2 shows the operating costs and safety violation values of operating schemes 1-7 under different uncertainty ratios, from which we can see that as the uncertainty ratio of wind power and photovoltaic output increases, the operating cost of the park increases, and the maximum possible safety violation value of the park also increases. However, with the continuous addition of energy conversion equipment such as heat storage equipment, P2G and its gas storage equipment, the park can continuously adjust the energy storage state in real time to cope with the real-time changes in wind power and photovoltaic output, reduce the system operation risk, consider demand response, and flexibly respond to system uncertainty. Considering the combined thermal power demand response, the operating cost of the park in the basic scenario is further reduced.

Table 2 Operating costs and safety violation values of operating schemes 1-7 under different uncertainty ratios

Table 3 shows the number of iterations and the value of violation of safety regulations for running scheme 1 under different wind power and photovoltaic uncertainty budgets. It can be concluded that the larger the uncertainty budget value of renewable energy, the fewer iterations are required to obtain the optimization scheme that meets the system requirements, and the greater the value of violation of safety regulations that may occur in the system. The robustness of the system can be adjusted by adjusting the uncertainty budget value.

Table 3. The number of iterations and safety violation values of operation scheme 1 under different wind power and photovoltaic uncertainty budgets

Table 4 shows the operating costs and safety violation values of operating schemes 8-14 under different uncertainty ratios, from which it can be concluded that: when the system load demand is significantly greater than the renewable energy output, the renewable energy output can be fully absorbed by the system, gas-to-electricity is better than electricity-to-gas, P2G equipment does not work, and the introduction of energy storage equipment cannot improve the robustness of the system. When the uncertainty ratio of renewable energy output is 0.2, compared with schemes 8-10, it can be seen that the more comprehensive the equipment contained in the system, the lower the operating cost of the system in the basic scenario. Compared with schemes 10-14, it can be seen that the introduction of demand-side response can further reduce the operating cost of the system. Comparing the operating costs and safety violation values of schemes 11-14 under different renewable energy uncertainty ratios, it can be seen that the combined thermal power demand response can reduce the power load loss by reducing the thermal load, greatly enhancing the robustness, flexibility and economy of the park's integrated energy system.

Table 4 Operating costs and safety violation values of operation schemes 8-14 under different uncertainty ratios

The above description is only a specific embodiment of the present invention, but it does not limit the patent protection scope of the present invention. Any equivalent changes or substitutions made by using the contents of the present invention and the drawings, directly or indirectly applied to other related technical fields, should be included in the protection scope of the present invention.

Claims (1)

1. A park comprehensive energy system day-ahead economic dispatching method considering wind and light uncertainty is characterized by comprising the following steps:

step 1: determining the specific composition of the comprehensive energy system of the park, including the introduced new energy form and the specific equipment composition;

step 2: establishing models of various energy conversion equipment in the park;

and 3, step 3: establishing a demand response model;

and 4, step 4: on the premise of meeting system safety constraints, establishing a two-stage adjustable robust model considering combined heat and power demand response by taking the minimum running cost of a basic scene as a target function;

step 5, obtaining an abstract expression of a two-stage adjustable robust optimization model of the park comprehensive energy system day-ahead economic dispatching;

step 6, establishing the maximum and minimum subproblems of the worst scene identification;

step 7, solving a two-stage adjustable robust model considering combined heat and power demand response by using a CCG method;

and step 8: inputting energy access, new energy output data, equipment parameters and operating parameters of the park integrated energy system, and solving a two-stage robust optimization model of the park integrated energy system day-ahead economic dispatching considering wind-solar uncertainty by adopting a commercial solver to obtain a dispatching strategy of the park integrated energy system day-ahead economic dispatching model;

the park comprehensive energy system in the step 1 specifically comprises the following components:

(1) The new energy form of the integrated energy system accessed to the park is as follows: wind power and photovoltaic power generation;

(2) The energy conversion equipment for introducing the park comprehensive energy system comprises: electricity-to-gas equipment, an electric boiler, a gas turbine, a cogeneration unit, and gas/heat storage equipment;

step 2, each energy conversion equipment model is as follows;

(1) Electric gas conversion equipment model

In the formula: t is scheduling time; m is an index of the electric-to-gas equipment;

respectively gas making power, consumed power and electricity-to-gas efficiency, L, of an electricity-to-gas (P2G) facility ^HANG Taking 9.7kWh/m as low heating value of natural gas ³ ；

The minimum and maximum breathing power of the mth P2G;

(2) Electric boiler model

In the formula: t is scheduling time; n is an electric boiler index;

and &>

Respectively the power consumption and the heat production of the nth electric boiler in the t period;

Is the electric heat conversion efficiency of the nth electric boiler>

Dividing into the minimum and maximum heating power of the nth electric boiler;

Starting and stopping the nth electric boiler at a time period t;

(3) Gas turbine model

In the formula: t is scheduling time; q is a gas turbine index;

and &>

Respectively representing the power generation power and the gas consumption power of the gas turbine; f (-) represents the gas turbine energy consumption curve;

And &>

Respectively representing the consumption of natural gas required by the startup and shutdown of the gas turbine; a is _q 、b _q And c _q A gas coefficient representing F ('); l is ^HANG Taking 9.7kWh/m as low heating value of natural gas ³ ；

Dividing the power into the minimum and maximum power generation powers of the q gas turbines;

Representing the generated power of the qth gas turbine in the t-1 period;

For the start-stop state of the qth gas turbine in the time period t, ->

Starting and stopping states of a qth gas turbine in a t-1 time period;

Is the climbing rate and the descending rate of the qth gas turbine>

For the continuous start-up and shutdown time of the qth gas turbine in the t-1 time period, based on the preset time interval>

The minimum startup and shutdown time of the qth gas turbine in the time period t is defined;

(4) Combined heat and power generation unit model

In the formula: t is scheduling time; p is an index of the cogeneration unit;

and &>

Respectively representing heat production power and gas consumption power of a combined heat and power generation unit (CHP);

The generating power and the generating efficiency of the micro-combustion engine in the t period are based on the pressure value>

For the heat dissipation loss rate, the device can be used for>

And &>

The heating coefficient and the flue gas recovery rate of the bromine refrigerator are respectively; l is a radical of an alcohol ^HANG Taking 9.7kWh/m as low heating value of natural gas ³ ；

Dividing into the p (th) CHP unit minimum and maximum generating power;

Representing the generated power of the p-th combined heat and power generation unit in a t-1 period;

For the start-stop state of the pth CHP unit in the t-1 time period, the system is turned on or off>

The starting and stopping state of the P-th cogeneration unit in the t-1 time period is shown;

The climbing rate and the descending rate of the pth CHP unit;

for the continuous startup and shutdown time of the pth CHP unit in the t-1 period>

The minimum startup and shutdown time of the pth CHP unit in the time period t is obtained;

(5) Energy storage equipment model

ESS(t)＝(1-σ ^S )·ESS(t-1)+ES ⁱⁿ (t)·η ⁱⁿ -ES ^out (t)/η ^out

In the formula: t is scheduling time; ESS (t), ES ⁱⁿ (t)、ES ^out (t) storing energy, stored energy power and released energy power of the energy storage device at a time period t, respectively; ESS (t-1) represents the stored energy of the energy storage device during the t-1 period; sigma ^S Is the self-consumption rate of the energy storage system; eta ⁱⁿ 、η ^out Respectively storing energy and releasing energy efficiency for the energy storage equipment;

gas storage, gas discharge power, GS, for a period of t ^in,max 、GS ^out,max The maximum gas storage and gas release power of the gas storage device are respectively;

The air storage quantity of the air storage device is greater or less for a time period t>

The gas storage capacity of the gas storage equipment is t-1 time; c ^GS,min 、C ^GS,max Respectively, the minimum and maximum gas storage capacities, eta, of the gas storage equipment ^CGS 、η ^GS,in 、η ^GS,out The self-consumption rate, the gas storage efficiency and the gas release efficiency of the gas storage equipment are improved;

For the heat-storage and heat-release power of t time period, HS ^{in ,max} 、HS ^out,max The maximum heat storage power and the maximum heat release power of the heat storage equipment are respectively;

For a gas reserve of the heat storage device for a period t>

The gas storage capacity of the heat storage equipment is t-1 time period; c ^HS,min 、C ^HS,max Minimum and maximum heat storage capacity, eta, of the heat storage apparatus ^CHS 、η ^HS,in 、η ^HS,out The self consumption rate, the heat storage efficiency and the heat release efficiency of the heat storage equipment are obtained;

step 3, the demand response model is as follows:

P _t ^DR ＝P _t ^DR,inte +P _t ^DR,shif

in the formula: t is scheduling time;

P _t ^DR 、P _t ^DR,inte 、P _t ^DR,shif 、P _t ^LD,max predicted value of electric load, demand side response electric load, demand side interruptible electric load, and demand side transfer for t periodElectric load, maximum electric load allowed by the system, P ^inte,max The maximum interruptible electrical load within the scheduling time;

A maximum interruptible and exportable electrical load ratio for the t period; p is _t ^DR,shif Positive represents the load that can be shifted out, and vice versa represents the load that can be shifted in;

in the formula: t is scheduling time;

and &>

Respectively a predicted value of the heat load, a response heat load of a demand side, a response heat load proportion of the demand side, the maximum heat load allowed by a system and the heat load after considering the demand response in a time period t; h ^DR,max The maximum interruptible thermal load within the scheduling time; n is a radical of _t Scheduling the time period for the whole;

step 4 the two-stage tunable robust model considering the combined heat and power demand response is as follows:

(1) Objective function

In the formula: t is scheduling time;

the electricity and gas purchase costs are respectively;

The starting-up and shutdown costs of the pth CHP are respectively;

Penalty fee for wind abandon, based on the sum of the wind and the wind>

Punishment of cost for light abandonment;

To earn electricity sales, C ^E,cc A compensation cost to interrupt the electrical load for demand side response; p _t ⁱⁿ For purchasing power, is turned on or off>

For purchasing air power, is selected>

Wind abandoning power and light abandoning power of the ith fan and the jth group of photovoltaic cells in the t period are respectively set; p is _t ^out To sell electric power, P _t ^DR,inte For interrupting the electrical load on the demand side, Δ t is the scheduling time interval, N _t For scheduling period number, ->

The unit purchase price of electricity, gas and electricity, phi _t ^WT 、φ _t ^PV A penalty price for wind abandon and light abandon of the unit, respectively>

For the compensation price of the unit demand side in response to an interrupted electrical load, based on the value of the compensation>

The starting and stopping states of the pth CHP in the period t, 1 represents starting, 0 represents stopping,

the cost of starting up and stopping the p-th CHP once respectively, N ^WT 、N ^PV 、N ^CHP The number of the fan, the photovoltaic cell, the electric boiler and the CHP are respectively;

(2) Constraint conditions

P _t ^u,DR ＝P _t ^u,DR,inte +P _t ^u,DR,shif :(λ _{10_6,t} )

In the formula: t is scheduling time; (. Cndot.) ^u Is a corresponding adjusted variable under the real-time change of wind and light output; v. of _1,t And v _2,t A power balance constraint relaxation variable; v. of _3,t And v _4,t Constraint relaxation variables for thermal balance; omega ^WT 、Ω ^PV Respectively representing uncertain collections of wind power output and photovoltaic output;

deviation of wind power, photovoltaic output and predicted value respectively>

The ratio of the wind power and photovoltaic output deviation value to the predicted value is obtained;

Is a variable of 0 to 1 in the uncertain convergence set; delta _i 、Δ _j Respectively obtaining wind power and photovoltaic output uncertainty precalculated values;

Correcting the climbing and descending of the pth CHP unit;

Correcting the climbing and descending of the qth gas turbine unit; lambda (-) is a dual variable corresponding to a constraint condition>

And &>

Respectively taking the average values of the predicted values of the wind power output and the photovoltaic output;

And &>

Respectively the abandoned wind power and the abandoned light power after the uncertainty parameters are considered;

Representing the wind abandoning proportion;

and 5, an abstract expression of the two-stage adjustable robust optimization model of the park comprehensive energy system economic dispatch in the day ahead is as follows:

s.t.Ax+By≤b

in the formula: x represents the starting and stopping states of units related to the CHP and the gas turbine, y and z represent the basic scene and the scheduling output of other units of the system adjusted according to the wind-light output transformation, u is an uncertain variable related to the uncertainty of the wind power output and the photovoltaic output, and c _b 、c _g A, B, B, F, h, C, D, E, F, G can be derived from the objective function and constraint conditions, C _b And c _g The method comprises the following steps that A, B, C, D, E, F and G are abstract coefficient vector matrixes of variables in inequality constraints respectively; b. h represents an abstract matrix of constants in the inequality constraint; f is an abstract coefficient vector corresponding to a max-min double-layer problem objective function;

step 6, the maximum and minimum subproblems of the worst scene recognition are as follows:

ΔD＝ΔD ₁ +ΔD ₂

s.t.λ _{9_1,t} ≤0,λ _{9_2,t} ≤0,λ _{9_3,t} ≤0,λ _{9_4,t} ≤0,λ _{9_5,t} ≤0,λ _{9_6,t} ≤0,t∈N ^T

λ _{6_2,q,t} ≤0,λ _{6_3,q,t} ≤0,λ _{6_4,q,t} ≤0,λ _{6_6,q,t} ≤0,λ _{6_7,q,t} ≤0,q∈N ^GT ,t∈N ^T

λ _{7_2,p,t} ≤0,λ _{7_3,p,t} ≤0,λ _{7_4,p,t} ≤0,λ _{7_6,p,t} ≤0,λ _{7_7,p,t} ≤0,p∈N ^CHP ,t∈N ^T

λ _{8_1,t} ≤0,λ _{8_2,t} ≤0,λ _{8_3,t} ≤0,λ _{8_4,t} ≤0,t∈N ^T

λ _{13_1,t} ≤0,λ _{13_2,t} ≤0,λ _{13_3,t} ≤0,λ _{13_4,t} ≤0,t∈N ^T

λ _{4_2,m,t} ≤0,λ _{4_3,m,t} ≤0,t∈N ^T ,m∈N ^P2G ,t∈N ^T

λ _{5_2,n,t} ≤0,λ _{5_3,n,t} ≤0,t∈N ^T ,n∈N ^EB ,t∈N ^T

λ _{10_1,t} ≤0,λ _{10_2,t} ≤0,λ _{10_3,t} ≤0,λ _{10_4,t} ≤0,t∈N ^T

λ _{11_3,t} ≤0,λ _{11_3,t} ≤0,λ _{12_1,t} ≤0,λ _{12_2,t} ≤0,t∈N ^T

λ _{10_7} ≤0,λ _{10_8} ≤0,λ _{11_4} ≤0,λ _{11_5} ≤0,t∈N ^T

s.t.-λ _1,t -λ _{6_1,q,t} ·b _q -λ _{6_2,q,t} +λ _{6_3,q,t} -λ _{6_4,q,t+1} +

λ _{6_4,q,t} +λ _{6_5,p,t+1} -λ _{6_5,p,t} -λ _{6_6,q,t} +λ _{6_7,q,t} ≤0

-λ _1,t -λ _{9_1,t} +λ _{9_2,t} ≤0,t∈N ^T

λ _1,t -λ _{9_3,t} +λ _{9_4,t} ≤0,t∈N ^T

λ _1,t +λ _{10_5,t} ≤0,t∈N ^T

-λ _{10_1,t} -λ _{10_2,t} +λ _{10_3,t} +λ _{10_5,t} -λ _{10_7} +λ _{10_8} ＝0,t∈N ^T

-λ _{10_1,t} +λ _{10_4,t} +λ _{10_5,t} +λ _{10_5,t} +λ _{10_6} ＝0,t∈N ^T

λ _1,t +λ _{12_1,i,t} ≤0,t∈N ^T ,i∈N ^WT

λ _1,t +λ _{12_2,j,t} ≤0,t∈N ^T ,j∈N ^PV

-λ _2,t +λ _{7_1,p,t} ≤0,t∈N ^T ,p∈N ^CHP

-λ _2,t +λ _{6_1,q,t} ≤0,t∈N ^T ,q∈N ^GT

-λ _2,t +λ _{4_1,m,t} -λ _{4_2,m,t} +λ _{4_3,m,t} ≤0,t∈N ^T ,m∈N ^P2G

λ _2,t -λ _{9_5,t} +λ _{9_6,t} ≤0,t∈N ^T

λ _2,t +λ _{8_2,t} +λ _{8_5,t} /η ^GS,out ≤0,t∈N ^T

-λ _2,t +λ _{8_1,t} -λ _{8_5,t} ·η ^GS,in ≤0,t∈N ^T

-λ _{8_3,t} +λ _{8_4,t} -λ _{8_5,t+1} ·η ^GS,in +λ _{8_5,t} ≤0,t∈N ^T -λ _3,t ·η ^h +λ _{13_2,t} +λ _{13_5,t} /η ^HS,out ≤0,t∈N ^T

λ _3,t ·η ^h +λ _{13_1,t} -λ _{13_5,t} ·η ^HS,in ≤0,t∈N ^T -λ _{13_3,t} +λ _{13_4,t} -λ _{13_5,t+1} ·η ^HS,in +λ _{13_5,t} ≤0,t∈N ^T

-λ _3,t ·η ^h +λ _{5_1,n,t} -λ _{5_2,n,t} +λ _{5_3,n,t} ≤0,t∈N ^T ,n∈N ^EB -λ _3,t ·η ^h +λ _{7_8,n,t} ≤0,t∈N ^T ,n∈N ^EB

λ _3,t +λ _{11_1,t} ≤0,t∈N ^T λ _{11_1,t} -λ _{11_2,t} +λ _{11_3,t} -λ _{11_4} +λ _{11_5,t} ≤0,t∈N ^T

-1≤λ _1,t ≤1,-1≤λ _3,t ≤1,t∈N ^T ；

and 7, solving a two-stage adjustable robust model considering combined heat and power demand response by using a CCG method as follows:

(1) Major problems

Ax+By≤b

(2) Sub-problems

(3) CCG solving step

Step 1: let the iteration counter s =0 set the maximum value epsilon allowed by the system for violating the safety regulations ^RO ；

Step 2: solving the main problem, if the main problem is solved, obtaining a system unit starting and stopping state x and a unit output arrangement y, and performing the step 3; otherwise, stopping iteration and outputting no solution;

and 3, step 3: solving the maximum and minimum subproblems according to the x and y obtained by the solving in the step 2, and finding out the wind power and photovoltaic output under the worst scene which causes the maximum possibility of violating the safety specified value;

and 4, step 4: if solved in step 3Maximum possible violation of the safety prescribed value less than epsilon ^RO Then x and y are the final optimization solution and the iteration is stopped; otherwise, let s = s +1, according to the wind power and photovoltaic output value under the worst scene solved in step 3

Adding CCG constraint shown below into the main problem, and returning to the step 2;

and 8, the park comprehensive energy system data further comprises the specific composition of the park comprehensive energy system, the energy price, the equipment parameters and values of each energy conversion equipment, the demand response proportion, the new energy output fluctuation situation under the basic scene and the worst scene, the maximum violation safety specified value and the predicted value of the new energy output and load.

Images