CN109861302B - Master-slave game-based energy internet day-ahead optimization control method - Google Patents

Abstract

An energy Internet optimization control method based on master-slave game, first, initialize the system and obtain relevant parameters, first set the initial internal price by the network group control center, each ELN subnet makes decisions according to the initial price, and calculates the corresponding The optimal strategy of , the network group control center integrates the strategy set, aiming at maximizing the benefits of the network group control center, and calculates the updated internal price. Then output the optimization strategy set at this time as the optimization result of the day before. The invention can effectively improve the energy Internet's ability to absorb new energy and the system reliability in case of failure, and increase the economic benefits of the energy Internet to a certain extent.

Description

A day-ahead optimization control method for energy internet based on master-slave game

technical field

The present invention relates to an energy internet optimization control method based on master-slave game.

Background technique

Scholars from various countries have put forward the concept of energy Internet that deeply couples energy network and information network by combining smart grid, microgrid and other power grid technologies and multi-dimensional interconnected Internet technologies. Energy Local Network (ELN) is the most basic optimization of energy Internet. Since it was proposed, it has become a research hotspot in academia and industry.

Energy Internet is a comprehensive energy system that deeply integrates electricity, natural gas, and new energy, and interconnects energy production, manufacturing, storage, and transmission, and maximizes the optimal allocation of resources. However, there are many distributed devices in the energy Internet, and the amount of data received and analyzed by the dispatch center is large, and the communication time increases accordingly. Moreover, with the continuous access of large-scale distributed renewable energy to the power grid, the utilization of new energy is unstable in most occasions, the load demand is fluctuating, and the real-time electricity price is random, which introduces a lot of uncertainty to the supply and demand side. It causes power imbalance in the output and consumption of the whole system. The imbalance between supply and demand not only causes energy waste, but also seriously affects the safe and stable operation of the Energy Internet. Therefore, the optimal operation and management of multi-energy systems has become the key to solving the problem of energy Internet.

The traditional power grid adopts centralized control and dispatching, and the optimal dispatching can only be realized by obtaining the global information through the central controller, so the calculation amount is large, the reliability is poor, the investment cost is high, and the energy transmission lacks flexibility. With the advent of the energy Internet era, the application of energy LAN groups has become more and more extensive. The energy local area network realizes real-time sharing of information through two-way interaction between the supply and demand sides, enabling each ELN to interact according to their own conditions, and real-time decision-making between ELN groups to update, solve the problem of power imbalance, and enhance system security and reliability. Multi-energy complementarity is carried out among the sub-networks to improve the penetration rate of renewable energy. In order to promote the coordination and optimization of source, load and storage in the energy local area network, we urgently need to do a deeper research on the optimization management of the energy Internet.

SUMMARY OF THE INVENTION

In order to effectively complete the interconnected operation control of the energy local area network group and overcome the instability problem of new energy utilization in most occasions, the present invention conducts research on the energy Internet optimization management scheme based on the master-slave game. The present invention introduces predictive control technology, establishes an energy local area network group model based on master-slave game theory, reduces the uncertainty of predictive data, realizes rational allocation of energy, and maximizes benefits.

In order to achieve the above object, the technical scheme of the present invention is:

An energy internet optimization control method based on a master-slave game, comprising the following steps:

S1: First construct the energy local area network cluster model, initialize the system and obtain the parameters required for optimization, including the forecast data of wind energy, solar energy and energy storage;

S2: Establish a master-slave game model, the network group control center is the leader, which sets the initial internal price, and each ELN subnet is the follower, making decisions based on the initial internal price, and calculating the corresponding optimal strategy;

S3: The network group control center integrates the strategy sets of each ELN subnet, and recalculates the internal price with the goal of maximizing the benefits of the network group control center, which is defined as the update internal price;

S4: Each ELN subnet makes a decision based on the updated internal price, and calculates the optimal strategy under which the internal price should be updated;

S5: When the game reaches the Stackelberg Equilibrium (SE) and the internal price is no longer updated, the final optimization set is output as the result of the previous optimization of the energy local area network group;

S6: If the game does not reach the Stackelberg equilibrium, return to step S2 to re-optimize according to the updated state information.

In the present invention, the energy internet environment is composed of multiple energy local area networks, the power supply side of each energy local area network individual is composed of photovoltaics, fans, gas turbines, energy storage and other power grids, and the demand side is composed of basic load and electric refrigeration. The heat load is supplied by gas turbines, heat collectors and gas boilers, the cooling load is supplied by gas turbines and electric refrigerators, and the overall ELN group is supplied or consumed by individual ELNs and external power grids.

Further, in the step S1, the system model includes the following components:

S1-1. Basic load model: ELN includes three types of loads, namely heating load, cooling load, and electrical load. The models are as follows:

Heat load: Provided by gas boilers, heat exchangers and collectors:

是ELNi中燃气锅炉的热功率；

是热交换器对外输出的热功率；

是ELNi中热负荷的总功率；in,

is the thermal power of the gas boiler in ELNi;

is the thermal power output by the heat exchanger;

is the total power of the thermal load in the ELNi;

The thermal power output by the gas boiler is related to the fuel usage and the heat production efficiency of the boiler.

是ELNi中燃气锅炉的最大热功率；η_GB是燃气锅炉的产热效率；V_GB,i为锅炉在一个时段内的燃气使用量；L_NG是天然气的热值，为9.7kWh/m³；in,

is the maximum thermal power of the gas boiler in ELNi; η _GB is the heat production efficiency of the gas boiler; V _GB,i is the gas consumption of the boiler in a period of time; LNG is the calorific value of natural gas, which is _9.7kWh /m ³ ;

From the fuel consumption of the gas turbine and the gas boiler, the total gas consumption V _SUM of the system can also be obtained as:

Among them, V _GT,i is the natural gas consumption of the ELNi gas turbine in period t;

As the main controllable energy supply equipment in the ELN system, the gas turbine not only provides electrical energy for the ELN, but also the heat carried by the high-temperature flue gas generated by the gas turbine is recovered by the heat recovery device, and heat is supplied to the heat load through the heat exchanger and the absorption refrigeration device. The relationship between the output, heat and gas consumption of ELNi's gas turbine to supply cooling for the cooling load is as follows:

为t时段ELNi燃气轮机的发电功率；

为ELNi燃气轮机的最大发电功率；

为t时段ELNi燃气轮机的余热回收功率；η_c和η_r为ELNi燃气轮机的发电效率和余热回收效率；

是t时段ELNi燃气轮机的天然气消耗量；in,

is the power generation of the ELNi gas turbine in period t;

is the maximum generating power of the ELNi gas turbine;

is the waste heat recovery power of the ELNi gas turbine in period t; η _c and η _r are the power generation efficiency and waste heat recovery efficiency of the ELNi gas turbine;

is the natural gas consumption of the ELNi gas turbine in period t;

In addition, the power generation efficiency and waste heat recovery efficiency of the gas turbine are greatly affected by the unit load rate. The relationship between the two and the unit load rate β is as follows:

0.25≤β≤1 (10)

为燃气轮机的额定发电效率；

为燃气轮机的额定余热回收效率；β为机组负载率；In the formula,

is the rated power generation efficiency of the gas turbine;

is the rated waste heat recovery efficiency of the gas turbine; β is the unit load rate;

The heat exchanger will exchange the part used for heating from the waste heat recovered by the gas turbine with water to obtain the output thermal power;

是

中分配出来用于制热的部分；η_HX为热交换器的换热效率；In the formula,

Yes

The part allocated for heating; η _HX is the heat exchange efficiency of the heat exchanger;

The cooling load is provided by absorption refrigeration units and electric chillers:

是ELNi中冷负荷的总功率；

是吸收式制冷装置对外输出的冷功率；

为ELNi的电制冷机制冷功率；in,

is the total power of the cooling load in the ELNi;

is the cooling power output by the absorption refrigeration device;

It is the cooling power of the electric refrigerator of ELNi;

The absorption refrigeration device provides the part of the waste heat used for refrigeration to the heat exchanger in the device, so that the device can convert cold energy;

是

中分配出来用于制冷的部分；η_AC为吸收式制冷装置的制冷效率；In the formula,

Yes

The part allocated for refrigeration; η _AC is the refrigeration efficiency of the absorption refrigeration device;

The electric refrigerator is a special load used to generate the cooling load, and it can also be adjusted. When the cooling load is known, it belongs to the determined party. The output of the electric refrigerator of the i-th ELN should meet the following requirements:

为ELNi的电制冷机输入功率；η_EC为电制冷机的制冷效率；

为 ELNi的电制冷机最大输入功率；in,

is the input power of the electric refrigerator of ELNi; η _EC is the cooling efficiency of the electric refrigerator;

It is the maximum input power of the electric refrigerator of ELNi;

Electric load: It is the only one that can be regulated among the three mentioned above. According to whether it is related to the other two types, the present invention mainly divides the electric load into two types: basic load and electric refrigerator. The basic load model is as follows:

For ELNi∈I, the base load is as follows:

是采用外部电网电价的情况下ELNi的基础负荷；λ_b代表从外部电网购买的电能价格；λ_s代表卖给外部电网的电能价格，r_b代表ELN群的内部购电价格，r_s代表ELN群的内部售电价格，电价应满足以下约束：in,

is the base load of ELNi in the case of using the external grid electricity price; λ _b represents the electricity price purchased from the external grid; λ _s represents the electricity price sold to the external grid, _rb represents the internal electricity purchase price of the ELN group, and _rs represents the ELN The price of electricity within the group should meet the following constraints:

λ _s ≤r _s <r _b ≤λ _b (17)

S1-2. Energy storage system model

The energy storage system reduces the net load of a single ELN and the entire ELN group through two controllable operations of charging and discharging. The SoC of each time period is related to the charge and discharge state and charge and discharge amount of the previous period. In the t time period , the working model of the energy storage system of the i-th ELN is as follows:

为t时段ELNi储能系统储存的能量；

t时段ELNi储能的剩余容量；Q_BES,i为ELNi储能系统的总容量；

为t时段ELNi储能系统的充电功率；

为t时段ELNi储能系统的放电功率；η_ch和η_dch为储能系统的充电效率和放电效率；In the formula,

is the energy stored by the ELNi energy storage system for the period t;

Remaining capacity of ELNi energy storage in period t; Q _BES,i is the total capacity of ELNi energy storage system;

is the charging power of the ELNi energy storage system in period t;

is the discharge power of the ELNi energy storage system in the t period; _ηch and _ηdch are the charging efficiency and discharging efficiency of the energy storage system;

Not only that, the energy storage system in the ELN also needs to constrain its own charge and discharge power and the state of the SoC, and at the same time meet the requirement that the state of the SoC remains unchanged before and after one day of operation:

和

分别表示ELNi储能系统的充放电功率；

和

为ELNi 储能系统的最大充放电功率；

与

代表ELNi储能系统的充放电状态，取 0或1，其中0表示不处于充放电状态，1表示处于充放电状态；

表示ELN i储能系统在t时段的充放电功率；

则为ELNi储能系统剩余容量的上下限；In the formula,

and

respectively represent the charging and discharging power of the ELNi energy storage system;

and

is the maximum charge and discharge power of the ELNi energy storage system;

and

Represents the charge and discharge state of the ELNi energy storage system, taking 0 or 1, where 0 means not in charge and discharge state, and 1 means in charge and discharge state;

Represents the charge and discharge power of the ELN i energy storage system in period t;

is the upper and lower limits of the remaining capacity of the ELNi energy storage system;

S1-3. Renewable Energy Model

The photovoltaic system has the maximum power point tracking function, which can be adjusted according to the light intensity and ambient temperature, and can track and output the maximum power of the time period. The output of ELNi's photovoltaic system is:

In the formula, P _pv,i is the average value of the power of all photovoltaic systems;

The fan system converts the mechanical energy of the wind into electrical energy, and the output power fluctuates with the change of the local average wind speed during the time period. The output of the fan system of ELNi is:

In the formula, P _WT,i is the average power of all fan systems;

S1-4.ELN Power Balance Model

Through the various components in the ELN on the supply side and the demand side, the electric power balance model of the i-th ELN is obtained:

为ELNi与网群的交换功率；

为输送线功率约束。in,

It is the exchange power between ELNi and the network group;

Power constraints for the transmission line.

Further, the establishment of the master-slave game model in the step S2 is as follows:

S2-1. The revenue model of a single ELN

的平方相关，电能交易的收益是在网群内部交易的收益，需要使用网群控制中心和各ELN子网之间的购电价格与售电价格，此外，当ELN联络线上的交换功率为正时，收益为

否则，收益为

燃气消耗的收益表示为燃气消耗成本的相反数；The function of the power balance is to reduce the net load of the ELN, here it is set to exchange power between the power balance and the ELN tie line

The square correlation of , the income of electric energy trading is the income of trading within the network group, it is necessary to use the electricity purchase price and electricity selling price between the network group control center and each ELN sub-network, in addition, when the exchange power on the ELN connection line is time, the profit is

Otherwise, the payoff is

The benefit of gas consumption is expressed as the inverse of the gas consumption cost;

From this, the utility function of ELN is obtained as:

相同；r_NG为燃气价格；In the formula, ρ is the correlation coefficient of the power balance utility, which is in the order of magnitude with

The same; _rNG is the gas price;

S2-2. The income model of the ELN group network group control center

Because the ELN in the system exchanges power through a bus, and exchanges energy with the external power grid through the bus, there are certain constraints on the power of the bus itself;

The network group control center is an artificially set organization in the system. As the leader of the master-slave game model, its optimization goal is to maximize its own interests. The profit function of the network group control center is set as follows:

和

为所有在t时段购电和售电的ELN交换功率的总和。in,

and

is the sum of the power exchanged by all ELNs that purchase and sell electricity in period t.

In step S3, the implementation process of the master-slave game is as follows:

The network group control center provides an initial internal price, first simulates the response of each ELN to the initial internal price, obtains the determined electric load size, and performs local optimization according to the utility function of each ELN to obtain all the optimal strategies of each ELN, and then obtains The interactive power u _grid between each ELN and the network group control center is obtained. According to the response of each ELN to the initial internal electricity price, and considering the influence of the internal electricity price on the electric load, the network group control center optimizes with the goal of maximizing its own interests, so as to set r _b and _rs , and the network group control center uses this Iterate for the initial value of the next optimization, and finally obtain the optimal r _b and _rs , and obtain the scheduling scheme of each ELN accordingly;

The resulting game model is as follows:

L={(I∪{GM}),{P _Load,i } _i∈I ,{R _b },{R _s },{U _i } _i∈I ,R} (31)

It contains the following components:

1) The set I of ELN is a follower and responds to the internal interactive electricity price set by the network group control center GM as the leader;

2) P _Load,i is the strategy set for ELN i to adjust the load, and the corresponding variable P _Load,i contains constraints

3) R _b and R _s are the strategy sets of GM, corresponding to the decision variables r _b and _rs of GM;

4) U _i is the profit function of ELNi, which can be represented by the variable P _Load,i and constraints (5)-(11), (14)-(15), (17)-(21) in the upper optimization;

5) R is the revenue function of GM, and its function is to obtain the profit of energy interaction within the ELN group and conduct transactions with the external power grid.

In step S5, the realization process of Stackelberg equilibrium is as follows:

Based on the master-slave game model in the above step S3, the network group control center determines the optimal price through the best response of each ELN, and each ELN determines the optimal value of its own decision based on the price, and achieves the above through the master-slave game. The way of the scheme is Stackelberg Equilibrium (SE);

为In the Stackelberg game L defined by (31), when a strategy set

for

都有

即

为

的集合；而

Among them, for

have

which is

for

collection; and

与

以及SE时的负荷需求

唯一确定。Therefore, when the above conditions are met, that is, when the game reaches SE, the SE electricity price

and

and load requirements at SE

Only sure.

In step S6, it is judged whether the updated internal electricity price maximizes the benefit of the network group control center, and if the updated internal price does not change any more, the final optimization strategy set is output as the result of the previous optimization of the energy local area network group, otherwise, skip to step S2 Optimize again.

The beneficial effects of the present invention are:

1. Complete the interconnection control of the energy local area network group, realize the rational allocation of energy and the construction of the energy Internet group, promote the consumption of renewable energy, and enhance the reliability of the power grid.

2. Guarantee the load demand, coordinate the output plan of multi-energy and energy storage in the ELN subnet and the change of the internal real-time electricity price, track the output of new energy in time, smooth the fluctuation of the net load, balance the power of the ELN group, and effectively improve the stability of the system operation .

3. The proposed optimization method has strong robustness in the presence of uncertainty in the forecast data, which can effectively alleviate the influence of system instability and uncertainty, and ensure the effective implementation of the scheduling plan and the stable operation of the system.

Description of drawings

Fig. 1 is the ELN group system structure.

Figure 2 is a schematic diagram of the master-slave game method.

Figure 3 is the net load curve of the ELN group in three modes.

Figure 4 is the change of the total objective function and the total cost of the network group during the iterative process.

Figure 5 is the change of net load fluctuation rate of network group in the iterative process.

Figure 6 shows the relationship between the total cost of the ELN group, the net load fluctuation rate and the power balance coefficient ρ.

Fig. 7 is a flow chart of a method for optimizing the control method of the energy Internet based on the master-slave game.

Specific implementation method

The present invention will be further described below with reference to the accompanying drawings.

Referring to Figures 1 to 7, a method for optimizing control of energy internet based on master-slave game a few days ago, includes the following steps:

S1: First build an energy local area network cluster model, initialize the system and obtain the parameters required for optimization, including the day-ahead forecast data of wind energy, solar energy and energy storage;

S2: Establish a master-slave game model, the network group control center is the leader, which sets the initial internal price, and each ELN subnet is the follower, making decisions based on the initial internal price, and calculating the corresponding optimal strategy;

S3: The network group control center integrates the strategy sets of each ELN subnet, and recalculates the internal price with the goal of maximizing the benefits of the network group control center, which is defined as the update internal price;

S4: Each ELN subnet makes a decision based on the updated internal price, and calculates the optimal strategy under which the internal price should be updated;

S5: When the game reaches the Stackelberg Equilibrium (SE) and the internal price is no longer updated, the final optimization set is output as the result of the previous optimization of the energy local area network group;

S6: If the game does not reach the Stackelberg equilibrium, return to step S2 to re-optimize according to the updated state information.

In the present invention, the energy internet environment is composed of multiple energy local area networks, the power supply side of each energy local area network individual is composed of photovoltaics, fans, gas turbines, energy storage and other power grids, and the demand side is composed of basic load and electric refrigeration. The heat load is supplied by gas turbines, heat collectors and gas boilers, the cooling load is supplied by gas turbines and electric refrigerators, and the overall ELN group is supplied or consumed by individual ELNs and external power grids.

Further, in the step S1, the system model includes the following components:

S1-1. Basic load model: ELN includes three types of loads, namely heating load, cooling load, and electrical load. The models are as follows:

Heat load: Provided by gas boilers, heat exchangers and collectors:

是ELNi中燃气锅炉的热功率；

是热交换器对外输出的热功率；

是ELNi中热负荷的总功率；in,

is the thermal power of the gas boiler in ELNi;

is the thermal power output by the heat exchanger;

is the total power of the thermal load in the ELNi;

The thermal power output by the gas boiler is related to the fuel usage and the heat production efficiency of the boiler;

是ELNi中燃气锅炉的最大热功率；η_GB是燃气锅炉的产热效率；V_GB,i为锅炉在一个时段内的燃气使用量；L_NG是天然气的热值，为9.7kWh/m³；in,

is the maximum thermal power of the gas boiler in ELNi; η _GB is the heat production efficiency of the gas boiler; V _GB,i is the gas consumption of the boiler in a period of time; LNG is the calorific value of natural gas, which is _9.7kWh /m ³ ;

From the fuel consumption of the gas turbine and the gas boiler, the total gas consumption V _SUM of the system can also be obtained as:

Among them, V _GT,i is the natural gas consumption of the ELNi gas turbine in period t;

As the controllable energy supply equipment in the ELN system of the present invention, the gas turbine not only provides electrical energy for the ELN, but also the heat carried by the high-temperature flue gas generated by the gas turbine can be recovered by the heat recovery device, and supplied to the heat load through the heat exchanger and the absorption refrigeration device. For heating and cooling the cooling load, the relationship between the output, heat and gas consumption of ELNi's gas turbines is as follows:

为t时段ELNi燃气轮机的发电功率；

为ELNi燃气轮机的最大发电功率；

为t时段ELNi燃气轮机的余热回收功率；η_c和η_r为ELNi燃气轮机的发电效率和余热回收效率；

是t时段ELNi燃气轮机的天然气消耗量；in,

is the power generation of the ELNi gas turbine in period t;

is the maximum generating power of the ELNi gas turbine;

is the waste heat recovery power of the ELNi gas turbine in period t; η _c and η _r are the power generation efficiency and waste heat recovery efficiency of the ELNi gas turbine;

is the natural gas consumption of the ELNi gas turbine in period t;

In addition, the power generation efficiency and waste heat recovery efficiency of the gas turbine are greatly affected by the unit load rate. According to the relevant literature, the relationship between the two and the unit load rate β can be found as follows:

0.25≤β≤1 (10)

为燃气轮机的额定发电效率；

为燃气轮机的额定余热回收效率；β为机组负载率；In the formula,

is the rated power generation efficiency of the gas turbine;

is the rated waste heat recovery efficiency of the gas turbine; β is the unit load rate;

3. The heat exchanger exchanges the part used for heating from the waste heat recovered from the gas turbine with water to obtain the output thermal power.

是

中分配出来用于制热的部分；η_HX为热交换器的换热效率；In the formula,

Yes

The part allocated for heating; η _HX is the heat exchange efficiency of the heat exchanger;

The cooling load is provided by absorption refrigeration units and electric chillers:

是ELNi中冷负荷的总功率；

是吸收式制冷装置对外输出的冷功率；

为ELNi的电制冷机制冷功率；in,

is the total power of the cooling load in the ELNi;

is the cooling power output by the absorption refrigeration device;

It is the cooling power of the electric refrigerator of ELNi;

Absorption refrigeration device: The absorption refrigeration device provides the part of the waste heat used for refrigeration to the heat exchanger in the device, so that the device can convert cold energy;

是

中分配出来用于制冷的部分；η_AC为吸收式制冷装置的制冷效率；In the formula,

Yes

The part allocated for refrigeration; η _AC is the refrigeration efficiency of the absorption refrigeration device;

The electric refrigerator is a special load used to generate the cooling load, and it can also be adjusted. When the cooling load is known, it belongs to the determined party. The output of the electric refrigerator of the i-th ELN should meet the following requirements:

为ELNi的电制冷机输入功率；η_EC为电制冷机的制冷效率；

为ELNi的电制冷机最大输入功率；in,

is the input power of the electric refrigerator of ELNi; η _EC is the cooling efficiency of the electric refrigerator;

It is the maximum input power of the electric refrigerator of ELNi;

Electric load: It is the only one that can be regulated among the three mentioned above. According to whether it is related to the other two types, the present invention mainly divides the electric load into two types: basic load and electric refrigerator. The basic load model is as follows:

For ELNi∈I, the base load is as follows:

是采用外部电网电价的情况下ELNi的基础负荷；λ_b代表从外部电网购买的电能价格；λ_s代表卖给外部电网的电能价格，r_b代表ELN群的内部购电价格，r_s代表ELN群的内部售电价格，电价应满足以下约束：in,

is the base load of ELNi in the case of using the external grid electricity price; λ _b represents the electricity price purchased from the external grid; λ _s represents the electricity price sold to the external grid, _rb represents the internal electricity purchase price of the ELN group, and _rs represents the ELN The price of electricity within the group should meet the following constraints:

λ _s ≤r _s <r _b ≤λ _b (17)

S1-2. Energy storage system model

The energy storage system reduces the net load of a single ELN and the entire ELN group through two controllable operations of charging and discharging. The SoC of each time period is related to the charge and discharge state and charge and discharge amount of the previous period. In the t time period , the working model of the energy storage system of the i-th ELN is as follows:

为t时段ELNi储能系统储存的能量；

t时段ELNi储能的剩余容量；Q_BES,i为ELNi储能系统的总容量；

为t时段ELNi储能系统的充电功率；

为t时段ELNi储能系统的放电功率；η_ch和η_dch为储能系统的充电效率和放电效率；In the formula,

is the energy stored by the ELNi energy storage system for the period t;

Remaining capacity of ELNi energy storage in period t; Q _BES,i is the total capacity of ELNi energy storage system;

is the charging power of the ELNi energy storage system in period t;

is the discharge power of the ELNi energy storage system in the t period; _ηch and _ηdch are the charging efficiency and discharging efficiency of the energy storage system;

Not only that, the energy storage system in the ELN also needs to constrain its own charge and discharge power and the state of the SoC, and at the same time meet the requirement that the state of the SoC remains unchanged before and after one day of operation:

和

分别表示ELNi储能系统的充放电功率；

和

为ELNi 储能系统的最大充放电功率；

与

代表ELNi储能系统的充放电状态，取 0或1，其中0表示不处于充放电状态，1表示处于充放电状态；

表示ELN i储能系统在t时段的充放电功率；

则为ELNi储能系统剩余容量的上下限；In the formula,

and

respectively represent the charging and discharging power of the ELNi energy storage system;

and

is the maximum charge and discharge power of the ELNi energy storage system;

and

Represents the charge and discharge state of the ELNi energy storage system, taking 0 or 1, where 0 means not in charge and discharge state, and 1 means in charge and discharge state;

Represents the charge and discharge power of the ELN i energy storage system in period t;

is the upper and lower limits of the remaining capacity of the ELNi energy storage system;

S1-3. Renewable Energy Model

The photovoltaic system has the maximum power point tracking function, which can be adjusted according to the light intensity and ambient temperature, and can track and output the maximum power of the time period. The output of ELNi's photovoltaic system is:

In the formula, P _pv,i is the average value of the power of all photovoltaic systems;

The fan system converts the mechanical energy of the wind into electrical energy, and the output power fluctuates with the change of the local average wind speed during the time period. The output of the fan system of ELNi is:

In the formula, P _WT,i is the average power of all fan systems;

S1-4.ELN Power Balance Model

Through the various components in the ELN on the supply side and the demand side, the electric power balance model of the i-th ELN is obtained:

为ELNi与网群的交换功率；

为输送线功率约束。in,

It is the exchange power between ELNi and the network group;

Power constraints for the transmission line.

Further, the establishment of the master-slave game model in the step S2 is as follows:

S2-1. The revenue model of a single ELN

的平方相关，电能交易的收益是在网群内部交易的收益，需要使用网群控制中心和各ELN子网之间的购电价格与售电价格；此外，当ELN联络线上的交换功率为正时，收益为

否则，收益为

燃气消耗的收益表示为燃气消耗成本的相反数；The function of the power balance is to reduce the net load of the ELN, here it is set to exchange power between the power balance and the ELN tie line

The square correlation of , the revenue of electric energy trading is the revenue of transactions within the network group, and the electricity purchase price and sales price between the network group control center and each ELN sub-network need to be used; in addition, when the exchange power on the ELN connection line is time, the profit is

Otherwise, the payoff is

The benefit of gas consumption is expressed as the inverse of the gas consumption cost;

From this, the utility function of ELN is obtained as:

相同；r_NG为燃气价格；In the formula, ρ is the correlation coefficient of the power balance utility, which is in the order of magnitude with

The same; _rNG is the gas price;

S2-2. The income model of the ELN group network group control center

Because the ELN in the system exchanges power through a bus, and exchanges energy with the external power grid through the bus, there are certain constraints on the power of the bus itself;

The network group control center is an artificially set organization in the system. As the leader of the master-slave game model, its optimization goal is to maximize its own interests. The profit function of the network group control center is set as follows:

和

为所有在t时段购电和售电的ELN交换功率的总和。in,

and

is the sum of the power exchanged by all ELNs that purchase and sell electricity in period t.

In step S3, the implementation process of the master-slave game is as follows:

The network group control center provides an initial internal price, first simulates the response of each ELN to the initial internal price, obtains the determined electric load size, and performs local optimization according to the utility function of each ELN to obtain all the optimal strategies of each ELN, and then obtains The interactive power u _grid of each ELN and the network group control center is calculated, and the network group control center optimizes with the goal of maximizing its own interests according to the response of each ELN to the initial internal electricity price and considering the influence of the internal electricity price on the electricity load. Thus, r _b and _rs are set more reasonably. The network group control center iterates with this as the initial value of the next optimization, and finally obtains the optimal r _b and _rs , and obtains the scheduling scheme of each ELN accordingly;

The resulting game model is as follows:

L={(I∪{GM}),{P _Load,i } _i∈I ,{R _b },{R _s },{U _i } _i∈I ,R} (31)

It contains the following components:

1) The set I of ELN is a follower and responds to the internal interactive electricity price set by the network group control center GM as the leader.

2) P _Load,i is the strategy set for ELN i to adjust the load, and the corresponding variable P _Load,i contains constraints

3) R _b and R _s are the strategy sets of GM, corresponding to the decision variables r _b and _rs of GM.

4) U _i is the profit function of ELNi, which can be represented by the variable P _Load,i and constraints (5)-(11), (14)-(15), (17)-(21) in the upper optimization.

5) R is the revenue function of GM, and its function is to obtain the profit of energy interaction within the ELN group and conduct transactions with the external power grid.

In step S5, the realization process of Stackelberg equilibrium is as follows:

Based on the master-slave game model in the above step S3, the network group control center determines the optimal price through the best response of each ELN, and each ELN determines the optimal value of its own decision on the basis of the price. The way to achieve the above scheme through master-slave game is Stackelberg Equilibrium (SE);

为In the Stackelberg game L defined by (31), when a strategy set

for

都有

即

为

的集合；而

Among them, for

have

which is

for

collection; and

与

以及SE时的负荷需求

唯一确定。Therefore, when the above conditions are met, that is, when the game reaches SE, the SE electricity price

and

and load requirements at SE

Only sure.

In step S6, it is judged whether the updated internal electricity price maximizes the benefit of the network group control center, and if the updated internal price does not change any more, the final optimization strategy set is output as the result of the previous optimization of the energy local area network group, otherwise, skip to step S2 Optimize again.

In order for those skilled in the art to better understand the benefits of the present invention, the applicant uses the ELN group net load characteristics and economics under three operating modes to analyze and compare. Take an ELN group consisting of 4 ELNs in a certain place as an example to verify the effectiveness of the proposed energy optimization management method, where ELN1 includes cooling and heating loads; ELN2 includes heating loads; ELN3 includes cooling loads and ELN4 without cooling and heating loads. to simulate various ELN structures. The three operating modes are as follows:

Mode 1: Before the optimization, the power generation equipment is fully loaded, and the energy storage does not participate in the dispatch mode.

Mode 2: The internal electricity price is the same as the external electricity price, and a non-cooperative game mode is adopted.

Mode 3: The present invention uses the real-time electricity price mode based on the Stackelberg master-slave game.

The statistics related to the net load characteristics are shown in Table 1. The RPR in the table is the fluctuation reduction rate before the relative optimization. Combining the net load curve Figure 3 and Table 1, it can be seen that mode 3 reduces the peak-to-valley difference by 82.44% and 29.22%, and the volatility decreases by 80.05% and 27.08% compared with modes 1 and 2, which reduces the ELN group. The power difference of the system improves the stability of the system. In addition, compared with Mode 1, the energy utilization rate of Modes 2 and 3 has been significantly improved, reaching or approaching the level of full energy utilization.

model Peak-to-valley difference/kW Volatility/kW RPR energy utilization 1 8701.1 1894.4 —— 0.9537 2 2158.1 518.4 0.7397 1 3 1527.6 378.0 0.7939 0.9996

Table 1

According to the economic calculation model, the statistical data related to the economy can be obtained as shown in Table 2.

Table 2

By analyzing the data in Table 2 and Figure 4, the following conclusions can be drawn:

1) Modes 2 and 3 are compared to Mode 1 before optimization. The energy utilization rate of wind turbines and photovoltaics has been significantly improved, the loss of abandoned wind and solar power has been reduced to close to zero, and the photovoltaic subsidies have increased by 4.85% and 4.82% respectively. Energy storage also participates in the adjustment process, and there are charge and discharge losses and electrical energy losses in the cost, and the operation and maintenance costs also increase slightly. In addition, the charging and discharging loss, power loss, and operation and maintenance cost of mode 3 are slightly higher than those of mode 2, indicating that the optimization process of the present invention has a higher degree of dispatching energy storage, and has little impact on economy.

2) The total costs of the three modes are all negative, indicating that the system is stable and profitable. Among them, compared with modes 1 and 2, the total revenue of mode 3 has increased by 290.06% and 123.31% respectively. This is because the total cost before optimization is relatively small compared with the large cost, which makes the improvement very obvious. At the same time, it is explained that the master-slave game used in the present invention is The optimization process greatly improves the economics of the ELN group.

3) Compared with Mode 1, the natural gas consumption of Modes 2 and 3 is increased by 11.01% and 4.52%, respectively, and the average unit power generation cost of gas turbines is increased by 19.92% and 15.11%, respectively, which indicates that the optimization process of Modes 2 and 3 will be in certain time periods. The power generation of the gas turbine unit is reduced, and other methods are used to supply cooling and heating, which increases the consumption of natural gas, but the mode 3 used in the present invention increases less, reflecting the present invention through the electricity price set by the network group control center. Optimization effectiveness.

网群的总成本为群内所有ELN成本的总和In the simulation process planned a few days ago, because the selected CPLEX solver has the limitation of solving conditions, the master-slave game cannot be written into the same solving process. Therefore, the game needs to be iterated to obtain the final optimization effect. If the power balance coefficient ρ=200 is set, the iterative process of the system is shown in Figure 5. Among them, the total objective function of the network group refers to the sum of the objective functions of all responding subjects, which is

The total cost of the network group is the sum of all ELN costs in the group

In the iterative process, with the increase of the number of iterations, the total objective function and the total cost of the ELN group decrease monotonically, and the sum of the variance of the ELN group's payload also shows a decreasing trend, and they all converge to a constant when the number of iterations is 12. , reaching the Stackelberg equilibrium. Under the operating state corresponding to this value, the ELN group can achieve the smallest overall net load fluctuation and the best economy at the same time.

In addition, the relationship between the power balance coefficient ρ in the ELN revenue model and the ELN group optimization results is also explored, and the obtained relationship is shown in Figure 6. Among them, when ρ <= 1500, the total cost of the ELN group fluctuates to a certain extent with the increase of ρ but is basically unchanged, and when ρ > 1500, the total cost of the ELN group will increase sharply with the increase of ρ. On the other hand, the net load volatility of the ELN group takes the minimum value around ρ=200. When ρ<200, it decreases with ρ and when ρ>200, it increases with ρ, which will lead to an increase in the net load volatility. To sum up, when ρ is about 200, the system can reach the optimal operating state of economy and stability.

Facing the ELN group system considering the interactive response, the present invention models various types of equipment involving multiple energy sources in the ELN, and on this basis designs an internal real-time electricity price model based on the master-slave game for the interactive response between the ELNs . The present invention completes the energy optimization management method of the ELN group on the basis of these models, so as to find the best economic operation strategy of the ELN group.

The method used coordinates the output plan of controllable sources, loads, and storage in each ELN and changes in the internal real-time electricity price, and at the same time promotes the overall ELN group closer to power balance, which significantly improves the stability of each ELN and ELN group system. Compared with the system model with constant electricity price, the proposed real-time electricity price system model can greatly reduce the consumption cost of natural gas at the expense of slightly increasing the loss and operation and maintenance costs in the dispatching process, so that the overall economical efficiency can be obtained. a large improvement.

In the description of this specification, case comparison and analysis, random scene analysis, etc. are used to describe the specific features, structures and benefits of the present invention. In this specification, the schematic representations of the present invention are not necessarily directed to the same embodiment or example, and those skilled in the art may combine and combine different embodiments or examples described in this specification. In addition, the content described in the embodiments of this specification is only an enumeration of the realization forms of the inventive concept, and the protection scope of the present invention should not be regarded as limited to the specific forms stated in the implementation cases, and the protection scope of the present invention also includes the field of Equivalent technical means that can be conceived by a skilled person according to the inventive concept.

Claims (5)

1. An energy internet optimization control method based on a master-slave game is characterized by comprising the following steps:

s1: firstly, an energy local area network group model is constructed, a system is initialized, and parameters required by optimization, including day-ahead prediction data of wind energy, light energy and stored energy, are obtained;

s2: establishing a master-slave game model, taking a network group control center as a leader, setting an initial internal price by the leader, taking each ELN sub-network as a follower, making a decision according to the initial internal price, and calculating a corresponding optimal strategy;

s3: the network group control center integrates the strategy set of each ELN sub-network, and calculates the internal price again by taking the benefit maximization of the network group control center as a target, and the internal price is defined as the updated internal price;

s4: each ELN subnet carries out decision making according to the updated internal price, and an optimal strategy corresponding to the updated internal price is calculated;

s5: when the game reaches Stackelberg equilibrium SE and the internal price is not updated any more, outputting a final optimization set as a day-ahead optimization result of the energy local area network group;

s6: if the game does not reach the Stackelberg balance, returning to the step S2 to optimize again according to the updated state information;

in step S1, the system model includes the following components:

s1-1, basic load model: the ELN contains three types of loads, i.e., heat load, cold load, electrical load, and the model is as follows:

heat load: provided by a gas boiler, a heat exchanger and a heat collector:

wherein,

is the thermal power of the gas boiler in the ELNi;

is the heat power output by the heat exchanger;

is the total power of the thermal load in the ELNi;

the thermal power output by the gas boiler is related to the fuel consumption and the heat production efficiency of the boiler;

wherein,

η is the maximum thermal power of the gas boiler in ELNi_GBIs the heat production efficiency of the gas boiler; v_GB,iThe gas consumption of the boiler in a period of time; l is_NGIs the heat value of natural gas and is 9.7kWh/m³；

The total gas consumption V of the system can be obtained according to the fuel consumption of the gas turbine and the gas boiler_SUMComprises the following steps:

wherein, V_GT,iIs the natural gas consumption of the ELNi gas turbine during the period t;

the gas turbine is used as a main controllable energy supply device in the ELN system, not only provides electric energy for the ELN, but also recovers heat carried by high-temperature flue gas generated by the gas turbine by a heat recovery device, supplies heat for a heat load and supplies cold for a cold load through a heat exchanger and an absorption refrigeration device, and the relationship among the output, the heat and the gas consumption of the ELNi gas turbine is as follows:

wherein,

generating power of the ELNi gas turbine for a period t;

the maximum power generation power of the ELNi gas turbine;

η waste heat recovery power of ELNi gas turbine in t period_cAnd η_rThe generating efficiency and the waste heat recovery efficiency of the ELNi gas turbine are achieved;

is the natural gas consumption of the ELNi gas turbine during the period t;

further, the power generation efficiency and the heat recovery efficiency of the gas turbine are greatly affected by the unit load factor, and the relationship between these two and the unit load factor β is as follows:

0.25≤β≤1 (10)

in the formula,

the rated power generation efficiency of the gas turbine;

β is the unit load factor;

the heat exchanger exchanges a part for heating in the waste heat recovered from the gas turbine with water to obtain output thermal power;

in the formula,

is that

A portion assigned for heating η_HXThe heat exchange efficiency of the heat exchanger;

the cold load is provided by the absorption refrigeration device and the electric refrigerator:

wherein,

is the total power of the cold load in the ELNi;

the cold power output by the absorption refrigeration device is the cold power output by the absorption refrigeration device;

the refrigeration power of the electric refrigerator is ELNi;

the absorption refrigeration device provides a part of the waste heat for refrigeration to a heat exchanger in the device, so that the device converts cold energy;

in the formula,

is that

Middle portion allocated for cooling η_ACThe refrigeration efficiency of the absorption refrigeration device;

the electric refrigerator is a special load for generating a cooling load, and can be adjusted, and belongs to a determined party under the condition that the cooling load is known, and the output of the electric refrigerator of the ith ELN meets the following requirements:

wherein,

input power of electric refrigerator being ELNi η_ECThe refrigeration efficiency of the electric refrigerator;

maximum input power of the electric refrigerator of ELNi;

electrical loading: the electric load is mainly divided into two types of basic load and electric refrigerator according to whether the electric load is related to other two types, and the basic load model is as follows:

for ELNi ∈ I, the base load is as follows:

wherein,

the method is the basic load of ELNi under the condition of adopting the electricity price of an external power grid; lambda [ alpha ]_bRepresents the price of electrical energy purchased from an external power grid; lambda [ alpha ]_sRepresenting the price of electricity sold to an external grid, r_bRepresenting the internal purchase price, r, of the ELN group_sRepresenting the internal electricity selling price of the ELN group, the electricity price should satisfy the following constraint:

λ_s≤r_s＜r_b≤λ_b(17)

s1-2. energy storage system model

The energy storage system reduces the net load of a single ELN and the whole ELN group through two controllable operations of charging and discharging, the SoC of each time period is related to the charging and discharging state and the charging and discharging amount of the previous time period, and in the t time period, the working model of the energy storage system of the ith ELN is as follows:

in the formula,

the energy stored by the ELNi energy storage system is t time period;

surplus capacity of time interval ELNi energy storage; q_BES,iThe total capacity of the ELNi energy storage system;

the charging power of the ELNi energy storage system is t time period;

η discharge power of ELNi energy storage system in t period_chAnd η_dchThe charging efficiency and the discharging efficiency of the energy storage system are obtained;

moreover, the energy storage system in the ELN still needs to constrain the charging and discharging power of itself and the state of SoC, and simultaneously satisfies the requirement that the SoC state is not changed before and after operating one day:

in the formula,

and

respectively representing the charge and discharge power of the ELNi energy storage system;

and

the maximum charge-discharge power of the ELNi energy storage system;

and

representing the charge-discharge state of the ELNi energy storage system, and taking 0 or 1, wherein 0 represents that the ELNi energy storage system is not in the charge-discharge state, and 1 represents that the ELNi energy storage system is in the charge-discharge state;

representing the charge and discharge power of the ELNi energy storage system in a t period;

the residual capacity of the ELNi energy storage system is the upper and lower limits;

s1-3 renewable energy model

The photovoltaic system has a maximum power point tracking function, can be adjusted according to the illumination intensity and the ambient temperature, tracks and outputs the maximum power in the time period, and the output of the photovoltaic system of the ELNi is as follows:

in the formula, P_PV,iAverage value of all photovoltaic system power;

the fan system converts the mechanical energy of wind into electric energy, the output power fluctuates along with the change of the local average wind speed in the time period, and the output of the ELNi fan system is as follows:

in the formula, P_WT,iThe average value of all fan system power is obtained;

s1-4.ELN Power balance model

Obtaining an electric power balance model of an ith ELN through various components of the ELN on a power supply side and a demand side:

wherein,

the exchange power of ELNi and the network group;

is a transmission line power constraint.

2. The energy internet optimization control method based on the master-slave game as claimed in claim 1, wherein the master-slave game model in step S2 is established as follows:

s2-1. yield model of Single ELN

The role of power balancing is to reduce the net load of the ELN, where power balancing is set to exchange power with the ELN link

The profit of the electric energy transaction is the profit of the transaction within the grid, the electricity purchase price and the electricity sale price between the grid control center and each ELN sub-network are used, and the profit is the profit when the exchange power on the ELN connection is positive

Otherwise, the benefit is

The benefit of gas consumption is expressed as the inverse of the gas consumption cost;

thus, the utility function for ELN is obtained as:

where ρ is workCorrelation coefficient of rate balancing effect, order of magnitude and

the same; r is_NGIs the gas price;

s2-2. revenue model of ELN cluster control center

Since the ELN in the system performs power interaction through one bus and performs energy interaction with an external power grid through the bus, certain constraint is provided for the power of the bus;

the network group control center is a mechanism manually set in the system, is used as a leader of a master-slave game model, and has the optimization goal of maximizing the benefits of the network group control center, and the revenue function of the network group control center set for the purpose is as follows:

wherein,

and

the sum of power is exchanged for all ELNs purchasing and selling electricity during the time period t.

3. The energy internet optimization control method based on the master-slave game as claimed in claim 2, wherein in step S3, the master-slave game is implemented as follows:

the network group control center provides an initial internal price, and simulates each ELN to make a sound to the initial internal priceAnd obtaining the determined electric load size, and carrying out local optimization according to the utility function of each ELN to obtain all optimal strategies of each ELN, thereby obtaining the interactive power u of each ELN and the network group control center_grid(ii) a The network group control center optimizes the optimization with the maximization of the benefit as the target according to the reaction of each ELN to the initial internal electricity price and the influence of the internal electricity price on the electric load_bAnd r_sThe network group control center uses the value as the initial value of the next optimization to carry out iteration, and finally the optimal r is obtained_bAnd r_sAnd obtaining a scheduling scheme of each ELN according to the scheduling scheme;

the game model thus formed is as follows:

L＝{(I∪{GM}),{P_Load,i}_i∈I,{R_b},{R_s},{U_i}_i∈I,R} (31)

the composition comprises the following components:

1) the set I of the ELNs is a follower, and a response network group control center GM is used as an internal interaction electricity price set by the leader;

2)P_Load,iis a strategy set of ELNi adjusting load and the corresponding variable P_Load,iContaining constraints

3)R_b、R_sIs a policy set of GM, corresponding to decision variable r of GM_bAnd r_s；

4)U_iFor the yield function of ELNi, the upper layer optimization can be performed by the variable P_Load,iAnd constraints (5) - (11), (14) - (15), (17) - (21);

5) and R is a yield function of the GM, and the effect is to obtain the profit of energy interaction in the ELN group and trade with an external power grid.

4. The energy internet optimization control method based on the master-slave game as claimed in claim 3, wherein in step S5, the Stackelberg balancing is implemented as follows:

based on the master-slave game model in the step S3, the cluster control center determines the optimal price through the optimal reaction of each ELN, each ELN determines the optimal value of its own decision on the basis of the price, and the mode of achieving the above scheme through the master-slave game is the Stackelberg balance;

in the Stackelberg game L defined in (31), when a policy set is in use

Is composed of

Wherein, for

Are all provided with

Namely, it is

Is composed of

A set of (a); while

Therefore, when the above condition is satisfied, namely the game reaches SE, the SE electricity price

And

and load demand at SE

And (4) uniquely determining.

5. The energy Internet optimization control method based on the master-slave game as claimed in any one of claims 1 to 4, wherein in step S6, it is determined whether the updated internal price of electricity maximizes the profit of the network group control center, if the updated internal price is not changed any more, the final optimization strategy set is output as the optimization result of the energy local area network group before the day, otherwise, the step S2 is skipped to perform the optimization again.

Images