CN112636330A

CN112636330A - Comprehensive energy system peak regulation method based on user contribution behaviors

Info

Publication number: CN112636330A
Application number: CN202011400192.7A
Authority: CN
Inventors: 朱大风; 杨博; 刘琦; 陈彩莲; 关新平
Original assignee: Shanghai Jiaotong University
Current assignee: Shanghai Jiaotong University
Priority date: 2020-12-02
Filing date: 2020-12-02
Publication date: 2021-04-09
Anticipated expiration: 2040-12-02
Also published as: CN112636330B

Abstract

The invention discloses a comprehensive energy system peak shaving method based on user contribution behaviors, which relates to the field of industrial energy management and comprises the following steps of 1, establishing a multi-energy storage, supply, transaction and supply and demand balance model; step 2, establishing a Lagrange function of energy demand management; step 3, establishing an energy scheduling optimization distributed algorithm; and 4, carrying out energy distribution management of the park. The step 1 comprises the following steps: establishing a Battery B_k(t) and hot water tank energy storage W_k(t) a dynamic model; establishing a Combined Heat and Power (CHP) Unit Electricity Generation E_kCHP(t) and Heat H_kCHP(t) and boiler heat generation H_kb(t) a model; establishing an energy transaction model with an external energy company; establishing a multi-energy supply and demand balance model of electric energy, natural gas and heat energy; energy demand management of industrial users is structured as a non-cooperative game. The invention solves the problem of scheduling and optimizing multiple energy sources in the industrial park, protects the privacy of users, optimizes energy scheduling and improvesEnergy efficiency and reduced energy consumption.

Description

Comprehensive energy system peak regulation method based on user contribution behaviors

Technical Field

The invention relates to the field of industrial energy, in particular to a comprehensive energy system peak shaving method based on user contribution behaviors.

Background

With the continuous expansion of industrial production scale and the rapid increase of demand, the problems of low efficiency and high cost of industrial parks need to be solved urgently. There have been many studies on the multi-energy management of industrial parks. However, most of the existing research focuses on system optimization, and there is no consideration for a real-time multi-energy user model, that is, the benefit and social welfare of a user are improved by adjusting the time and space multi-energy load of the user.

Second, by encouraging users to participate in energy management, some users shift the load during peak hours to off-peak hours, which not only reduces the cost of other users, but also relieves power supply pressure. However, in an actual industrial park, users schedule energy sources for maximizing their profits, and particularly in the case of high energy consumption load, energy sources of different types are coupled with each other, which may affect energy transactions between industrial users and park managers. Meanwhile, existing research does not consider that some users help other users by shifting the need for resilience in situations where the total energy is limited.

The peak load transfer of single energy electric energy is only considered in most of the peak regulation of the existing industrial park. The existing implementation scheme most similar to the method considers the scheduling optimization of the industrial park of the peak regulation requirement of the power grid, and ignores the cost optimization and the scheduling optimization of other types of energy caused by coordination and complementation of various energy sources. The photovoltaic unit, the battery energy storage system and the industrial load are only used as scheduling units participating in peak shaving of the power grid, the types of the scheduling units are few, and particularly common combined heat and power generation units in industrial parks are not considered. Meanwhile, the adopted centralized optimization method does not consider the user privacy, and neglects the rolling or random optimization problem of the dynamic energy storage model.

Accordingly, those skilled in the art are devoted to developing an integrated energy system peak shaving method based on user contribution behavior to improve the utilization of distributed energy and the efficiency of a multi-energy network.

Disclosure of Invention

In view of the above-mentioned drawbacks of the prior art, the technical problem to be solved by the present invention is: how to solve the problem of scheduling and optimizing multiple energy sources of an industrial park on the premise of protecting the privacy of users, how to consider the coordination and complementation of multiple energy sources and the requirement transfer of the multiple energy sources across space-time scales, how to fully mobilize the users to actively participate in energy scheduling, and how to deal with various complex conditions occurring in the control process.

In order to achieve the above object, the present invention provides a peak shaving method of an integrated energy system based on user contribution behavior, comprising the following steps:

step 1, establishing a multifunctional storage, supply, transaction and supply and demand balance model;

the step 1 comprises the following steps:

step 1.1, set up Battery B_k(t) and hot water tank energy storage W_k(t) a dynamic model;

step 1.2, establishing a Combined Heat and Power (CHP) unit to generate electricity E_kCHP(t) and Heat H_kCHP(t) and boiler heat generation H_kb(t) a model;

step 1.3, establishing an energy transaction model with an external energy company;

step 1.4, establishing a multi-energy supply and demand balance model of electric energy, natural gas and heat energy;

step 1.5, constructing the energy demand management of industrial users into a non-cooperative game;

step 2, establishing a Lagrange function of energy demand management;

step 3, establishing an energy scheduling optimization distributed algorithm;

and 4, carrying out energy distribution management of the park.

Further, the step 1.1 comprises:

establishing a Battery B_k(t) and hot water tank energy storage W_k(t) dynamic model, as follows:

in the formula eta_dkeAnd η_dkchAll represent energy efficiency;

battery charging C_ke(t) and discharge D_ke(t) do not occur simultaneously, and the hot water tank stores heat energy C_kh(t) and transporting the heat quantity D outwards_kh(t) do not occur simultaneously, they exist in the following relationship:

the battery and the hot water tank are constrained by

B_k，min≤B_k(t)≤B_k，max#

W_k，min≤W_k(t)≤W_k，max#

The battery and the hot water tank are constrained in charging and discharging energy

0≤C_ke(t)≤C_ke，max，0≤D_ke(t)≤D_ke，max#

0≤C_kh(t)≤C_kh，max，0≤D_kh(t)≤D_kh，max#

Combining that the battery charge and discharge do not simultaneously occur and charge and discharge constraint, and cannot exceed the battery capacity limit, combining that the hot water tank heat charge and discharge do not simultaneously occur and heat charge and discharge constraint, and cannot exceed the hot water tank capacity limit, can obtain the constraint:

0≤C_ke(t)≤min[B_k，max-B_k(t)，C_ke，max]#

0≤D_ke(t)≤min[B_k(t)-B_k，min，D_ke，max]#

0≤C_kh(t)≤min[W_k，max-W_k(t)，C_kh，max]#

0≤D_kh(t)≤min[W_k(t)-W_k，min，D_kh，max]#

further, the step 1.2 comprises:

establishing a Combined Heat and Power (CHP) Unit Electricity Generation E_kCHP(t) and Heat H_kCHP(t) and boiler heat generation H_kb(t) model:

E_kCHP(t)＝η_kpg G_kCHP(t)，H_kCHP(t)＝η_khg G_kCHP(t)#

H_kb(t)＝η_kbgG_kb(t)#

the natural gas consumed by the cogeneration unit and the boiler is G respectively_kCHP(t) and G_kb(t) their capacity constraints are:

0≤E_kCHP(t)≤E_kCHP，max，0≤H_kCHP(t)≤H_kCHP，max#

0≤H_kb(t)≤H_kb，max#

further, the step 1.3 includes:

establishing an energy transaction model with an external energy company, mainly purchasing electricity E (t) and heat G (t) from the energy company, and selling electricity E from the energy company when the local renewable energy is more_o(t)：

0≤E(t)≤E_max，0≤G(t)≤G_max，0≤E_o(t)≤E_o，max#

Further, the step 1.4 includes:

establishing a multi-energy supply and demand balance model of electric energy, natural gas and heat energy, wherein the electric energy balance relates to the cogeneration unit, the battery, the renewable energy R (t), the trade with the power grid and the demand E_tot(t); the natural gas balance relates to the cogeneration unit and the boilerFurnace and demand G_tot(t); the heat energy balance relates to the cogeneration unit, the boiler, the hot water tank and the demand H_tot(t); the industrial park is supplied with energy by an energy hub, the energy hub K in the industrial park consists of a battery, renewable energy, a cogeneration unit, a boiler and a hot water tank, and K belongs to K; the supply and demand balance model of the industrial park is as follows:

the set of various energies is denoted by X, i.e., X ═ { E, G, H }, the industrial park supplies the total available energy X to the users_tot(t) is represented by

Simultaneous user i energy X_i(t) the following constraints need to be satisfied

X_i，min(t)≤X_i(t)≤X_i，max(t)#

Further, the step 1.5 includes:

building energy demand management for industrial users as a non-cooperative game, wherein each user is dedicated to maximizing a respective benefit; benefit U of the user_id(t) consists of energy use satisfaction and energy use cost:

p_px(t) means forPrice for energy sold to users in a park, a_ixAnd b_ixA user's energy use satisfaction factor;

benefit U of park manager_d(t) consists of the energy benefit sold to the user and the cost of trading with external energy companies:

p_o(t) shows the electricity rate of the park for the electric power company, p_e(t) and p_g(t) prices for electricity and natural gas purchased from external energy companies in the park, respectively;

social welfare U_td(t) consists of the sum of the revenues of all industrial users and the revenues of the campus manager:

further, the step 2 comprises:

the Lagrange multiplier is introduced to obtain the Lagrange function, lambda, of the user_ix，min(t)、λ_ix，max(t) and μ_ix(t) is the Lagrangian multiplier:

further, the energy scheduling optimization distributed algorithm in the step 3 realizes optimization, communication and contribution functions.

Furthermore, the optimization function means that the energy demand management problem is separately solved according to the energy utilization limit of each user under the condition that the overload is not considered by the user; the communication function refers to that a user communicates with others to process common constraints, namely overload control; the contribution function means that the user can reduce the energy stress of the overload period by changing the elastic electricity demand.

Further, the step 4 comprises: after the optimal energy demand of the user is obtained, the social welfare maximization problem is solved, the mutual coupling coordination and complementation of multiple energy sources are utilized, and the optimal energy dispatching of the electric heat of the industrial park is obtained by adopting a linear programming method.

Compared with the prior art, the invention has the beneficial technical effects that:

(1) the invention fully considers the coordination and complementation of various energy sources and the requirement transfer of the multiple energy sources across the space-time scale.

(2) The invention adopts a distributed algorithm based on a consistency method to solve the problem of peak shaving of the industrial park and protect the privacy of users.

(3) The invention fully mobilizes the user to actively participate in energy scheduling, and improves the income of other users under the condition of ensuring that the income of part of users is not changed.

(4) The invention adopts a control method of rolling optimization to correct various complex conditions in time in the control process.

The conception, the specific structure and the technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, the features and the effects of the present invention.

Drawings

FIG. 1 is a flow chart of a preferred embodiment of the present invention;

FIG. 2 is a multi-energy management system framework for an industrial park in accordance with a preferred embodiment of the present invention;

FIG. 3 is a schematic diagram of an energy scheduling optimization distributed algorithm;

FIG. 4 is a flow diagram of an energy scheduling optimization distributed algorithm including optimization, communication, and contribution functions;

FIG. 5 is a schematic diagram of the electricity prices of a utility company in an embodiment of the invention;

FIG. 6 is a simulation result obtained by the first method according to the embodiment of the present invention;

FIG. 7 is a simulation result obtained by the second method according to an embodiment of the present invention;

fig. 8 is a simulation result obtained by the third method in the embodiment of the present invention.

Detailed Description

The technical contents of the preferred embodiments of the present invention will be more clearly and easily understood by referring to the drawings attached to the specification. The present invention may be embodied in many different forms of embodiments and the scope of the invention is not limited to the embodiments set forth herein.

Fig. 1 shows a flow chart of the method of the present invention. Figure 2 is a multi-energy management system framework for an industrial park.

The invention provides a comprehensive energy system peak regulation method based on user contribution behaviors, which comprises the following steps of:

step 1, establishing a system model. The method comprises the following substeps:

step 1.1, set up Battery B_k(t) and hot water tank energy storage W_k(t) dynamic model, as follows:

in the formula eta_dkeAnd η_dkhBoth represent energy efficiency.

the battery and the hot water tank are constrained in capacity

B_k，min≤B_k(t)≤B_k，max# (4)

W_k，min≤W_k(t)≤W_k，max# (5)

The charging and discharging energy of the battery and the hot water tank are restricted as

0≤C_ke(t)≤C_ke，max，0≤D_ke(t)≤D_ke，max# (6)

0≤C_kh(t)≤C_kh，max，0≤D_kh(t)≤D_kh，max# (7)

In combination with the non-simultaneous occurrence of battery charging and discharging and the charging and discharging constraints, and the inability to exceed the battery capacity limit, the following constraints can be obtained. The method is also suitable for heat charging and discharging of the hot water tank.

0≤C_ke(t)≤min[B_k，max-B_k(t)，C_ke，max]# (8)

0≤D_ke(t)≤min[B_k(t)-B_k，min，D_ke，max]# (9)

0≤C_kh(t)≤min[W_k，max-W_k(t)，C_kh，max]# (10)

0≤D_kh(t)≤min[W_k(t)-W_k，min，D_kh，max]# (11)

Step 1.2, establishing a Combined Heat and Power (CHP) unit to generate electricity E_kCHP(t) and Heat H_kCHP(t) and boiler heat generation H_kb(t) model:

E_kCHP(t)＝η_kpg G_kCHP(t)，H_kCHP(t)＝η_khg G_kCHP(t)# (12)

H_kb(t)＝η_kbg G_kb(t)# (13)

0≤E_kCHP(t)≤E_kCHP，max，0≤H_kCHP(t)≤H_kCHP，max# (14)

0≤H_kb(t)≤H_kb，max# (15)

step 1.3, establishing an energy transaction model with an external energy company, mainly purchasing electricity E (t) and heat G (t) from the energy company, and selling electricity E from the energy company when the local renewable energy sources are more_o(t)：

0≤E(t)≤E_max，0≤G(t)≤G_max，0≤E_o(t)≤E_o，max# (16)

Step 1.4, establishing a multi-energy supply and demand balance model of electric energy, natural gas and heat energy, wherein the electric energy balance relates to a cogeneration unit, a battery, renewable energy R (t), and the trade and demand E with a power grid_tot(t); natural gas balance involving cogeneration unit, boiler and demand G_tot(t); the heat energy balance relates to a combined heat and power generation unit, a boiler, a hot water tank and a demand H_tot(t); the industrial park is supplied with energy by an energy hub, the energy hub K in the industrial park is composed of a battery, renewable energy, a cogeneration unit, a boiler and a hot water tank, and K belongs to K. The supply and demand balance model of the industrial park is as follows;

X_i，min(t)≤X_i(t)≤X_i，max(t)# (18)

Step 1.5, the energy demand management of the industrial users is constructed as a non-cooperative game, where each user is working to maximize its respective benefit. Benefit U of the user_id(t) is composed of energy consumption satisfaction and energy consumption cost

p_px(t) represents the price at which the industrial park sells energy to the consumer, a_ixAnd b_ixThe user satisfaction coefficient. Benefit U of park manager_d(t) consisting of the energy benefit sold to the user and the cost of the transaction with the outside energy company

p_o(t) shows the electricity rate of the park for the electric power company, p_e(t) and p_g(t) shows prices of electricity and natural gas purchased from external energy companies, respectively, in the park. Social welfare U_td(t) is composed of the sum of the profits of all the industrial users and the profits of the park manager

And 2, establishing a Lagrange function for energy demand management.

The present invention describes a solution to the energy demand management problem that is based on a rolling optimization of the T time step, rather than a single time step. From the current moment, optimal control is obtained within a fixed, finite time frame. In optimal control over a limited time domain, only the first action is implemented as the current control action, and then the process is repeated over a limited time range from the next and implemented with new available information updates. The implementation of the update is a result of the dynamic environment. Therefore, the algorithm is suitable for dynamic environments with uncertain supply and demand. Meanwhile, after the optimal energy demand of each user is obtained, the energy distribution management problem is solved by using a linear programming method, so that the social welfare is maximized.

The energy demand management problem of the industrial user meets the conditions of Karush-Kuhn-Tucker (KKT), and the Lagrangian multiplier is introduced to obtain the Lagrangian function, lambda, of the user_ix，min(t)、λ_ix，max(t) and μ_ix(t) is the Lagrangian multiplier:

according to the KKT condition, it is obtained

The optimal solution can be expressed as

And 3, establishing an energy scheduling optimization distributed algorithm.

In order to solve the problem of energy scheduling optimization and protect user privacy, the technical scheme provided by the invention provides a distributed algorithm. In such a distributed architecture, users need to communicate with others to exchange their public information, namely X_i(t) and μ_ix(t) of (d). The user then executes the energy demand management algorithm with the contributing behavior, which is the first algorithm, as shown in fig. 3, 4. At algorithm initialization, the binary term of the contribution flag (CB) is set to 1. The algorithm implements the optimization, communication, and contribution functions, as described in detail below.

The optimization function means that under the condition that overload is not considered by users, the energy demand management problem is independently solved according to the energy utilization limit of each user(see step 2 of the algorithm of fig. 3). In this way, the present invention creates an initial distributed structure and demonstrates the ability of each user to make individual initial decisions. The best solution obtained is X_i(t) and μ_ix(t) of (d). The execution of this function is referred to as a first method.

The communication function means that the user communicates with others to handle a common constraint, i.e. overload control. The first algorithm terminates in nash equalization when there is no overload condition. However, if any overload condition occurs, the binary term of the overload flag (OL) is set to 1, and the energy consumption of the user needs to be suppressed to satisfy the overload constraint. In order to obtain the optimal solution, the invention needs to combine mu_ixAll values of (t) converge to a global value μ_x(t)) to suppress energy consumption by the current user, as discussed in more detail below.

First, the energy loss due to overload is Δ X_i(t) represents. Each user then iteratively interacts its public information (X) with other neighboring users using a consistency approach_i(t) and μ_ix(t)). User i then adds μ_ix(t) mu adjacent thereto_jx(t) the total weighted difference between (t) and the weighted term of Δ X (t) to update its μ_ix(t) (see step 12 of the algorithm of fig. 3). k is a radical of_ixAnd l_ixIs a weight coefficient, phi_ixIs a neighbor set of user i, and after convergence is achieved, user i updates its maximum available energy X_i，max(t) (see step 14 of the algorithm of FIG. 3), user i needs to be informed of the new maximum available energy X_i，max(t) rescheduling its energy consumption. In the first algorithm, e and ξ are small positive values. The performance of the optimization and communication functions is referred to as a second method.

The contribution function means that the user can reduce the energy stress of the overload period by changing the elastic electricity demand. These shifted demands of the user may be at other times T_nCompensation is performed at the same electricity rate. By shifting the demand for resilience of some users away from the overload period, the total energy available for the more demanding users during the overload period will be greater, i.e. the more demanding users may be stressed for the time T_oIncrease energy demand at high electricity priceThe energy demand is reduced.

The specific contribution process of the first algorithm is as follows: first, user i gives that he can be at T_nDuring which the transferable energy X is supplied_itI.e. it is at T_nThe difference between the maximum available energy and the actual energy used during the period (see step 21 of the algorithm of fig. 3). If the transferable energy is greater than its threshold X_it，thThen user i reduces its maximum available energy X for the current time period based on its transferable energy_i，max(t) to contribute to others. Thereafter, user i will be at time T_nCompensating for this part of the transferred energy X_it. Finally, after the overload problem is resolved and the contribution function is implemented, the first algorithm terminates. Performing the above three functions is referred to as a third method.

And 4, energy distribution management of the industrial park.

After the optimal energy demand of the user is obtained, the social welfare maximization problem can be solved accordingly, the mutual coupling coordination and complementation of multiple energy sources are utilized, and the optimal energy dispatching of the electric heat of the industrial park can be obtained by solving by a linear programming method.

In the present invention, consider a system comprising an energy hub and three factories, and the price of electricity is provided by power saving companies in Jiangsu of the national grid, as shown in FIG. 5. The simulation results obtained by applying the three methods proposed by the present invention are shown in fig. 6-8. The total duration in the simulation was 24 hours. The energy efficiency parameters in the present invention are η_cke＝η_dke＝η_ckh＝η_dkh＝98％，η_kpg＝η_khg＝35％，η_kbg80%. Other parameters are, respectively, the gas price p_g(t)＝0.3yuan/kWh，B_k，max＝4MWh，C_ke，max＝D_ke，max＝1MWh，W_k，max＝ 4MWh，C_kh，max＝D_kh，max＝1MWh。

Considering that electricity prices are dynamically changing, the present invention will focus on analyzing the power demand of the plant. The total power requirement per plant per day is 26.9 MWh. The daily working time of the factory 1 is 8:00-24:00, and the working time of the factories 2 and 3 is 8:00-21: 00. While each plant requires at least 0.6MWh of electrical energy for any one period of time (i.e., one hour) to maintain the necessary energy supply, such as lighting and keeping room temperature constant. Fig. 6 shows that all plants buy as much electricity as possible at a price of 0.6 yuan. Wherein the first method does not take overload control into account and thus overload occurs during the 12:00-15:00 time period, as shown by the black shading in the figure. Fig. 7 shows that the occurrence of overload is avoided under the communication function of the second method, but under this method, the factories 2 and 3 purchase more electric power at a high price of electricity, resulting in high energy costs. Fig. 8 shows that in the third method, the plant 1 shifts its load from the peak energy usage period 12:00-15:00 to the valley energy usage period 21:00-24:00, relieving the power demand of the plants 2 and 3 during the peak energy usage period, so that the plants 2 and 3 can increase their own power usage during the peak energy usage period, thereby reducing their energy usage at high electricity rates, i.e., reducing the power costs of the plants 2 and 3. By means of method 3, the energy consumption cost of plant 1 is unchanged, and the electric energy cost of plants 2 and 3 is reduced from 17745 yuan to 16807 yuan.

Fig. 5 shows that the electricity price is higher during the period of 8:00-23:00, and the industrial park generates electricity by utilizing cogeneration equipment, so that the electricity purchasing quantity is reduced, and heat is supplied to the factory. The simulation results show that the battery is charged at low electricity prices and low demand and discharged at high electricity prices and high demand. Cogeneration plants and boilers supply the heat requirements of the plant together. Therefore, the method provided by the invention flexibly realizes the transfer of various energy requirements, the transaction with electric power and gas companies and the energy supply of energy storage equipment.

The invention provides a distributed algorithm based on a consistency method and rolling optimization, aiming at the problem of energy demand management of an industrial park. The rolling horizon control is generally applied in dynamic management environments, such as supply and demand uncertainty. Unlike the general non-cooperative method, the invention adds a contribution behavior to selfish users, namely, the non-cooperative method and cooperative assistance are adopted to minimize the energy cost of each user. In addition, the present invention maximizes social welfare after obtaining the optimal energy demand of each user. The main contributions of the present invention are as follows.

A multi-energy management framework for industrial parks is proposed, where park managers sell electricity, heat and gas to industrial customers and transact with external energy companies.

A distributed algorithm based on a consistency method and rolling optimization is proposed to solve the energy demand management problem to protect user privacy.

Give contribution behavior to selfish users. Some users shift their elastic energy demand from peak hours to off-peak hours, which helps other users to reduce costs, especially in overload situations.

From simulation analysis and results, the proposed algorithm can achieve a better compromise between energy costs, energy trading and energy storage. Meanwhile, experimental simulation proves the effectiveness of the proposed algorithm.

The invention brings various energy forms such as cogeneration, photovoltaic power generation, energy storage, boilers and the like into an industrial park so as to improve the energy efficiency and reduce the operation cost. In order to facilitate unified management, the invention introduces an energy hub comprising a battery, a hot water tank, a cogeneration device and renewable energy power generation. For energy hubs, many energy conversion methods can be used to shift supply and demand across spatio-temporal scales between multi-energy networks. Obviously, the energy hub can realize the complementarity and flexibility of scheduling by controlling and optimizing the multi-energy network, thereby realizing the high reliability of the multi-energy network, improving the utilization rate of distributed energy and improving the efficiency of the multi-energy network.

The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.

Claims

1. A peak shaving method of an integrated energy system based on user contribution behaviors is characterized by comprising the following steps:

the step 1 comprises the following steps:

step 2, establishing a Lagrange function of energy demand management;

step 3, establishing an energy scheduling optimization distributed algorithm;

and 4, carrying out energy distribution management of the park.

2. The integrated energy system peak shaving method based on user contribution behavior according to claim 1, wherein the step 1.1 comprises:

in the formula eta_dkeAnd η_dkhAll represent energy efficiency;

battery charging C_ke(t) and discharge D_ke(t) do not occur simultaneously, and the hot water tank stores heat energy C_kh(t) and transporting the heat quantity D outwards_kh(t) do not occur simultaneouslyThey have the following relationships:

the battery and the hot water tank are constrained by

B_k，min≤B_k(t)≤B_k，max#

W_k，min≤W_k(t)≤W_k，max#

0≤C_ke(t)≤C_ke，max，0≤D_ke(t)≤D_ke，max#

0≤C_kh(t)≤C_kh，max，0≤D_kh(t)≤D_kh，max#

0≤C_ke(t)≤min[B_k，max-B_k(t)，C_ke，max]#

0≤D_ke(t)≤min[B_k(t)-B_k，min，D_ke，max]#

0≤C_kh(t)≤min[W_k，max-W_k(t)，C_kh，max]#

0≤D_kh(t)≤min[W_k(t)-W_k，min，D_kh，max]#。

3. the integrated energy system peak shaving method based on user contribution behavior according to claim 2, wherein the step 1.2 comprises:

E_kCHP(t)＝η_kpgG_kCHP(t)，H_kCHP(t)＝η_khgG_kCHP(t)#

H_kb(t)＝η_kbgG_kb(t)#

0≤E_kCHP(t)≤E_kCHP，max，0≤H_kCHP(t)≤H_kCHP，max#

0≤H_kb(t)≤H_kb，max#。

4. the integrated energy system peak shaving method based on user contribution behavior according to claim 3, wherein the step 1.3 comprises:

0≤E(t)≤E_max，0≤G(t)≤G_max，0≤E_o(t)≤E_o，max#。

5. The integrated energy system peak shaving method based on user contribution behavior according to claim 4, wherein the step 1.4 comprises:

establishing a multi-energy supply and demand balance model of electric energy, natural gas and heat energy, wherein the electric energy balance relates to the cogeneration unit, the battery, the renewable energy R (t), the trade with the power grid and the demand E_tot(t); natural gas balance involving the cogeneration unit, boiler and demand G_tot(t); the heat energy balance relates to the cogeneration unit, the boiler, the hot water tank and the demand H_tot(t); the industrial park is supplied with energy by an energy hub, and the energy hub k in the industrial park is supplied with energy byThe system comprises a battery, a renewable energy source, a cogeneration unit, a boiler and a hot water tank, wherein K belongs to K; the supply and demand balance model of the industrial park is as follows:

X_i，min(t)≤X_i(t)≤X_i，max(t)#。

6. The integrated energy system peak shaving method based on user contribution behavior according to claim 5, wherein the step 1.5 comprises:

p_px(t)representing the price of energy sold to the user in the industrial park, a_ixAnd b_ixA user's energy use satisfaction factor;

7. the integrated energy system peak shaving method based on user contribution behavior according to claim 1, wherein the step 2 comprises:

8. the method according to claim 1, wherein the energy scheduling optimization distributed algorithm in step 3 implements optimization, communication and contribution functions.

9. The peak shaving method for integrated energy system based on user contribution behavior according to claim 8, wherein the optimization function means that the user individually solves the energy demand management problem according to the energy usage limit of each user without considering overload; the communication function refers to that a user communicates with others to process common constraints, namely overload control; the contribution function means that the user can reduce the energy stress of the overload period by changing the elastic electricity demand.

10. The integrated energy system peak shaving method based on user contribution behavior according to claim 1, wherein the step 4 comprises: after the optimal energy demand of the user is obtained, the social welfare maximization problem is solved, the mutual coupling coordination and complementation of multiple energy sources are utilized, and the optimal energy dispatching of the electric heat of the industrial park is obtained by adopting a linear programming method.