Park optimization operation method based on low-carbon demand response and interval number model

Abstract

The invention relates to the technical field of power utilization optimization scheduling, and discloses a park optimization operation method based on low carbon demand response and an interval number model, comprising the following steps: step 1, proposing a carbon metering method considering uncertainty; 2, constructing a master-slave game model between the source loads of the park according to the model and the method; 3, proposing a solving method of the model, wherein the solving method comprises interval possibility degree conversion and KKT conditions; the invention provides a carbon metering method considering uncertainty, which not only pays attention to static carbon emission data, but also can dynamically process uncertainty factors related to carbon emission, such as wind and light output and load demand response; according to the method, the accuracy and reliability of carbon metering are improved, so that the evaluation of carbon emission can better reflect the actual situation, and a firmer data basis is provided for low-carbon operation of a park.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity optimization scheduling, and in particular to a park optimization operation method based on low-carbon demand response and interval number model. Background Art

[0002] Under the current global climate change and energy crisis, all industries are seeking to reduce energy consumption and carbon emissions through technological innovation and management optimization. As a concentrated area of ​​industry, commerce and residents' lives, the park's energy consumption and carbon emissions are particularly prominent. Therefore, achieving low-carbon operation of the park is one of the key measures to achieve sustainable development.

[0003] Traditional energy management focuses more on cost-effectiveness and less on environmental impact. Although many parks have implemented energy optimization measures, such as demand response strategies, these strategies are mainly aimed at balancing grid loads and reducing electricity costs, but often ignore the impact of carbon emissions. In the context of global climate change, strategies that only focus on energy costs and ignore carbon emissions are no longer in line with the requirements of sustainable development.

[0004] The basic idea of ​​low-carbon demand response is to combine carbon emission costs with energy prices, and dynamically adjust price signals so that users can consider both economic costs and carbon costs when making energy decisions. This approach helps guide users to reduce energy consumption during periods of high carbon emissions, or give priority to the use of low-carbon or zero-carbon energy, thereby reducing overall carbon emissions.

[0005] Implementing low-carbon demand response in the park requires an efficient management and coordination mechanism. In this context, the master-slave game model provides a theoretical basis for solving multi-agent decision-making in the park. The master-slave game is a hierarchical decision-making model that is usually used to describe the interactive relationship between leaders and followers.

[0006] In addition, the implementation of low-carbon demand response also needs to consider a variety of uncertain factors, such as the unstable output of renewable energy. Therefore, the park's optimal operation method needs to have good adaptability and robustness, and be able to quickly adjust strategies when external conditions change. This requires the park management system to not only integrate efficient real-time data processing and prediction models, but also to have flexible decision support tools. Summary of the invention

[0007] The purpose of the present invention is to provide a park optimization operation method based on low-carbon demand response and interval number model to solve the problems raised in the above background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a park optimization operation method based on low-carbon demand response and interval number model, comprising the following steps:

[0009] Step 1: Propose a carbon accounting method that takes uncertainty into account, which includes an interval number model for source load prediction uncertainty and a carbon emission flow model;

[0010] Step 2: Based on the above model and method, a master-slave game model between the park source and load is constructed. The master-slave game model includes a demand response model on the load side and a profit maximization model on the source side.

[0011] Step 3: Propose a solution method for the above model, which includes interval possibility transformation and KKT conditions.

[0012] The interval number model for source load prediction uncertainty specifically includes:

[0013] Renewable energy generation and load are highly random and volatile. Since the probability distribution of this error is difficult to determine accurately and the fluctuation range is relatively small, interval numbers are used to deal with this uncertainty:

[0014]

[0015] In the formula, is the interval number; A is the true value; A - , A + Respectively represent the lower and upper bounds of the interval number;

[0016] The possible intervals selected are:

[0017]

[0018] In the formula, b is the allowed range of the variable, which is degenerated into a constant; is the probability density function of the uncertainty u, which is obtained by analyzing the historical operation data.

[0019] The carbon emission flow model specifically includes:

[0020] Carbon flow tracking can realize the transfer of carbon emissions from the source side to the load side. The carbon potential of each park sub-network can be obtained by the following formula:

[0021]

[0022] Where: is the carbon potential at time t; is the branch set of the flow into node i; P s,t is the branch active power; ρ s,t is the branch carbon flow density, which is numerically equal to the node carbon potential of the source node j of branch s is the active power of the power source connected to the node; is the carbon intensity of power generation, which is determined by the carbon emission factor and consumption rate of the fuel; P Nii,t is the node active flux, which means the sum of active powers flowing into the node in the direction of the flow.

[0023] The load-side demand response model specifically includes:

[0024]

[0025] Where: e is the e-th type of energy; R is the set of energy types, R = {E (electricity), H (heat), G (gas)}; is the load after demand response, is the user's baseline load; are the time-shiftable and interruptible loads of energy source e respectively; They are the minimum and maximum values ​​of the load that can be time-shifted, and the maximum value of the load that can be interrupted;

[0026]

[0027]

[0028] In the formula, Indicates energy efficiency; It indicates that users are dissatisfied with the cost; Indicates the energy cost of the user; is the benefit coefficient; is the user's satisfaction loss parameter for energy e; are the energy sales price of energy e and the unit compensation price of demand response; CET is the unit carbon price; is the average carbon emission factor.

[0030] The profit maximization model on the source side specifically includes:

[0031]

[0032] In the formula, Represents the revenue from energy sales from the park to the load, and satisfies They represent the interaction cost with the upper-level energy network, energy storage maintenance cost, new energy equipment maintenance cost, and energy conversion equipment maintenance cost respectively; They represent the gas purchase price, electricity purchase price and reverse electricity sales price respectively; eES The unit maintenance cost of energy storage for category e; Respectively represent the unit operation and maintenance costs of wind power and photovoltaic power; CHP , EB , GBRespectively represent the unit operation and maintenance costs of cogeneration units, electric boilers, and gas boilers; are the gas power and electric power transmitted from the upstream energy grid to park i, and the electric power transmitted from park i to the upstream energy grid; are the power of charging and discharging the energy storage respectively; They represent the power generated by photovoltaic, wind power, cogeneration units, electric boilers and gas boilers respectively.

[0033] The interval may be transformed, specifically including:

[0034] The objective function is transformed into the following formula:

[0035]

[0036] The equality constraints and inequality constraints are transformed as follows:

[0037]

[0038] In the formula, λ i1 and λ i2 is a pre-given possibility; x is a decision variable.

[0039] The KKT conditions specifically include:

[0040] The Lagrangian function of the lower objective function of the master-slave model is as follows:

[0041]

[0042] Where: The dual variable corresponding to the equality constraint; are the dual variables of the inequality constraints respectively;

[0043] Taking partial derivatives of the constructed Lagrangian function, we get the corresponding KKT conditions:

[0044]

[0045] The complementary relaxation condition is:

[0046]

[0047] Where: 0≤a⊥b≥0 means a≥0, b≥0 and a*b=0;

[0048] Since the above equation is a nonlinear constraint, the big-M method is used to introduce 0-1 variables to transform it into a mixed integer linear constraint. The first constraint in the above equation is taken as an example, and the rest of the transformations are similar;

[0049]

[0050] Where: M is a sufficiently large positive number; is a 0-1 variable;

[0051] Combining the KKT condition and the duality theorem, the nonlinear term can be eliminated and the upper objective function can be transformed into:

[0052]

[0053] Where: The dual variables of the corresponding inequality constraints respectively; Corresponding to the dual variables of the equality constraints, at this point, the master-slave model has been transformed into a single-layer linear model for solution.

[0054] The method for solving the master-slave game model specifically includes the following steps:

[0055] S1. Initialize and set the basic parameters and iteration parameters of the park energy system;

[0056] S2, using the interval number model for source load prediction uncertainty to describe source load uncertainty;

[0057] S3, using interval possibility transformation to transform uncertain constraints into certain constraints;

[0058] S4. Based on the KKT condition, the double-layer problem is equivalent to a single-layer problem;

[0059] S5, solving the transformed single-layer model and updating the load demand and energy price;

[0060] S6: If the convergence condition is met, the process ends. If the convergence condition is not met, S1-S5 are repeated until the convergence condition is met.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] 1. The present invention provides a carbon measurement method that takes uncertainty into account. This method not only focuses on static carbon emission data, but also can dynamically process uncertainty factors related to carbon emissions, such as wind and solar output and load demand response. This method improves the accuracy and reliability of carbon measurement, so that the assessment of carbon emissions can better reflect the actual situation, thereby providing a more solid data foundation for the low-carbon operation of the park.

[0063] 2. The present invention constructs a master-slave game model between park sources and loads based on the carbon metering method. The model takes the carbon metering results as key parameters and integrates them into the game process between energy supply (source) and demand (load) within the park. Through this model, park managers (masters) can formulate carbon emission targets and strategies, while different users (slaves) within the park make optimized decisions within the framework of these strategies. This model effectively coordinates the interests of all parties and promotes the reduction of overall carbon emissions and the optimal allocation of resources.

[0064] 3. The present invention provides a method for transforming a master-slave game model, which can effectively improve the solution speed while taking into account the uncertainty of the prediction. Through this transformation method, the theoretical model can be more smoothly applied to practical scenarios, ensuring the feasibility and effectiveness of the strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic diagram of the process of the park optimization operation method of the present invention;

[0066] Figure 2 This is a schematic diagram of the master-slave game in the park of the present invention;

[0067] Figure 3 The figure is a schematic diagram of the algorithm flow of the park master-slave game of the present invention. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0069] See also Figure 1 The present invention proposes a park optimization operation method based on low-carbon demand response and interval number model, comprising the following steps:

[0070] Step 1: Propose a carbon accounting method that takes uncertainty into account, which includes an interval number model for source load prediction uncertainty and a carbon emission flow model;

[0071] Step 2: Based on the above model and method, a master-slave game model between the park source and load is constructed. The master-slave game model includes a demand response model on the load side and a profit maximization model on the source side.

[0072] Step 3: Propose a solution method for the above model, which includes interval possibility transformation and KKT conditions.

[0073] The interval number model for source load prediction uncertainty specifically includes:

[0074] Renewable energy generation and load are highly random and volatile. Since the probability distribution of this error is difficult to determine accurately and the fluctuation range is relatively small, interval numbers are used to deal with this uncertainty:

[0075]

[0076] In the formula, is the interval number; A is the true value; A - , A + Respectively represent the lower and upper bounds of the interval number;

[0077] The possible intervals selected are:

[0078]

[0079] In the formula, b is the allowed range of the variable, which is degenerated into a constant; is the probability density function of the uncertainty u, which is obtained by analyzing the historical operation data.

[0080] The carbon emission flow model specifically includes:

[0081] Carbon flow tracking can realize the transfer of carbon emissions from the source side to the load side. The carbon potential of each park sub-network can be obtained by the following formula:

[0082]

[0083] Where: is the carbon potential at time t; is the branch set of the flow into node i; P s,t is the branch active power; ρ s,t is the branch carbon flow density, which is numerically equal to the node carbon potential of the source node j of branch s is the active power of the power source connected to the node; is the carbon intensity of power generation, which is determined by the carbon emission factor and consumption rate of the fuel; P Nii,t is the node active flux, which means the sum of active powers flowing into the node in the direction of the flow.

[0084] See also Figure 2 , the load-side demand response model specifically includes:

[0085]

[0086] Where: e is the e-th type of energy; R is the set of energy types, R = {E (electricity), H (heat), G (gas)}; is the load after demand response, is the user's baseline load; are the time-shiftable and interruptible loads of energy source e respectively; They are the minimum and maximum values ​​of the load that can be time-shifted, and the maximum value of the load that can be interrupted;

[0087]

[0088] In the formula, Indicates energy efficiency; It indicates that users are dissatisfied with the cost; Indicates the energy cost of the user; is the benefit coefficient; is the user's satisfaction loss parameter for energy e; are the energy sales price of energy e and the unit compensation price of demand response; CET is the unit carbon price; is the average carbon emission factor.

[0090] See also Figure 2 , the profit maximization model on the source side specifically includes:

[0091]

[0092] In the formula, Represents the revenue from energy sales from the park to the load, and satisfies They represent the interaction cost with the upper-level energy network, energy storage maintenance cost, new energy equipment maintenance cost, and energy conversion equipment maintenance cost respectively; They represent the gas purchase price, electricity purchase price and reverse electricity sales price respectively; eES The unit maintenance cost of energy storage for category e; Respectively represent the unit operation and maintenance costs of wind power and photovoltaic power; CHP , EB , GB Respectively represent the unit operation and maintenance costs of cogeneration units, electric boilers, and gas boilers; are the gas power and electric power transmitted from the upstream energy grid to park i, and the electric power transmitted from park i to the upstream energy grid; are the power of charging and discharging the energy storage respectively; They represent the power generated by photovoltaic, wind power, cogeneration units, electric boilers and gas boilers respectively.

[0093] See also Figure 3 , the interval may be transformed, specifically including:

[0094] The objective function is transformed into the following formula:

[0095]

[0096] The equality constraints and inequality constraints are transformed as follows:

[0097]

[0098] In the formula, λ i1 and λ i2 is a pre-given possibility; x is a decision variable.

[0099] See also Figure 3 The KKT conditions specifically include:

[0100] The Lagrangian function of the lower objective function of the master-slave model is as follows:

[0101]

[0102] Where: The dual variable corresponding to the equality constraint; are the dual variables of the inequality constraints respectively;

[0103] Taking partial derivatives of the constructed Lagrangian function, we get the corresponding KKT conditions:

[0104]

[0105] The complementary relaxation condition is:

[0106]

[0107] Where: 0≤a⊥b≥0 means a≥0, b≥0 and a*b=0;

[0108] Since the above equation is a nonlinear constraint, the big-M method is used to introduce 0-1 variables to transform it into a mixed integer linear constraint. The first constraint in the above equation is taken as an example, and the rest of the transformations are similar;

[0109]

[0110] Where: M is a sufficiently large positive number; is a 0-1 variable;

[0111] Combining the KKT condition and the duality theorem, the nonlinear term can be eliminated and the upper objective function can be transformed into:

[0112]

[0113] Where: The dual variables of the corresponding inequality constraints respectively; Corresponding to the dual variables of the equality constraints, at this point, the master-slave model has been transformed into a single-layer linear model for solution.

[0114] See also Figure 3 The method for solving the master-slave game model specifically includes the following steps:

[0115] S1. Initialize and set the basic parameters and iteration parameters of the park energy system;

[0116] S2, using the interval number model for source load prediction uncertainty to describe source load uncertainty;

[0117] S3, using interval possibility transformation to transform uncertain constraints into certain constraints;

[0118] S4. Based on the KKT condition, the double-layer problem is equivalent to a single-layer problem;

[0119] S5, solving the transformed single-layer model and updating the load demand and energy price;

[0120] S6: If the convergence condition is met, the process ends. If the convergence condition is not met, S1-S5 are repeated until the convergence condition is met.

[0121] In summary, the present invention provides a carbon measurement method that takes uncertainty into account. The method not only focuses on static carbon emission data, but also can dynamically process uncertainty factors related to carbon emissions, such as wind and solar output and load demand response; this method improves the accuracy and reliability of carbon measurement, so that the assessment of carbon emissions can better reflect the actual situation, thereby providing a more solid data foundation for the low-carbon operation of the park; the present invention constructs a master-slave game model between park sources and loads based on the carbon measurement method, which takes the carbon measurement results as key parameters and integrates them into the game process between energy supply (source) and demand (load) within the park; through this model, the park manager (master) can formulate carbon emission goals and strategies, and different users (slaves) in the park make optimization decisions within these strategy frameworks; this model effectively coordinates the interests of all parties, promotes the reduction of overall carbon emissions and the optimal allocation of resources; the present invention provides a master-slave game model transformation method, which can effectively improve the solution speed under the premise of taking into account the uncertainty of the prediction, and through this transformation method, the theoretical model can be more smoothly applied to actual scenarios to ensure the feasibility and effectiveness of the strategy.

[0122] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A park optimization operation method based on low-carbon demand response and interval number model, characterized in that: The following steps are involved: Step 1: Propose a carbon accounting method that takes uncertainty into account, which includes an interval number model for source load prediction uncertainty and a carbon emission flow model; Step 2: Based on the above model and method, a master-slave game model between the park source and load is constructed. The master-slave game model includes a demand response model on the load side and a profit maximization model on the source side. Step 3: Propose a solution method for the above model, which includes interval possibility transformation and KKT conditions.

2. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The interval number model for source load prediction uncertainty specifically includes: Renewable energy generation and load are highly random and volatile. Since the probability distribution of this error is difficult to determine accurately and the fluctuation range is relatively small, interval numbers are used to deal with this uncertainty: In the formula, is the interval number; A is the true value; A - , A + Respectively represent the lower and upper bounds of the interval number; The possible intervals selected are: In the formula, b is the allowed range of the variable, which is degenerated into a constant; is the probability density function of the uncertainty u, which is obtained by analyzing the historical operation data.

3. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The carbon emission flow model specifically includes: Carbon flow tracking can realize the transfer of carbon emissions from the source side to the load side. The carbon potential of each park sub-network can be obtained by the following formula: Where: is the carbon potential at time t; is the branch set of the flow into node i; P s,t is the branch active power; ρ s,t is the branch carbon flow density, which is numerically equal to the node carbon potential of the source node j of branch s is the active power of the power source connected to the node; is the carbon intensity of power generation, which is determined by the carbon emission factor and consumption rate of the fuel; P Nii,t is the node active flux, which means the sum of active powers flowing into the node in the direction of the flow.

4. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The load-side demand response model specifically includes: Where: e is the e-th type of energy; R is the set of energy types, R = {E (electricity), H (heat), G (gas)}; is the load after demand response, is the user's baseline load; are the time-shiftable and interruptible loads of energy source e respectively; They are the minimum and maximum values ​​of the load that can be time-shifted, and the maximum value of the load that can be interrupted; In the formula, Indicates energy efficiency; It indicates that users are dissatisfied with the cost; Indicates the energy cost of the user; is the benefit coefficient; is the user's satisfaction loss parameter for energy e; are the energy sales price of energy e and the unit compensation price of demand response; CET is the unit carbon price; is the average carbon emission factor.

5. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The profit maximization model on the source side specifically includes: In the formula, Represents the revenue from energy sales from the park to the load, and satisfies They represent the interaction cost with the upper-level energy network, energy storage maintenance cost, new energy equipment maintenance cost, and energy conversion equipment maintenance cost respectively; They represent the gas purchase price, electricity purchase price and reverse electricity sales price respectively; eES The unit maintenance cost of energy storage for category e; Respectively represent the unit operation and maintenance costs of wind power and photovoltaic power; CHP , EB , GB Respectively represent the unit operation and maintenance costs of cogeneration units, electric boilers, and gas boilers; are the gas power and electric power transmitted from the upstream energy grid to park i, and the electric power transmitted from park i to the upstream energy grid; are the power of charging and discharging the energy storage respectively; They represent the power generated by photovoltaic, wind power, cogeneration units, electric boilers and gas boilers respectively.

6. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The interval may be transformed, specifically including: The objective function is transformed into the following formula: The equality constraints and inequality constraints are transformed as follows: In the formula, λ i1 and λ i2 is a pre-given possibility; x is a decision variable.

7. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The KKT conditions specifically include: The Lagrangian function of the lower objective function of the master-slave model is as follows: Where: The dual variable corresponding to the equality constraint; are the dual variables of the inequality constraints respectively; Taking partial derivatives of the constructed Lagrangian function, we get the corresponding KKT conditions: The complementary relaxation condition is: Where: 0≤a⊥b≥0 means a≥0, b≥0 and a*b=0; Since the above equation is a nonlinear constraint, the big-M method is used to introduce 0-1 variables to transform it into a mixed integer linear constraint. The first constraint in the above equation is taken as an example, and the rest of the transformations are similar; Where: M is a sufficiently large positive number; is a 0-1 variable; Combining the KKT condition and the duality theorem, the nonlinear term can be eliminated and the upper objective function can be transformed into: Where: The dual variables of the corresponding inequality constraints respectively; Corresponding to the dual variables of the equality constraints, at this point, the master-slave model has been transformed into a single-layer linear model for solution.

8. The park optimization operation method based on low-carbon demand response and interval number model according to claim 1 is characterized by: The method for solving the master-slave game model specifically includes the following steps: S1. Initialize and set the basic parameters and iteration parameters of the park energy system; S2, using the interval number model for source load prediction uncertainty to describe source load uncertainty; S3, using interval possibility transformation to transform uncertain constraints into certain constraints; S4. Based on the KKT condition, the double-layer problem is equivalent to a single-layer problem; S5, solving the transformed single-layer model and updating the load demand and energy price; S6: If the convergence condition is met, the process ends. If the convergence condition is not met, S1-S5 are repeated until the convergence condition is met.

Images