A new energy consumption optimization scheduling method and device considering multi-agent carbon emission responsibility allocation

Abstract

A new energy consumption optimization scheduling method and device considering multi-agent carbon emission responsibility allocation, the method comprises the following steps: according to the use amount of user clean electric energy, the responsibility of carbon emission of user is allocated; considering the output power fluctuation, prediction accuracy and other factors, the reliability of new energy output is evaluated; a model with different optimization objectives is built, and the realization of distribution network layer safety and stability, regional carbon emission reduction and the improvement of the benefits of each participant are taken as the objectives; through the reinforcement learning method, the scheduling plan of new energy generation and user is obtained layer by layer, and the economic and safe consumption of new energy considering different subject carbon emission responsibility allocation is realized. The present application can fairly consider the responsibility and contribution allocation of multiple subjects in carbon emission, reduce new energy power fluctuation, reduce new energy power generation, improve the consumption level of clean energy to reduce carbon emission, and realize the fair optimization scheduling of safety, economy and carbon environmental protection in the distribution network with high proportion of new energy access.

Description

Technical Field

[0001] This invention relates to the field of clean energy consumption and optimal scheduling, specifically a method and apparatus for optimal scheduling of energy consumption that considers the sharing of carbon emission responsibilities among multiple stakeholders. Background Technology

[0002] Excessive carbon emissions contribute to global warming, seriously threatening the planet's future sustainable development. While renewable energy sources such as wind and solar power offer advantages like cleanliness and low pollution, they also exhibit significant volatility and uncertainty. This necessitates the development of reasonable dispatch plans that involve user participation in demand response, facilitating the safe and economical smooth absorption of renewable energy. However, existing electricity market mechanisms lack a clear delineation of responsibilities among different stakeholders, particularly regarding carbon emission liability. This fails to guarantee fairness in electricity market transactions and discourages proactive cooperation from all parties in dispatching. Therefore, it is necessary to implement a carbon emission responsibility sharing mechanism among different stakeholders, comprehensively considering carbon emissions, safe operation, and economic optimization during dispatching optimization to protect the interests of renewable energy generators, electricity users, and the distribution network.

[0003] Existing research on absorbing a high proportion of renewable energy mainly focuses on market mechanisms, optimal scheduling, and demand response mining. One study established a decentralized two-tier trading mechanism, allowing numerous small users of different types to directly participate in market transactions. Another study established a P2P market, utilizing continuously adjustable load matching to mitigate renewable energy curtailment. A proposed distributed power trading market fully utilizes electric vehicle resources for highly accurate and low-cost load regulation. However, these studies largely focus on improving trading economics through demand response, with little attention paid to optimal scheduling related to safety and low-carbon environments. A third study, based on interval estimation, extends from single-objective optimization to multi-objective optimization, jointly considering safety, economics, and power quality. A hybrid economic environment optimization model was proposed, simplifying the CEF carbon emission model using a Bayesian model. A bilateral carbon tax mechanism was studied, attempting to transfer carbon costs to end-users. However, these studies still face a significant challenge: the lack of clear delineation of responsibilities among stakeholders fails to guarantee long-term trading fairness, hinders the stimulation of user demand response potential, and fails to achieve joint optimization of economics, safety, and environmental protection.

[0004] References

[0005] 1.0F. Luo, Z. Y. Dong, G. Liang, J. Murata and Z. Xu, "A DistributedElectricity Trading System in Active Distribution Networks Based on Multi-Agent Coalition and Blockchain," in IEEE Transactions on Power Systems, vol.34, no. 5, pp. 4097-4108, Sept. 2019.

[0006] 2.0Z. Zhang, R. Li and F. Li, "A Novel Peer-to-Peer Local ElectricityMarket for Joint Trading of Energy and Uncertainty," in IEEE Transactions onSmart Grid, vol. 11, no. 2, pp. 1205-1215, March 2020.

[0007] 3.0Y. Li and B. Hu, "An Iterative Two-Layer Optimization Charging andDischarging Trading Scheme for Electric Vehicle Using Consortium Blockchain,"in IEEE Transactions on Smart Grid, vol. 11, no. 3, pp. 2627-2637, May 2020.

[0008] 4.0Kong.Xiangyu, Kong. Deqian, Yao.Jingtao, "8.Online pricing ofdemand response based on long short-term memory and reinforcement learning,"in Applied Energy, vol. 271, pp. 114945, Aug 2020.

[0009] 5.0Y. Li, P. Wang, "Multi-Objective Optimal Dispatch of MicrogridUnder Uncertainties via Interval Optimization," in IEEE Transactions on SmartGrid, vol. 10, no. 2, pp. 2046-2058, March 2019. Summary of the Invention

[0010] The purpose of this invention is to provide a new energy consumption optimization scheduling method based on considering the carbon emission responsibility sharing of multiple entities. This method can fairly consider the responsibilities and contributions of new energy generators and users in terms of carbon emissions, promote the active participation of all entities in new energy consumption scheduling, and facilitate fair optimization scheduling that balances safety, economy and carbon environmental protection in distribution networks with a high proportion of new energy access.

[0011] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0012] A renewable energy consumption optimization scheduling method considering the sharing of carbon emission responsibilities among multiple stakeholders includes the following steps:

[0013] Based on the amount of clean energy used by users, the carbon emission responsibility of users is allocated to them, resulting in a carbon emission responsibility factor.

[0014] The reliability of new energy power generators is evaluated to obtain the new energy reliability factor;

[0015] To achieve the goals of ensuring the safety and stability of the distribution network, reducing regional carbon emissions, and improving the benefits of all participants, models with different optimization objectives are built, including a distribution network safety optimization model, a regional carbon emission control optimization model, a new energy equipment economic optimization model, and an optimization model for power users. The carbon emission responsibility factor and the new energy reliability factor are added as parameters to the objective equation.

[0016] Based on the established models with different optimization objectives, the scheduling plans for new energy power generation and users are obtained layer by layer through reinforcement learning methods, so as to realize the economical and safe consumption of new energy considering the carbon emission reduction responsibility sharing of different entities.

[0017] Furthermore, the step of allocating carbon emission responsibility to users based on their clean energy consumption to obtain a carbon emission responsibility factor includes the following steps:

[0018] Based on the inherent characteristics of user nodes and the number of connection branches, determine the carbon emission parameters of each user node. and ,in, and These represent the carbon emission intensity coefficients of branch current and user node injected power, respectively.

[0019] A carbon emission flow model is introduced, which uses the nodal power on the demand side to calculate the relevant carbon emission intensity and calculates the nodal carbon emissions based on the active power of the load flowing into each node and the carbon emission parameters:

[0020]

[0021] For user node j at time t, The nodal carbon emission intensity refers to the energy consumption intensity of user loads. Indicates the power flowing through the branch. This indicates the amount of electricity generated by the node.

[0022] Based on the principle of carbon emission sharing, the carbon emission reduction ratio of each user node is calculated using the carbon emission amounts of all user nodes within the scope, that is, the proportion of the user's carbon emission reduction to the total user carbon emission reduction:

[0023]

[0024] in, This refers to user j's carbon emission responsibility factor at time t. This refers to the power consumption of user nodes;

[0025] Calculate the carbon emission responsibility factor based on user demand response characteristics. To comprehensively evaluate users' contributions during the dispatching process, the responsibility factor considers users' willingness and ability in demand response and carbon emission reduction. It consists of two parts: the first part represents users' efforts to compensate for insufficient renewable energy output by adjusting energy consumption, thereby reducing renewable energy curtailment and power fluctuations; the second part represents users' contribution to reducing carbon emissions.

[0026]

[0027]

[0028]

[0029]

[0030] For user i at time t, This refers to demand response power; This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption.

[0031] Furthermore, the assessment of the reliability of new energy power generators to obtain a new energy reliability factor includes the following steps:

[0032] Based on the predicted power output and actual available power output data of new energy sources, the prediction accuracy factor of new energy power output is calculated. :

[0033]

[0034] Among them, for the new energy generator i at time t, To predict the power output of new energy sources, This represents the actual output power of new energy sources;

[0035] Calculate the active power output fluctuation factor of the new energy source based on the power output of the new energy source at two adjacent moments:

[0036]

[0037] Based on the clean energy power generation of new energy sources, and according to their output proportion, calculate the contribution of each new energy power generator to the reduction of carbon emissions:

[0038]

[0039] in, This refers to the carbon emission reduction intensity factor brought about by renewable energy power generators using clean energy supply. Let i be the power generation of the renewable energy generator at time t;

[0040] Calculating the reliability factor of renewable energy generators based on the concept of Mahalanobis distance :

[0041]

[0042]

[0043] in, This represents the column vector of the obtained new energy generator i. It is a column vector of mean. This describes the process of covariance matrix operations.

[0044] Furthermore, the steps of building models with different optimization objectives—namely, achieving distribution network security and stability, regional carbon emission reduction, and improving the benefits for all participants—include the following:

[0045] A distribution network layer security optimization model is established. The distribution network company is responsible for the security of the entire distribution network. The optimization objective is to ensure power balance, handle power fluctuations caused by the uncertainty of new energy sources, and ensure good power quality. The formula of the distribution network layer security optimization model includes two terms: the first term is the impact of active power fluctuations from new energy sources, and the second term is the limitation on unbalanced power output.

[0046]

[0047]

[0048]

[0049] st

[0050]

[0051]

[0052] The constraints include the balance between load power consumption and renewable energy output at any given time, renewable energy output limits, and the range of load adjustment in response to user demand. A piecewise function representing output power fluctuations. It is a punitive factor that ensures a balance of power. This indicates the upper limit of the output of the new energy power generator i. and These are the upper and lower limits for adjusting the energy load of user j;

[0053] A regional carbon emission control optimization model is established. At the regional level, the optimization objective of the power trading center is to control carbon emissions, which includes two parts: reducing the curtailment of renewable energy by replacing traditional thermal power with more renewable energy generation; and promoting the local consumption of renewable energy within the region to reduce energy loss and carbon emissions caused by long-distance inter-regional transmission. The regional carbon emission control optimization model is as follows:

[0054]

[0055] st

[0056] The constraint is the upper and lower limits of the transmission capacity of the transmission feeder. This refers to the maximum possible output of new energy power generators; It is the line loss factor of line n; It is the power of branch n at time t. and Indicates the power limit of the branch;

[0057] According to the new energy reliability factor described in step [1] An economic optimization model for new energy equipment is constructed. The resource equipment layer includes two main entities: power generators and electricity users. They adjust transaction prices based on reference electricity prices, participate in market negotiations, and obtain their respective maximum benefits. The optimization objective of new energy power generators is to obtain the maximum net revenue from electricity sales, which is divided into three parts: electricity sales revenue, demand response compensation costs, and the potential value of reliability factors. The economic optimization model for new energy equipment is as follows:

[0058] st

[0059] The constraint is the fluctuation range of the electricity price for electricity generated from new energy sources. This represents the normal bidding price for renewable energy generator i at time t. Indicates demand response electricity price;

[0060] Based on the calculated carbon emission responsibility factor An optimization model for electricity users is established, aiming to minimize electricity costs. It consists of three parts: demand response revenue, electricity costs, and the potential value of return factors, including carbon emission responsibility factors. This will affect the user's long-term trading profits, so it is included in the optimization objective. The optimization model for electricity users is as follows:

[0061]

[0062] in, This represents the revenue price for the j-th user's demand response during time period t. The electricity price for user j during time period t. This represents the estimated power consumption in response to previous plans. This represents the user's actual power consumption, NT represents the total number of response periods, and NU represents the total number of users.

[0063] Furthermore, based on the established models with different optimization objectives, the scheduling plans for new energy power generation and users are obtained layer by layer through reinforcement learning methods, realizing the economical and secure consumption of new energy considering the carbon emission reduction responsibilities of different entities. This includes the following steps:

[0064] Based on the aforementioned distribution network layer security optimization model, the reinforcement learning Q-learning method is used to first solve the overall security optimization scheduling plan of the distribution network, thereby obtaining the distribution network scheduling plan. The distribution network scheduling plan includes the overall renewable energy output power and total user demand response of the entire network in each time period.

[0065] Based on the distribution network scheduling plan, and according to the regional carbon emission control optimization model, the regional carbon emission reduction optimization model is solved to obtain the regional scheduling plan. The regional scheduling plan includes the total output power of all new energy sources and the total demand response of all users in each region.

[0066] Based on the regional dispatch plan, dispatch instructions are obtained according to the economic optimization model of the new energy equipment and the optimization model of the power users. The dispatch instructions include the output plan of each new energy generator and the demand response load adjustment of the users in different time periods within the region.

[0067] The dispatch instructions are issued to users and new energy sources, namely the output plans of new energy power generators and the demand response load adjustments of users at different times, and the optimization of new energy consumption taking into account the carbon emission responsibility sharing is executed and completed.

[0068] A renewable energy consumption optimization scheduling device considering the sharing of carbon emission responsibilities among multiple stakeholders includes:

[0069] The carbon emission responsibility factor acquisition module is used to allocate carbon emission responsibility to users based on their clean energy consumption, and obtain carbon emission responsibility factors.

[0070] The new energy reliability factor acquisition module is used to evaluate the reliability of new energy generators and obtain the new energy reliability factor.

[0071] The optimization model building module is used to build models with different optimization objectives, namely, to achieve distribution network security and stability, regional carbon emission reduction and improve the benefits of each participant. These include distribution network security optimization model, regional carbon emission control optimization model, new energy equipment economic optimization model and power user optimization model. The obtained carbon emission responsibility factor and new energy reliability factor are added as parameters to the objective equation.

[0072] The scheduling plan acquisition module is used to obtain the scheduling plans of new energy power generation and users layer by layer through reinforcement learning based on the models with different optimization objectives, so as to realize the economic and safe consumption of new energy considering the carbon emission reduction responsibility of different entities.

[0073] Furthermore, the carbon emission responsibility factor acquisition module allocates carbon emission responsibility to users based on their clean energy consumption, resulting in carbon emission responsibility factors, including:

[0074] Based on the inherent characteristics of user nodes and the number of connection branches, determine the carbon emission parameters of each user node. and ,in, and These represent the carbon emission intensity coefficients of branch current and user node injected power, respectively.

[0075] A carbon emission flow model is introduced, which uses the nodal power on the demand side to calculate the relevant carbon emission intensity and calculates the nodal carbon emissions based on the active power of the load flowing into each node and the carbon emission parameters:

[0076]

[0077] For user node j at time t, The nodal carbon emission intensity refers to the energy consumption intensity of user loads. Indicates the power flowing through the branch. This indicates the amount of electricity generated by the node.

[0078] Based on the principle of carbon emission sharing, the carbon emission reduction ratio of each user node is calculated using the carbon emission amounts of all user nodes within the scope, that is, the proportion of the user's carbon emission reduction to the total user carbon emission reduction:

[0079]

[0080] in, This refers to user j's carbon emission responsibility factor at time t. This refers to the power consumption of user nodes;

[0081] Calculate the carbon emission responsibility factor based on user demand response characteristics. To comprehensively evaluate users' contributions during the dispatching process, the responsibility factor considers users' willingness and ability in demand response and carbon emission reduction. It consists of two parts: the first part represents users' efforts to compensate for insufficient renewable energy output by adjusting energy consumption, thereby reducing renewable energy curtailment and power fluctuations; the second part represents users' contribution to reducing carbon emissions.

[0082]

[0083]

[0084]

[0085]

[0086] For user i at time t, This refers to demand response power; This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption.

[0087] Furthermore, the new energy reliability factor acquisition module evaluates the reliability of new energy power generators to obtain new energy reliability factors, including:

[0088] Based on the predicted power output and actual available power output data of new energy sources, the prediction accuracy factor of new energy power output is calculated. :

[0089]

[0090] Among them, for the new energy generator i at time t, To predict the power output of new energy sources, This represents the actual output power of new energy sources;

[0091] Calculate the active power output fluctuation factor of the new energy source based on the power output of the new energy source at two adjacent moments:

[0092]

[0093] Based on the clean energy power generation of new energy sources, and according to their output proportion, calculate the contribution of each new energy power generator to the reduction of carbon emissions:

[0094]

[0095] in, This refers to the carbon emission reduction intensity factor brought about by renewable energy power generators using clean energy supply. Let i be the power generation of the renewable energy generator at time t;

[0096] Calculating the reliability factor of renewable energy generators based on the concept of Mahalanobis distance :

[0097]

[0098]

[0099] in, This represents the column vector of the obtained new energy generator i. It is a column vector of mean. This describes the process of covariance matrix operations.

[0100] Furthermore, the optimization model building module aims to achieve distribution network security and stability, regional carbon emission reduction, and improve the benefits for all participants, building models with different optimization objectives, including:

[0101] A distribution network layer security optimization model is established. The distribution network company is responsible for the security of the entire distribution network. The optimization objective is to ensure power balance, handle power fluctuations caused by the uncertainty of new energy sources, and guarantee good power quality. The formula of the distribution network layer security optimization model includes two terms: the first term represents the impact of active power fluctuations from new energy sources, and the second term represents the limitation on unbalanced power output.

[0102]

[0103]

[0104]

[0105] st

[0106]

[0107]

[0108] The constraints include the balance between load power consumption and renewable energy output at any given time, renewable energy output limits, and the range of load adjustment in response to user demand. A piecewise function representing output power fluctuations. It is a punitive factor that ensures a balance of power. This indicates the upper limit of the output of the new energy power generator i. and These are the upper and lower limits for adjusting the energy load of user j;

[0109] A regional carbon emission control optimization model is established. At the regional level, the optimization objective of the power trading center is to control carbon emissions, which includes two parts: reducing the curtailment of renewable energy by replacing traditional thermal power with more renewable energy generation; and promoting the local consumption of renewable energy within the region to reduce energy loss and carbon emissions caused by long-distance inter-regional transmission. The regional carbon emission control optimization model is as follows:

[0110]

[0111] st

[0112] The constraint is the upper and lower limits of the transmission capacity of the transmission feeder. This refers to the maximum possible output of new energy power generators; It is the line loss factor of line n; It is the power of branch n at time t. and Indicates the power limit of the branch;

[0113] According to the new energy reliability factor described in step [1] An economic optimization model for new energy equipment is constructed. The resource equipment layer includes two main entities: power generators and electricity users. They adjust transaction prices based on reference electricity prices, participate in market negotiations, and obtain their respective maximum benefits. The optimization objective of new energy power generators is to obtain the maximum net revenue from electricity sales, which is divided into three parts: electricity sales revenue, demand response compensation costs, and the potential value of reliability factors. The economic optimization model for new energy equipment is as follows:

[0114] st

[0115] The constraint is the fluctuation range of the electricity price for electricity generated from new energy sources. This represents the normal bidding price for renewable energy generator i at time t. Indicates demand response electricity price;

[0116] Based on the calculated carbon emission responsibility factor An optimization model for electricity users is established, aiming to minimize electricity costs. It consists of three parts: demand response revenue, electricity costs, and the potential value of return factors, including carbon emission responsibility factors. This will affect the user's long-term trading profits, so it is included in the optimization objective. The optimization model for electricity users is as follows:

[0117]

[0118] in, This represents the revenue price for the j-th user's demand response during time period t. The electricity price for user j during time period t. This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption. NT represents the total number of response periods, and NU represents the total number of users.

[0119] Furthermore, the scheduling plan acquisition module, based on the established models with different optimization objectives, obtains the scheduling plans for new energy power generation and users layer by layer through reinforcement learning methods, realizing the economical and secure consumption of new energy considering the carbon emission reduction responsibility sharing of different entities, including:

[0120] Based on the aforementioned distribution network layer security optimization model, the reinforcement learning Q-learning method is used to first solve the overall security optimization scheduling plan of the distribution network, thereby obtaining the distribution network scheduling plan. The distribution network scheduling plan includes the overall renewable energy output power and total user demand response of the entire network in each time period.

[0121] Based on the distribution network scheduling plan, and according to the regional carbon emission control optimization model, the regional carbon emission reduction optimization model is solved to obtain the regional scheduling plan. The regional scheduling plan includes the total output power of all new energy sources and the total demand response of all users in each region.

[0122] Based on the regional dispatch plan, dispatch instructions are obtained according to the economic optimization model of the new energy equipment and the optimization model of the power users. The dispatch instructions include the output plan of each new energy generator and the demand response load adjustment of the users in different time periods within the region.

[0123] The dispatch instructions are issued to users and new energy sources, namely the output plans of new energy power generators and the demand response load adjustments of users at different times, and the optimization of new energy consumption taking into account the carbon emission responsibility sharing is executed and completed.

[0124] The beneficial effects of the technical solution provided by this invention are:

[0125] (1) In terms of new energy consumption, the optimization method of new energy consumption that considers the sharing of carbon emission responsibilities among multiple entities should be considered. Starting from the accuracy of predicted output and the fluctuation of grid-connected power, the clean power quality of different new energy generators at the time of grid connection should be fully considered. This can promote new energy generators to obtain better output benefits by improving power quality, forming a positive incentive cycle. This will not only promote the safe and stable operation of new energy grid connection, but also improve the economic benefits of new energy generators.

[0126] (2) At the residential user response level, this new energy consumption optimization method, which considers the sharing of carbon emission responsibilities among multiple stakeholders, can assess the user's responsibility for carbon emissions by combining their own electricity consumption habits and response capabilities. This allows for a more scientific and reasonable formulation of user energy consumption adjustment and scheduling plans. By calculating responsibility factors to comprehensively evaluate the user's contribution to carbon emission reduction and smooth new energy consumption, users can obtain economic benefits commensurate with their contributions. This incentivizes users to actively participate in carbon emission reduction and demand response, helping to achieve environmentally friendly and economical new energy consumption.

[0127] (3) At the societal level, this optimization method for new energy consumption that considers the sharing of carbon emission responsibilities among multiple stakeholders can provide policymakers, grid dispatchers, high-efficiency research teachers and students, sales personnel, etc., with a reference for the sharing of carbon emission reduction responsibilities and the formulation of dispatch plans. It can effectively promote the formulation of strategies to meet the future development needs of society, carry out practical carbon emission reduction, power dispatch and power user response management applications, provide researchers with a reference for exploring carbon emission reduction strategies under the new energy consumption environment, and enable energy sales managers to fully tap the potential of user-side response, mobilize user-side response to the smooth consumption demand of new energy from the whole, ensure the safe and stable operation of the power system, realize the use of clean and environmentally friendly electricity, reduce carbon emissions, and effectively improve the level of social energy intelligent interaction and the development of management technology. Attached Figure Description

[0128] Figure 1 A diagram illustrating the user responsibility factor calculation method;

[0129] Figure 2 A diagram illustrating the calculation method for the reliability factor of new energy sources;

[0130] Figure 3 A flowchart of a new energy consumption optimization scheduling method that considers the sharing of carbon emission reduction responsibilities among multiple stakeholders;

[0131] Figure 4 User load in different regions;

[0132] Figure 5 For different regions, the user demand response volume and corresponding incentive electricity price;

[0133] Figure 6 For comparison of changes in carbon emissions;

[0134] Figure 7 The impact of reliability factor on the revenue of new energy sources. Detailed Implementation

[0135] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below.

[0136] Please see Figure 1-3 The renewable energy consumption optimization scheduling method proposed in this invention, which considers the sharing of carbon emission responsibilities among multiple stakeholders, includes four key steps: user carbon emission responsibility sharing, renewable energy reliability assessment, hierarchical optimization objective determination, and reinforcement learning scheduling plan formulation. See the description below for details:

[0137] Step 1: User carbon emission responsibility sharing, which includes the following steps:

[0138] Step 1.1: Determine the carbon emission parameters of each user node based on its inherent characteristics and the number of connection branches. and ,in, and These represent the carbon emission intensity coefficients for branch current and user node injected power, respectively.

[0139] Step 1.2: Introduce a carbon emission flow model. This model uses the nodal power on the demand side to calculate the relevant carbon emission intensity. Based on the active power of the load flowing into each node and the carbon emission parameters, calculate the nodal carbon emissions:

[0140]

[0141] For user node j at time t, The nodal carbon emission intensity refers to the energy consumption intensity of user loads. Indicates the power flowing through the branch. This indicates the amount of power generated by the node.

[0142] Step 1.3: Based on the principle of carbon emission sharing, using the carbon emissions of all user nodes within the scope, calculate the carbon reduction ratio corresponding to each user node, that is, the proportion of the user's carbon emission reduction to the total user carbon emission reduction:

[0143]

[0144] in, This refers to the carbon emission reduction factor of user j at time t. This refers to the power consumption of user nodes.

[0145] Step 1.4: Calculate the carbon emission responsibility factor based on the user's demand response characteristics. To comprehensively evaluate users' contributions during the dispatching process, the responsibility factor considers users' willingness and ability in demand response and carbon emission reduction. It consists of two parts: the first part represents users' efforts to compensate for insufficient renewable energy output by adjusting energy consumption, thereby reducing renewable energy curtailment and power fluctuations; the second part represents users' contribution to reducing carbon emissions.

[0146]

[0147]

[0148]

[0149]

[0150] For user i at time t, This refers to demand response power; This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption.

[0151] Step 2: Renewable Energy Reliability Assessment. Based on historical data, a new energy reliability factor is introduced to assess the historical output quality and creditworthiness of different new energy power generators, serving as a reference indicator for their priority in electricity market transactions and settlement. This includes the following steps:

[0152] Step 2.1: Calculate the prediction accuracy factor of new energy output based on the predicted output and actual available output data of new energy sources. :

[0153]

[0154] Among them, for the new energy generator i at time t, To predict the power output of new energy sources, This refers to the actual output power of new energy sources.

[0155] Step 2.2: Calculate the active power output fluctuation factor of the new energy source based on the power output of the new energy source at two adjacent moments.

[0156]

[0157] Step 2.3: Based on the clean energy power generation of new energy sources, calculate the contribution of each new energy power generator to the reduction of carbon emissions according to its output proportion:

[0158]

[0159] in, This refers to the carbon emission reduction intensity factor brought about by renewable energy power generators using clean energy supply. Let t be the power generation of renewable energy generator i at time t.

[0160] Step 2.4: Calculate the reliability factor of the renewable energy generator. Because they have different units of measurement but are related to each other, the reliability factor is calculated based on the idea of ​​Mahalanobis distance:

[0161]

[0162]

[0163] in, Let represent the column vector of the new energy power generator i obtained in steps 2.1, 2.2, and 2.3. It is a column vector of mean. This describes the process of covariance matrix operations.

[0164] Step 3: Determining the Hierarchical Optimization Objective. To balance the economic, safe, and low-carbon environmental aspects of dispatching, this paper proposes a hierarchical multi-agent optimization model based on the distribution network layer, regional layer, and resource equipment layer. The carbon emission responsibility factor obtained in Step 1 and the renewable energy reliability factor obtained in Step 2 are added as parameters to the objective equation. The specific steps include the following:

[0165] Step 3.1: Establish a security optimization model for the distribution network layer. The distribution network layer is the responsibility of the distribution network company for the security of the entire distribution network. The optimization objective is to ensure power balance and handle power fluctuations caused by the uncertainty of new energy sources, thereby ensuring good power quality. Its formula includes two terms: the first term is the impact of active power fluctuations from new energy sources, and the second term is the limitation on unbalanced power generation.

[0166]

[0167]

[0168]

[0169] st

[0170]

[0171]

[0172] The constraints include the balance between load power consumption and renewable energy output at any given time, renewable energy output limits, and the range of load adjustment in response to user demand. A piecewise function representing output power fluctuations. It is a punitive factor that ensures a balance of power. This indicates the upper limit of the output of the new energy power generator i. and It is the upper and lower limit of the energy load adjustment for user j.

[0173] Step 3.2: Establish a regional carbon emission control optimization model. At the regional level, the optimization objective of the power trading center is primarily to control carbon emissions, which mainly includes two parts: reducing the curtailment of renewable energy and replacing traditional thermal power with more renewable energy generation; and promoting the local consumption of renewable energy within the region, reducing energy loss and carbon emissions caused by long-distance inter-regional transmission. The regional carbon emission control optimization model is as follows:

[0174]

[0175] st

[0176] The constraint is the upper and lower limits of the transmission capacity of the transmission feeder. This refers to the maximum possible output of new energy power generators; It is the line loss factor of line n; It is the power of branch n at time t. and This indicates the power limit of the branch.

[0177] Step 3.3: Based on the new energy reliability factor obtained in Step 2.4 An economic optimization model for new energy equipment is established. The resource and equipment layer comprises two main entities: power generators and electricity users. Based on a reference electricity price, they adjust transaction prices, participate in market negotiations, and strive to maximize their respective interests. The optimization objective for new energy power generators is to obtain maximum net revenue from electricity sales, which is divided into three parts: revenue from electricity sales, demand response compensation costs, and the potential value of reliability factors. The economic optimization model for new energy equipment is as follows:

[0178]

[0179] st

[0180] The constraint is the fluctuation range (upper and lower limits) of the electricity price for new energy power generation. This represents the normal bidding price for renewable energy generator i at time t. This indicates the demand response electricity price.

[0181] Step 3.4: Based on the carbon emission responsibility factor calculated in Step 1.4 An optimization model for electricity users is established. The objective is to minimize electricity costs, and it is divided into three parts: demand response revenue, electricity costs, and the potential value of return factors. This is because of the carbon emission responsibility factor. This will affect the user's long-term trading profits, so it is included in the optimization objective. The optimization model for electricity users is as follows:

[0182]

[0183] in, This represents the revenue price for the j-th user's demand response during time period t. The electricity price for user j during time period t. This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption. NT represents the total number of response periods, and NU represents the total number of users.

[0184] Step 4: Based on the model listed in Step 3, reinforcement learning is used to formulate a scheduling plan, which includes the following steps:

[0185] Step 4.1: Based on the distribution network security optimization model in Step 3.1, the Q-learning reinforcement learning method is used to solve the overall security optimization scheduling plan of the distribution network to obtain the distribution network scheduling plan. The distribution network scheduling plan includes the overall renewable energy output power and the total user demand response of the entire network in each time period.

[0186] Step 4.2: Based on the distribution network scheduling plan obtained in Step 4.1, and according to the regional carbon emission control optimization model in Step 3.2, solve the regional carbon emission reduction optimization model to obtain the regional scheduling plan. The regional scheduling plan includes the total output power of all new energy sources and the total demand response of all users in each region.

[0187] Step 4.3: Based on the regional dispatch plan obtained in Step 4.2, and according to the economic optimization model of new energy equipment in Step 3.3 and the optimization model of power users in Step 3.4, dispatch instructions are obtained, including the output plan of each new energy generator and the demand response load adjustment of users in different time periods within the region.

[0188] Step 4.4: Issue the scheduling instructions obtained in Step 4.3 to users and new energy sources, namely the output plans of new energy generators and the demand response load adjustment of users at different time periods, and execute and complete the optimization of new energy consumption taking into account the carbon emission responsibility sharing.

[0189] Note: All the above steps can be written in Python 3.7. For easier implementation and more efficient computation, it is recommended to implement them on the TensorFlow 2.2 platform using Keras version 2.3.1.

[0190] This invention also provides a renewable energy consumption optimization scheduling device that considers the sharing of carbon emission responsibilities among multiple stakeholders, comprising:

[0191] The carbon emission responsibility factor acquisition module is used to allocate carbon emission responsibility to users based on their clean energy consumption, and obtain carbon emission responsibility factors.

[0192] The new energy reliability factor acquisition module is used to evaluate the reliability of new energy generators and obtain the new energy reliability factor.

[0193] The optimization model building module is used to build models with different optimization objectives, namely, to achieve distribution network security and stability, regional carbon emission reduction and improve the benefits of each participant. These include distribution network security optimization model, regional carbon emission control optimization model, new energy equipment economic optimization model and power user optimization model. The obtained carbon emission responsibility factor and new energy reliability factor are added as parameters to the objective equation.

[0194] The scheduling plan acquisition module is used to obtain the scheduling plans of new energy power generation and users layer by layer through reinforcement learning based on the models with different optimization objectives, so as to realize the economic and safe consumption of new energy considering the carbon emission reduction responsibility of different entities.

[0195] This invention also provides a renewable energy consumption optimization scheduling system that considers the sharing of carbon emission responsibilities among multiple entities, comprising: a computer-readable storage medium and a processor;

[0196] The computer-readable storage medium is used to store executable instructions;

[0197] The processor is used to read executable instructions stored in the computer-readable storage medium and execute the new energy consumption optimization scheduling method that considers the sharing of carbon emission responsibilities among multiple entities.

[0198] This invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned new energy consumption optimization scheduling method that considers the sharing of carbon emission responsibilities among multiple entities.

[0199] Case Analysis

[0200] This invention validated the proposed model in a system comprising three regions: A, B, and C. The line power loss rate was 0.5%, assuming that line losses within the same region can be ignored, and only losses during inter-regional power transmission are considered. The maximum feeder capacity was 300kW, and the highest electricity price was $2kWh / kWh. The carbon emission density was 0.65kg / kWh. The simulation results are as follows:

[0201] 1. Incentive price and load demand response

[0202] The power consumption in the three regions before and after demand response, after considering the incentive price, is shown in the figure. Figure 4 It is evident that after users adjusted their electricity usage time, the overall power fluctuation and peak-valley difference decreased. Specific user demand response quantities and corresponding incentive electricity prices are as follows: Figure 5 As shown.

[0203] 2. Carbon emission reduction effect

[0204] Depend on Figure 6 It can be seen that after optimization using the method proposed in the patent, the total carbon emissions of the system decrease. The reduction is mainly due to the reduction in power line energy loss during dispatching, the increase in clean and renewable power generation, the reduction in abandoned power, and the reduction in the use of thermal power, all of which contribute to the reduction in carbon emissions.

[0205] 3. Reliability Factor and Long-Term Returns of New Energy Power Generators

[0206] This invention compares the optimized benefits of renewable energy generators with and without considering reliability factors, and the results are as follows: Figure 7 As shown, when considering reliability factors, generator owners experience lower returns in the early stages, but these returns steadily increase in subsequent phases. Overall, good reliability allows generators to reap greater long-term benefits.

[0207] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0208] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0209] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0210] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0211] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing the scheduling of new energy consumption that considers the sharing of carbon emission responsibilities among multiple stakeholders, characterized in that, Includes the following steps: Based on the amount of clean energy used by users, the carbon emission responsibility of users is allocated to them, resulting in a carbon emission responsibility factor. The reliability of new energy power generators is evaluated to obtain the new energy reliability factor; To achieve the goals of ensuring the safety and stability of the distribution network, reducing regional carbon emissions, and improving the benefits of all participants, models with different optimization objectives are built, including a distribution network safety optimization model, a regional carbon emission control optimization model, a new energy equipment economic optimization model, and an optimization model for power users. The carbon emission responsibility factor and the new energy reliability factor are added as parameters to the objective equation. Based on the established models with different optimization objectives, the scheduling plans for new energy power generation and users are obtained layer by layer through reinforcement learning methods, so as to realize the economic and safe consumption of new energy considering the carbon emission reduction responsibility of different entities. The process of allocating carbon emission responsibility to users based on their clean energy consumption to obtain a carbon emission responsibility factor includes the following steps: Based on the inherent characteristics of user nodes and the number of connection branches, determine the carbon emission parameters of each user node. and ,in, and These represent the carbon emission intensity coefficients of branch current and user node injected power, respectively. A carbon emission flow model is introduced, which uses the nodal power on the demand side to calculate the relevant carbon emission intensity and calculates the nodal carbon emissions based on the active power of the load flowing into each node and the carbon emission parameters: ； For user node j at time t, The nodal carbon emission intensity refers to the energy consumption intensity of user loads. Indicates the power flowing through the branch. This represents the amount of power generated by this node, where NU is the total number of users; Based on the principle of carbon emission sharing, the carbon emission reduction ratio of each user node is calculated using the carbon emission amounts of all user nodes within the scope, that is, the proportion of the user's carbon emission reduction to the total user carbon emission reduction: ； in, This refers to user j's carbon emission responsibility factor at time t. This refers to the power consumption of user nodes; Calculate the carbon emission responsibility factor based on user demand response characteristics. To comprehensively evaluate users' contributions during the dispatching process, the responsibility factor considers users' willingness and ability in demand response and carbon emission reduction. It consists of two parts: the first part represents users' efforts to compensate for insufficient renewable energy output by adjusting energy consumption, thereby reducing renewable energy curtailment and power fluctuations; the second part represents users' contribution to reducing carbon emissions. ； ； ； ； For user i at time t, This refers to demand response power; This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption.

2. The renewable energy consumption optimization scheduling method considering the sharing of carbon emission responsibilities among multiple entities as described in claim 1, characterized in that, The process of evaluating the reliability of new energy power generators to obtain a new energy reliability factor includes the following steps: Based on the predicted power output and actual available power output data of new energy sources, the prediction accuracy factor of new energy power output is calculated. : ； Among them, for the new energy generator i at time t, To predict the power output of new energy sources, Let i be the power generation of the renewable energy generator at time t; Calculate the active power output fluctuation factor of the new energy source based on the power output of the new energy source at two adjacent moments: ； Based on the clean energy power generation of new energy sources, and according to their output proportion, calculate the contribution of each new energy power generator to the reduction of carbon emissions: ； in, This refers to the carbon emission reduction intensity factor brought about by renewable energy power generators using clean energy supply. Let i be the power generation of the renewable energy generator at time t; Calculating the reliability factor of renewable energy generators based on the concept of Mahalanobis distance : ； ； in, This represents the column vector of the obtained new energy generator i. It is a column vector of mean. This describes the process of covariance matrix operations.

3. The renewable energy consumption optimization scheduling method considering the sharing of carbon emission responsibilities among multiple entities as described in claim 2, characterized in that, The aforementioned model, with the objectives of achieving distribution network security and stability, regional carbon emission reduction, and improving the benefits of all participants, is constructed with different optimization goals, including the following steps: A distribution network layer security optimization model is established. The distribution network company is responsible for the security of the entire distribution network. The optimization objective is to ensure power balance, handle power fluctuations caused by the uncertainty of new energy sources, and ensure good power quality. The formula of the distribution network layer security optimization model includes two terms: the first term is the impact of active power fluctuations from new energy sources, and the second term is the limitation on unbalanced power output. ； ； ； s.t. ; ； ； The constraints include the balance between load power consumption and renewable energy output at any given time, renewable energy output limits, and the range of load adjustment in response to user demand. A piecewise function representing output power fluctuations. It is a punitive factor that ensures a balance of power. This indicates the upper limit of the output of the new energy power generator i. and These are the upper and lower limits for adjusting the energy load of user j; A regional carbon emission control optimization model is established. At the regional level, the optimization objective of the power trading center is to control carbon emissions, which includes two parts: reducing the curtailment of renewable energy by replacing traditional thermal power with more renewable energy generation; and promoting the local consumption of renewable energy within the region to reduce energy loss and carbon emissions caused by long-distance inter-regional transmission. The regional carbon emission control optimization model is as follows: ； s.t. ; The constraint is the upper and lower limits of the transmission capacity of the transmission feeder. This refers to the maximum permissible power generation capacity of renewable energy generators; It is the line loss factor of line n; It is the power of branch n at time t. and Indicates the power limit of the branch; According to the new energy reliability factor described in step [1] An economic optimization model for new energy equipment is constructed. The resource equipment layer includes two main entities: power generators and electricity users. They adjust transaction prices based on reference electricity prices, participate in market negotiations, and obtain their respective maximum benefits. The optimization objective of new energy power generators is to obtain the maximum net revenue from electricity sales, which is divided into three parts: electricity sales revenue, demand response compensation costs, and the potential value of reliability factors. The economic optimization model for new energy equipment is as follows: ; s.t. ; The constraint is the fluctuation range of the electricity price for electricity generated from new energy sources. This represents the normal bidding price for renewable energy generator i at time t. Indicates demand response electricity price; Based on the calculated carbon emission responsibility factor An optimization model for electricity users is established, with the objective of minimizing electricity costs. It consists of three parts: demand response revenue, electricity costs, and the potential value of return factors. The optimization model for electricity users is as follows: ； in, This represents the revenue price for the j-th user's demand response during time period t. The electricity price for user j during time period t. This represents the estimated power consumption in response to previous plans. This represents the user's actual power consumption, NT represents the total number of response periods, and NU represents the total number of users.

4. The renewable energy consumption optimization scheduling method considering the sharing of carbon emission responsibilities among multiple entities as described in claim 3, characterized in that, The aforementioned model, based on different optimization objectives, uses reinforcement learning to progressively obtain scheduling plans for new energy power generation and users, thereby achieving the economical and secure consumption of new energy while considering the carbon emission reduction responsibilities of different stakeholders. The process includes the following steps: Based on the aforementioned distribution network layer security optimization model, the reinforcement learning Q-learning method is used to first solve the overall security optimization scheduling plan of the distribution network, thereby obtaining the distribution network scheduling plan. The distribution network scheduling plan includes the overall renewable energy output power and total user demand response of the entire network in each time period. Based on the distribution network scheduling plan, and according to the regional carbon emission control optimization model, the regional carbon emission reduction optimization model is solved to obtain the regional scheduling plan. The regional scheduling plan includes the total output power of all new energy sources and the total demand response of all users in each region. Based on the regional dispatch plan, dispatch instructions are obtained according to the economic optimization model of the new energy equipment and the optimization model of the power users. The dispatch instructions include the output plan of each new energy generator and the demand response load adjustment of the users in different time periods within the region. The dispatch instructions are issued to users and new energy sources, namely the output plans of new energy power generators and the demand response load adjustments of users at different times, and the optimization of new energy consumption taking into account the carbon emission responsibility sharing is executed and completed.

5. A renewable energy consumption optimization scheduling device considering the sharing of carbon emission responsibilities among multiple stakeholders, characterized in that, include: The carbon emission responsibility factor acquisition module is used to allocate carbon emission responsibility to users based on their clean energy consumption, and obtain carbon emission responsibility factors. The new energy reliability factor acquisition module is used to evaluate the reliability of new energy generators and obtain the new energy reliability factor. The optimization model building module is used to build models with different optimization objectives, namely, to achieve distribution network security and stability, regional carbon emission reduction and improve the benefits of each participant. These include distribution network security optimization model, regional carbon emission control optimization model, new energy equipment economic optimization model and power user optimization model. The obtained carbon emission responsibility factor and new energy reliability factor are added as parameters to the objective equation. The scheduling plan acquisition module is used to obtain the scheduling plans of new energy power generation and users layer by layer through reinforcement learning based on the models with different optimization objectives, so as to realize the economic and safe consumption of new energy considering the carbon emission reduction responsibility of different entities. The carbon emission responsibility factor acquisition module allocates carbon emission responsibility to users based on their clean energy consumption, resulting in carbon emission responsibility factors, including: Based on the inherent characteristics of user nodes and the number of connection branches, determine the carbon emission parameters of each user node. and ,in, and These represent the carbon emission intensity coefficients of branch current and user node injected power, respectively. A carbon emission flow model is introduced, which uses the nodal power on the demand side to calculate the relevant carbon emission intensity and calculates the nodal carbon emissions based on the active power of the load flowing into each node and the carbon emission parameters: ； For user node j at time t, The nodal carbon emission intensity refers to the energy consumption intensity of user loads. Indicates the power flowing through the branch. This represents the amount of power generated by this node, where NU is the total number of users; Based on the principle of carbon emission sharing, the carbon emission reduction ratio of each user node is calculated using the carbon emission amounts of all user nodes within the scope, that is, the proportion of the user's carbon emission reduction to the total user carbon emission reduction: ； in, This refers to user j's carbon emission responsibility factor at time t. This refers to the power consumption of user nodes; Calculate the carbon emission responsibility factor based on user demand response characteristics. To comprehensively evaluate users' contributions during the dispatching process, the responsibility factor considers users' willingness and ability in demand response and carbon emission reduction. It consists of two parts: the first part represents users' efforts to compensate for insufficient renewable energy output by adjusting energy consumption, thereby reducing renewable energy curtailment and power fluctuations; the second part represents users' contribution to reducing carbon emissions. ； ； ； ； For user i at time t, This refers to demand response power; This represents the estimated power consumption in response to previous plans. This indicates the user's actual power consumption.

6. The new energy consumption optimization scheduling device considering the sharing of carbon emission responsibilities among multiple entities as described in claim 5, characterized in that, The new energy reliability factor acquisition module evaluates the reliability of new energy power generators and obtains new energy reliability factors, including: Based on the predicted power output and actual available power output data of new energy sources, the prediction accuracy factor of new energy power output is calculated. : ； Among them, for the new energy generator i at time t, To predict the power output of new energy sources, Let i be the power generation of the renewable energy generator at time t; Calculate the active power output fluctuation factor of the new energy source based on the power output of the new energy source at two adjacent moments: ； Based on the clean energy power generation of new energy sources, and according to their output proportion, calculate the contribution of each new energy power generator to the reduction of carbon emissions: ； in, This refers to the carbon emission reduction intensity factor brought about by renewable energy power generators using clean energy supply. Let i be the power generation of the renewable energy generator at time t; Calculating the reliability factor of renewable energy generators based on the concept of Mahalanobis distance : ； ； in, This represents the column vector of the obtained new energy generator i. It is a column vector of mean. This describes the process of covariance matrix operations.

7. The new energy consumption optimization scheduling device considering the sharing of carbon emission responsibilities among multiple entities as described in claim 5, characterized in that, The optimization model building module aims to achieve distribution network security and stability, regional carbon emission reduction, and improved benefits for all participants, building models with different optimization objectives, including: A distribution network layer security optimization model is established. The distribution network company is responsible for the security of the entire distribution network. The optimization objective is to ensure power balance, handle power fluctuations caused by the uncertainty of new energy sources, and ensure good power quality. The formula of the distribution network layer security optimization model includes two terms: the first term is the impact of active power fluctuations from new energy sources, and the second term is the limitation on unbalanced power output. ； ； ； s.t. ; ； ； The constraints include the balance between load power consumption and renewable energy output at any given time, renewable energy output limits, and the range of load adjustment in response to user demand. A piecewise function representing output power fluctuations. It is a punitive factor that ensures a balance of power. This indicates the upper limit of the output of the new energy power generator i. and These are the upper and lower limits for adjusting the energy load of user j; A regional carbon emission control optimization model is established. At the regional level, the optimization objective of the power trading center is to control carbon emissions, which includes two parts: reducing the curtailment of renewable energy by replacing traditional thermal power with more renewable energy generation; and promoting the local consumption of renewable energy within the region to reduce energy loss and carbon emissions caused by long-distance inter-regional transmission. The regional carbon emission control optimization model is as follows: ； s.t. ; The constraint is the upper and lower limits of the transmission capacity of the transmission feeder. This refers to the maximum permissible power generation capacity of renewable energy generators; It is the line loss factor of line n; It is the power of branch n at time t. and Indicates the power limit of the branch; According to the new energy reliability factor described in step [1] An economic optimization model for new energy equipment is constructed. The resource equipment layer includes two main entities: power generators and electricity users. They adjust transaction prices based on reference electricity prices, participate in market negotiations, and obtain their respective maximum benefits. The optimization objective of new energy power generators is to obtain the maximum net revenue from electricity sales, which is divided into three parts: electricity sales revenue, demand response compensation costs, and the potential value of reliability factors. The economic optimization model for new energy equipment is as follows: ; s.t. ; The constraint is the fluctuation range of the electricity price for electricity generated from new energy sources. This represents the normal bidding price for renewable energy generator i at time t. Indicates demand response electricity price; Based on the calculated carbon emission responsibility factor An optimization model for electricity users is established, aiming to minimize electricity costs. It consists of three parts: demand response revenue, electricity costs, and the potential value of return factors, including carbon emission responsibility factors. This will affect the user's long-term trading profits, so it is included in the optimization objective. The optimization model for electricity users is as follows: ； in, This represents the revenue price for the j-th user's demand response during time period t. The electricity price for user j during time period t. This represents the estimated power consumption in response to previous plans. This represents the user's actual power consumption, NT represents the total number of response periods, and NU represents the total number of users.

8. The new energy consumption optimization scheduling device considering the sharing of carbon emission responsibilities among multiple entities as described in claim 5, characterized in that, The scheduling plan acquisition module, based on models with different optimization objectives, uses reinforcement learning to progressively obtain scheduling plans for renewable energy generation and users. This achieves the economical and secure absorption of renewable energy while considering the carbon emission reduction responsibilities of different stakeholders, including: Based on the aforementioned distribution network layer security optimization model, the reinforcement learning Q-learning method is used to first solve the overall security optimization scheduling plan of the distribution network, thereby obtaining the distribution network scheduling plan. The distribution network scheduling plan includes the overall renewable energy output power and total user demand response of the entire network in each time period. Based on the distribution network scheduling plan, and according to the regional carbon emission control optimization model, the regional carbon emission reduction optimization model is solved to obtain the regional scheduling plan. The regional scheduling plan includes the total output power of all new energy sources and the total demand response of all users in each region. Based on the regional dispatch plan, dispatch instructions are obtained according to the economic optimization model of the new energy equipment and the optimization model of the power users. The dispatch instructions include the output plan of each new energy generator and the demand response load adjustment of the users in different time periods within the region. The dispatch instructions are issued to users and new energy sources, namely the output plans of new energy power generators and the demand response load adjustments of users at different times, and the optimization of new energy consumption taking into account the carbon emission responsibility sharing is executed and completed.

Images