Power distribution network multi-type flexible resource low-carbon scheduling method based on carbon emission flow

Abstract

The invention discloses a power distribution network multi-type flexible resource low-carbon scheduling method based on carbon emission flow, and relates to the technical field of power system low-carbon scheduling, and the method comprises the steps: constructing a scheduling time period set of a discrete time domain, and building an equivalent lossless network model; power flow calculation is carried out, and a node-branch carbon flow mapping relation is established; according to the carbon emission intensity of the unit, the carbon flow rate and node carbon emission flow of each branch are calculated; performing differential approximate calculation on the node injection power to obtain a carbon marginal coefficient of each node; generating a time-varying carbon price signal of each node in combination with an external carbon price reference; constructing a weighted combination objective function based on the node time-varying carbon price signal; and solving a weighted combination objective function and constraint conditions by adopting a distributed multi-agent collaborative optimization strategy. According to the invention, collaborative scheduling of multiple types of flexible resources under a low-carbon target is realized, and the carbon emission reduction effect is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of low-carbon dispatching of power systems, and particularly relates to a low-carbon dispatching method for multiple types of flexible resources in a distribution network based on carbon emission flow. BACKGROUND

[0002] The power system is evolving from the traditional dispatching mode focusing on economy to the comprehensive dispatching mode considering low carbonization, flexibility and intelligence. The wide access of multiple types of flexible resources such as distributed photovoltaic, energy storage system, electric vehicle and adjustable load in the distribution network makes the system operation characteristics more complex and time-varying. How to realize the low-carbon collaborative optimization dispatching of multiple sources and loads under the premise of ensuring the safe and stable operation of the system has become one of the key research directions. The traditional economic dispatching method of the distribution network usually takes energy balance and cost minimization as the target and only considers economic factors such as fuel cost, electricity price or power loss, without taking into account the difference in carbon emission of power generation sources and the characteristics of carbon transmission path. This method cannot accurately depict the flow rule of carbon emission in the network and is difficult to realize the node-level carbon responsibility identification and price signal transmission in the multiple energy interconnection and distributed scenarios. The existing low-carbon dispatching research is mostly based on the carbon factor of the power generation side or the regional carbon quota constraint, and lacks the research on the dynamic allocation mechanism of carbon emission and the marginal carbon cost driving mechanism at the distribution network level. Especially under the background of collaborative operation of multiple types of flexible resources, there is a lack of quantitative model that can reflect the coupling relationship between energy flow and carbon flow, resulting in a coarse carbon emission control granularity and an optimization dispatching result that is difficult to reflect the real carbon effect.

[0003] Therefore, there is an urgent need for a new low-carbon dispatching method that can realize the traceability of carbon emission flow, the transmission of carbon price signal, and the joint optimization of carbon cost and energy cost at the distribution network level, so as to promote the collaborative operation and economic incentive response of flexible resources under the low-carbon target.

[0004] The present application is based on the above problems and proposes a low-carbon dispatching method for multiple types of flexible resources in a distribution network based on carbon emission flow, which realizes the integrated optimization dispatching decision of low carbon and economy. SUMMARY

[0005] In view of the above problems, the present application is proposed.

[0006] Therefore, the problem to be solved by the present application is how to accurately quantify the carbon emission flow in the distribution network, generate the time-varying carbon price signal of the node, and realize the low-carbon economic collaborative dispatching of multiple types of flexible resources under the premise of meeting the flow and resource time sequence constraints.

[0007] To solve the above technical problems, the present application provides the following technical solutions.

[0008] In a first aspect, the embodiments of the present application provide a multi-type flexible resource low-carbon scheduling method for a power distribution network based on carbon emission flow, which comprises collecting node, branch parameters required for operation of the power distribution network and operation data of various types of distributed power sources, energy storage units and adjustable loads, constructing a discrete time domain scheduling period set, initializing various scheduling decision variables and resource initial states, establishing an equivalent lossless network model, and performing power flow calculation according to node injection power and line parameters to obtain active and reactive power distribution of each node and branch power flow.

[0009] According to the node injection power and the line parameters, the power flow is calculated to obtain the active and reactive power distribution of each node and the branch power flow. The node carbon potential is defined to represent the linear relationship between the node carbon emission flow and the power flow, and the branch carbon flow density is calculated based on the node carbon potential to establish the node-branch carbon flow mapping relationship. According to the carbon emission intensity of the unit, the carbon flow rate of each branch and the node carbon emission flow are calculated.

[0010] The carbon emission flow rate of each node at each time is combined to form a carbon emission flow matrix. For new energy units, an equivalent carbon reduction load processing method is used to maintain the conservation of the total carbon flow rate, and the node power is combined to obtain the node time-varying carbon potential distribution.

[0011] A small active disturbance is introduced to perform a difference approximation calculation on the node injection power to obtain the carbon marginal coefficient of each node. Combined with the external carbon price benchmark, a time-varying carbon price signal of each node is generated.

[0012] Based on the time-varying carbon price signal of the node, a weighted combination objective function including economic cost, carbon emission cost and service cost is constructed in the discrete time domain.

[0013] The flow constraint and the resource time sequence constraint are constructed, and the carbon price signal and the Lagrange multiplier coupling mechanism are introduced to synchronize the response cost and the carbon target. A distributed multi-agent collaborative optimization strategy is used to solve the weighted combination objective function and the constraint condition.

[0014] As a preferred scheme of the multi-type flexible resource low-carbon scheduling method for the power distribution network based on carbon emission flow, the discrete time domain scheduling period set is constructed, the scheduling decision variables and the resource initial states are initialized, and the equivalent lossless network model is established, which comprises: The scheduling time window is discretized according to the predetermined period length Δt to form a scheduling period set T={t1,t2,…,tN}, wherein the Δt is selected as 1h, 15min control step length according to the scene, and switching rules are given between different levels.

[0015] The equivalent lossless network model is established as a benchmark for carbon emission flow calculation, and a construction process of the equivalent lossless network model comprises: solving a real power flow based on original node injection and network parameters at each time period, and counting actual active power loss of each branch and total system loss; according to a predetermined loss allocation rule, the total system loss is distributed to each generator to obtain equivalent injection values of the generators; the power flow is re-solved on the lossless model by using the distributed equivalent power generation injection and the original load vector, the equivalent model is verified to satisfy injection conservation and make key power flow error within a preset threshold; if the error is out of limit, the allocation weight is adjusted or local correction is adopted and correction information is recorded; on the equivalent lossless network, the equivalent injection of each generator is multiplied by the carbon intensity of the unit to obtain the power generation side carbon flow rate.

[0016] As a preferred scheme of the low-carbon scheduling method for multiple types of flexible resources in the power distribution network based on carbon emission flow, wherein: according to the node injection power and the line parameters, the power flow is solved to obtain the active and reactive power distribution of each node and the branch flow of each branch, including:

[0017] For each discrete time period t, the AC power flow is used to solve the power flow of the whole network by taking the node injection vector, the network topology and the line parameters as inputs, to obtain the active and reactive injection distribution of each node and the active and reactive power flow and line loss of each branch;

[0018] Based on the equivalent lossless injection of each generator at time period t and the real-time carbon emission intensity of the unit, the power generation side carbon flow rate of each generator at time period t is calculated as the carbon flow source input.

[0019] As a preferred scheme of the low-carbon scheduling method for multiple types of flexible resources in the power distribution network based on carbon emission flow, wherein: based on the node carbon potential, the branch carbon flow density is calculated, and the node-branch carbon flow mapping relationship is established; according to the carbon emission intensity of the unit, the carbon flow rate of each branch and the node carbon emission flow are calculated, including:

[0020] The node carbon potential of each node in the network at a given time period is determined by using the upstream branch carbon flow information and the local equivalent power generation injection carbon flow of the node in sequence, taking the power flow direction as the propagation direction;

[0021] For each branch from node i to node j in the network, the carbon flow density of the branch is set as the node carbon potential of the starting node i, and the mapping relationship of the node-branch carbon flow is established; the node carbon potential is expressed as: ;

[0022] Wherein, is the set of upstream nodes connected to node t and having power flow injection, and s=t is the carbon emission flow density of the upstream branch of node t. P(t) is the active power flow through the node t, s=t is the branch number;

[0023] Based on the branch carbon flow density and the active power flow of the corresponding branch, the carbon flow rate of each branch in the time period and its cumulative carbon flow are calculated;

[0024] All generators are converted into power generation side carbon flow rate according to equivalent lossless injection and respective carbon intensity, and combined with branch carbon flow distribution, to obtain node carbon emission flow distribution by time period according to node.

[0025] As a preferred scheme of the low-carbon scheduling method for multi-type flexible resources of the power distribution network based on carbon emission flow of the application, wherein: the carbon emission flow rate of each node at each time of the system is composed into a carbon emission flow matrix; for new energy units, an equivalent carbon reduction load processing method is adopted to maintain the conservation of the total carbon flow rate of the system, and the node time-varying carbon potential distribution is obtained by combining the node power, including:

[0026] The carbon emission flow rate results of the system at all discrete times and each node are arranged to form a carbon emission flow matrix; each row of the matrix corresponds to a node, each column corresponds to a time, and the matrix element represents the carbon flow rate of the node at the time;

[0027] For new energy units connected to the system, an equivalent carbon reduction load processing method is adopted to offset part of the carbon emission flow generated by the upstream thermal power unit by reducing the carbon flow injection of the equivalent node;

[0028] The actual load of the system node is composed of two parts of ordinary load and carbon reduction load, the ordinary load represents the actual electricity demand, and the carbon reduction load represents the emission reduction contribution of new energy; by combining the two types of loads, the equivalent combined power of the node is obtained, and the new node carbon potential distribution of the system is calculated accordingly.

[0029] As a preferred scheme of the low-carbon scheduling method for multi-type flexible resources of the power distribution network based on carbon emission flow of the application, wherein: a small active disturbance is introduced, the injected power of the node is calculated by difference approximation, and the carbon marginal coefficient of each node is obtained; combined with the external carbon price benchmark, the time-varying carbon price signal of each node is generated, including:

[0030] Under the premise of keeping the system topology structure and line parameters unchanged, a small active power disturbance is injected into the selected node, the total carbon emission of the system before and after the disturbance is obtained by recalculating the power flow distribution and the corresponding carbon emission flow matrix; according to the difference relationship between the total carbon emission change of the system and the energy increment of the node, the carbon marginal coefficient of each node is calculated;

[0031] In combination with the externally published carbon price benchmark, the carbon marginal coefficient of the node is multiplied by the carbon price benchmark to generate a time-varying carbon price signal corresponding to the node.

[0032] As a preferred scheme of the low-carbon scheduling method for multiple types of flexible resources in a power distribution network based on carbon emission flow, the time-varying carbon price signal of a node is used to construct a weighted combination target function including economic cost, carbon emission cost and service cost in a discrete time domain, including:

[0033] According to the obtained time-varying carbon price signal of the node, the scheduling period is divided into multiple discrete time periods in combination with the operating state of the power distribution system and the load prediction result.

[0034] For each time period, the output power and carbon emission of each unit, energy storage unit and load node in the system are calculated, and a comprehensive optimization target function of the system is established in a discrete time domain.

[0035] As a preferred scheme of the low-carbon scheduling method for multiple types of flexible resources in a power distribution network based on carbon emission flow, the time-varying carbon price signal of a node is used to construct a weighted combination target function including economic cost, carbon emission cost and service cost in a discrete time domain, including:

[0036] The node carbon price signal is coupled with the Lagrange multiplier to form a node-level carbon cost adjustment factor.

[0037] The power distribution network is divided into multiple functional agents, including a distributed power supply agent, an energy storage agent, a controllable load agent and a network agent. Each agent independently makes optimization decisions for economic cost and carbon cost within a local range, and interacts and coordinates prices based on Lagrange multiplier information.

[0038] Through multiple rounds of iteration and update, a collaborative consistent solution under global constraints is achieved, and the optimal system-level weighted combination target function is achieved.

[0039] In a second aspect, an embodiment of the present application provides a computer device, including a memory and a processor, the memory storing a computer program, wherein the computer program instructions are executed by the processor to realize the steps of the low-carbon scheduling method for multiple types of flexible resources in a power distribution network based on carbon emission flow according to the first aspect of the present application.

[0040] In a third aspect, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program instructions are executed by the processor to realize the steps of the low-carbon scheduling method for multiple types of flexible resources in a power distribution network based on carbon emission flow according to the first aspect of the present application.

[0041] The beneficial effects of this invention are as follows: This invention achieves dual-objective coordinated scheduling of cost and carbon emissions in distributed power systems by using a weighted objective function solution method based on carbon price signals and multi-agent collaborative optimization. This can significantly reduce the total carbon emissions of the system and improve energy utilization efficiency and scheduling intelligence level while ensuring the safe operation of the network. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Fig. 1 A flowchart of a low-carbon scheduling method for multiple flexible resources in a distribution network based on carbon emission flows; Fig. 2 This is a computer equipment diagram for a low-carbon scheduling method for multiple types of flexible resources in a distribution network based on carbon emission flows. Detailed Implementation

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0045] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0047] Example 1

[0048] Reference Figs. 1-2 This is the first embodiment of the present invention, which provides a method for low-carbon scheduling of multiple types of flexible resources in a distribution network based on carbon emission flows, including: S1: Collect node and branch parameters required for the operation of the distribution network, as well as the operation data of various distributed power sources, energy storage units and adjustable loads, construct a set of scheduling time periods in the discrete time domain, initialize each scheduling decision variable and the initial state of resources, and establish an equivalent lossless network model.

[0049] Furthermore, the scheduling time window is discretized according to the predetermined time period length Δt to form a scheduling time period set T={t1,t2,…,tN}, wherein Δt is selected as a control step size of 1 hour or 15 minutes according to the scenario, and switching rules are given between different levels;

[0050] An equivalent lossless network model is established as the benchmark for carbon emission flow calculation. The construction process of the equivalent lossless network model includes: solving the actual power flow based on the original node injection and network parameters at each time period, and calculating the actual active power loss of each branch and the total system loss; allocating the total system loss to each generator according to a predetermined loss allocation rule to obtain the equivalent injection value of each generator; resolving the power flow on the lossless model using the allocated equivalent power generation injection and the original load vector, verifying that the equivalent model satisfies injection conservation and that the key power flow error is within a preset threshold; if the error exceeds the limit, adjusting the allocation weight or using local correction and recording the correction information; multiplying the equivalent injection of each generator by the unit carbon intensity on the equivalent lossless network to obtain the carbon flow rate on the power generation side.

[0051] It should be noted that, in the lossless equivalent model, the carbon emission flow rate of the computer group is calculated as follows:

[0052] Unit carbon intensity determination: Based on the carbon emission intensity parameters corresponding to the power generation type (thermal power, gas-fired power, photovoltaic power, wind power, etc.) and its unit power generation, a unit carbon emission coefficient database is established. Generator-side carbon flow calculation: For each generator, the carbon flow rate of that node is obtained by multiplying its equivalent active power injection by the unit carbon intensity. The carbon flow rate characterizes the carbon emission intensity injected into the network by that node per unit time, providing basic data for subsequent node carbon potential distribution calculations.

[0053] S2: Based on the injected power at the nodes and the line parameters, power flow calculation is performed to obtain the active and reactive power distribution of each node and the power flow of each branch; the node carbon potential is defined to characterize the linear relationship between the node carbon emission flow and the power flow, and the branch carbon flow density is calculated based on the node carbon potential to establish the node-branch carbon flow mapping relationship; the carbon flow rate of each branch and the node carbon emission flow are calculated according to the carbon emission intensity of the unit.

[0054] Furthermore, for each discrete time period t, using the node injection vector, network topology, and line parameters as input, AC power flow is used to solve the power flow of the entire network, obtaining the active and reactive power injection distribution of each node, as well as the active and reactive power flow and line loss of each branch.

[0055] Based on the equivalent lossless injection of each generator in time period t and the real-time carbon emission intensity of the unit, the carbon flow rate of each generator on the power generation side in time period t is calculated as the carbon flow source input.

[0056] Furthermore, using the tidal current direction as the propagation direction, the carbon flow information of the upstream branch and the carbon flow injected by the local equivalent power generation of this node are used for sequential calculation to determine the node carbon potential of each node in the network within a given time period.

[0057] For each branch in the network flowing from node i to node j, the carbon flow density of that branch is set to the node carbon potential of its originating node i, establishing a mapping relationship between node and branch carbon flow; the node carbon potential Represented as: ;

[0058] in, Let ρ be the set of branches containing upstream nodes connected to node t and having power flow injection. s=t Let be the carbon emission flow density of the upstream branch of node t. For the active power flow passing through node t, s = t represents the carbon flow rate through the corresponding node t, where s = t is the branch number;

[0059] Based on the carbon flow density of the branch and the active power flow of the corresponding branch, calculate the carbon flow rate and cumulative carbon flow of each branch during the time period.

[0060] All generators are converted into carbon flow rate on the power generation side by equivalent lossless injection and their respective carbon intensity, and combined with the branch carbon flow distribution, the node carbon emission flow distribution by time period is obtained by summing them up by node.

[0061] S3: The carbon emission flow rate of each node at each time point in the system is used to form a carbon emission flow matrix; for new energy units, the equivalent carbon reduction load is used to maintain the total carbon flow rate of the system, and the node power is combined to obtain the time-varying carbon potential distribution of the nodes.

[0062] Furthermore, the carbon emission flow rate results of the system at all discrete time points and at each node are organized to form a carbon emission flow matrix; each row of the matrix corresponds to a node, each column corresponds to a time point, and the matrix elements represent the carbon flow rate of that node at that time point.

[0063] For the new energy units connected to the system, an equivalent carbon reduction load treatment method is adopted, which reduces the carbon flow injection at the equivalent node to offset part of the carbon emission flow generated by the upstream thermal power units.

[0064] The actual load of a system node consists of two parts: ordinary load and carbon reduction load. Ordinary load represents the actual electricity demand, while carbon reduction load represents the emission reduction contribution of new energy sources. By merging the two types of loads, the equivalent combined power of the node is obtained, and the new node carbon potential distribution of the system is calculated accordingly.

[0065] It should be noted that for the wind power, photovoltaic and other new energy units connected to the system, this embodiment adopts the modeling concept of "equivalent carbon reduction load", that is, the output of new energy is regarded as a negative contribution to the carbon flow injection of the system, and the carbon emission flow injection of the equivalent node is reduced to offset part of the carbon emission flow of the upstream thermal power unit.

[0066] This modeling approach does not alter the power balance of the electricity flow; it only modifies carbon injection at the carbon emission flow level, thereby achieving low-carbon redistribution under the condition of carbon flow conservation in the system. In this implementation, renewable energy units are modeled as "negative carbon sources," meaning their equivalent injected power is assigned a negative carbon emission rate. The total load of nodes in the system can then be represented as: ordinary load portion: representing the actual electricity demand of users; carbon reduction load portion: representing the emission reduction contribution of renewable energy. In this way, the carbon emission reduction effect of renewable energy can be accurately reflected in the network topology, quantifying its offsetting effect in the carbon flow transmission path. Based on the above equivalent processing of renewable energy, the ordinary load and carbon reduction load at the node are merged to obtain the equivalent merged power of the node. This power reflects both the actual electricity demand of the node and the local carbon emission reduction effect brought about by the access of renewable energy. Then, in the equivalent lossless network model of the system, the node carbon potential is recalculated with the new equivalent node power distribution. The carbon potential recalculation process includes: adjusting the node carbon flow injection amount based on the equivalent combined power of the nodes; updating the carbon flow transfer path between nodes according to the power flow direction; and comparing the differences in node carbon potential before and after the calculation correction to analyze the impact of new energy sources on the spatial distribution of system carbon potential. Through carbon potential reconstruction, the carbon potential of the output nodes of thermal power units in the system will be relatively reduced, while the carbon potential of the load nodes in areas with dense new energy access will be correspondingly increased, reflecting the emission reduction role of new energy sources in regional carbon flow balance.

[0067] S4: Introduce a small active power disturbance and perform differential approximation calculation on the injected power of the nodes to obtain the carbon marginal coefficient of each node; combine with the external carbon price benchmark to generate the time-varying carbon price signal of each node.

[0068] Furthermore, while keeping the system topology and line parameters unchanged, a small active power disturbance is injected into selected nodes. By recalculating the power flow distribution and the corresponding carbon emission flow matrix, the total carbon emissions of the system before and after the disturbance are obtained. Based on the difference between the change in total carbon emissions of the system and the energy increment of the nodes, the carbon marginal coefficient of each node is calculated.

[0069] By combining externally published carbon price benchmarks, the carbon marginal coefficient of a node is multiplied by the carbon price benchmark to generate a time-varying carbon price signal for that node.

[0070] It should be noted that this embodiment first establishes a power flow calculation model corresponding to the scheduling time window. The system topology consists of a node set N={1,2,...,n} and a branch set L={l1,l2,...,l m The structure is as follows: The injected power vector of each node is represented as:

[0071] P i =P G,i -P L,i ;

[0072] Among them, P G,i For node power generation injection, P L,i For node load;

[0073] At each discrete time t∈T, the AC power flow is solved using the node injection vector and branch parameters to obtain the voltage phase angle, power injection distribution, and active and reactive power flow of each node.

[0074] While keeping the system topology and line parameters unchanged, a small active power disturbance ΔP is injected into selected node i. i Its value ranges from 0.1% to 1% of the system's rated capacity.

[0075] Furthermore, this embodiment feeds back the time-varying carbon price signal of nodes to the system optimization scheduling model. By introducing carbon cost weights into power flow constraints and generator output constraints, it achieves simultaneous optimization of carbon emission targets and operational economics. When the carbon marginal coefficient of a node is high, its carbon price signal also increases synchronously, thereby suppressing the output tendency of high-emission nodes. For nodes connected to new energy or low-carbon sources, the carbon marginal coefficient approaches zero or is negative, and its carbon price signal can be transformed into a carbon incentive signal, realizing the dynamic quantification of the value of carbon emission reduction resources.

[0076] S5: Based on the node time-varying carbon price signal, a weighted combined objective function including economic cost, carbon emission cost and service cost is constructed in the discrete time domain.

[0077] Furthermore, based on the obtained node time-varying carbon price signals, combined with the operating status of the power distribution system and load forecast results, the scheduling cycle is divided into multiple discrete time periods.

[0078] For each time period, the power output and carbon emissions of each unit, energy storage unit and load node in the system are calculated, and a comprehensive optimization objective function of the system is established in the discrete time domain.

[0079] It should be noted that the system first uses the obtained nodal time-varying carbon price signal, combined with the real-time operating status of the power distribution system and future load forecasts, to discretize the scheduling cycle. The scheduling cycle is typically 24 hours or 48 hours, and is divided into multiple discrete time periods according to the control granularity:

[0080] T = {t1, t2, ..., tN}, and the distance step length Δt can be selected as 15 min, 30 min, or 1 h according to the system operating characteristics and scheduling level.

[0081] Furthermore, this embodiment employs a distributed multi-agent optimization algorithm for solving the problem. Each node agent calculates carbon cost and power allocation locally and coordinates network power flow constraints using Lagrange multipliers. The node carbon price signal serves as a local optimization weight at each time period, achieving synchronous response between economic cost and carbon objectives. When the system detects drastic fluctuations in the carbon price signal, it can trigger a dynamic adjustment strategy to optimize the carbon cost weight, achieving flexible temporal adjustment of the carbon signal.

[0082] S6: Construct power flow constraints and resource timing constraints, introduce carbon price signals and Lagrange multiplier coupling mechanism to synchronize response costs and carbon targets; adopt a distributed multi-agent collaborative optimization strategy to solve the weighted combined objective function and constraints.

[0083] Furthermore, the node carbon price signal is coupled with the Lagrange multiplier to form a node-level carbon cost adjustment factor; the distribution network is divided into multiple functional intelligent agents, including distributed power generation intelligent agents, energy storage intelligent agents, adjustable load intelligent agents and network agent intelligent agents; each intelligent agent independently makes optimization decisions on economic cost and carbon cost within its local scope, and interacts and coordinates prices based on Lagrange multiplier information.

[0084] Through multiple rounds of iterative updates, a coordinated and consistent solution under global constraints is achieved, reaching the optimal system-level weighted combination objective function.

[0085] It should be noted that in the carbon signal-driven distribution network dispatching process, the time-varying carbon price signal P of each node in the system is first obtained. carbon,i (t), which reflects the marginal cost of carbon at a node and its dynamic relationship with the carbon flow distribution of the system.

[0086] Subsequently, the nodal carbon price signal is compared with the system Lagrange multiplier λ. i (t) are coupled to form a node-level carbon cost adjustment factor:

[0087] Θ i (t)=f(P carbon,i (t),λ i (t));

[0088] This adjustment factor comprehensively characterizes the node's overall response under power balance constraints and carbon target constraints, guiding the local agent to balance economic benefits and carbon emission constraints during optimization; where λ i(t) represents the Lagrange multiplier corresponding to the power balance constraint of the distribution network, reflecting the marginal price signal of system power coupling. By weighted fusion of the carbon price signal and the multiplier, the node regulation has dual feedback characteristics.

[0089] To achieve distributed coordinated scheduling, this embodiment divides the distribution network into several functional intelligent agents, including: a distributed power generation intelligent agent: responsible for optimizing the output of local distributed photovoltaic, wind power, and small gas turbine units, and dynamically adjusting the power generation plan in conjunction with carbon adjustment factors, so that the power generation output can be economical while suppressing the operating time of units with high carbon intensity; an energy storage intelligent agent: responsible for optimizing the charging and discharging strategy of local energy storage devices, prioritizing discharging during periods of high carbon price, and charging during periods of low carbon price or high carbon intensity, to achieve the effect of "carbon reduction and valley filling"; energy storage optimization considers state of charge, power limits, energy balance, and lifetime cost;

[0090] Adjustable load agent: Responsible for adjusting flexible and transferable loads, participating in carbon signal response by adjusting operating time or power level; when the node carbon price is high, DL-Agent postpones the operation of high-energy-consuming equipment to achieve indirect carbon emission reduction; Network agent agent: Responsible for power flow constraints and Lagrange multiplier updates between nodes, maintaining global power and carbon flow balance; This agent coordinates information exchange between local agents to ensure that node voltage and line power flow are within safe ranges.

[0091] During the iteration process, if the overall power balance deviation, node carbon flow imbalance, and total cost change rate are all less than the preset thresholds, the algorithm is considered to have converged. At this point, the power allocation, energy storage strategy, adjustable load regulation, and carbon price signal of each node all reach steady-state consistency, achieving a dynamic balance between global economic efficiency and low carbon emissions.

[0092] This embodiment also provides a computer device applicable to the case of a multi-type flexible resource low-carbon scheduling method for distribution networks based on carbon emission flows, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the multi-type flexible resource low-carbon scheduling method for distribution networks based on carbon emission flows proposed in the above embodiment.

[0093] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0094] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the low-carbon scheduling method for multiple types of flexible resources in the distribution network based on carbon emission flows as proposed in the above embodiments.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A low-carbon scheduling method for multi-type flexible resources of a power distribution network based on carbon emission flow, characterized in that, The application relates to a method for scheduling a power grid, and belongs to the field of power grid scheduling. The method comprises the following steps: collecting node, branch parameters required for power grid operation and operation data of various distributed power sources, energy storage units and adjustable loads, constructing a discrete time domain scheduling period set, initializing various scheduling decision variables and resource initial states, and establishing an equivalent lossless network model; performing power flow calculation according to node injection power and line parameters to obtain active and reactive power distribution of each node and branch power flow; defining a node carbon potential to represent the linear relationship between node carbon emission flow and power flow, and calculating branch carbon flow density based on the node carbon potential; and calculating carbon flow rate of each branch and node carbon emission flow according to the carbon emission intensity of a unit; composing carbon emission flow matrixes of each node at each time point; for new energy units, an equivalent carbon reduction load processing mode is adopted to maintain the conservation of the total carbon flow rate, and node time-varying carbon potential distribution is obtained by combining node power; introducing a small active disturbance to perform differential approximation calculation on node injection power, and obtaining carbon marginal coefficients of each node; and combining an external carbon price benchmark to generate time-varying carbon price signals of each node; based on the time-varying carbon price signals of the nodes, a weighted combination target function including economic cost, carbon emission cost and service cost is constructed in the discrete time domain; 2. The low-carbon dispatching method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that, flow constraints and resource time sequence constraints are constructed, a carbon price signal and a Lagrange multiplier coupling mechanism are introduced to synchronously respond to the cost and carbon target, and a distributed multi-agent collaborative optimization strategy is adopted to solve the weighted combination target function and the constraint conditions. The method for scheduling a power grid comprises the following steps: discretize the scheduling time window according to a predetermined time period length Delta t to form a scheduling period set T={t1, t2, …, tN}, wherein the Delta t is selected as a control step length of 1 h or 15 min according to a scene, and switching rules are given between different levels; 3. The low-carbon dispatching method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that: an equivalent lossless network model is established as a benchmark for carbon emission flow calculation, and the construction process of the equivalent lossless network model comprises the following steps: real power flow is solved based on original node injection and network parameters in each period, and actual active power loss of each branch and total system loss are counted; the total system loss is distributed to each power generator according to a predetermined loss allocation rule to obtain equivalent injection values of the power generators; the power flow is re-solved on a lossless model by using the distributed equivalent power generation injection and the original load vector, the equivalent model is verified to meet the injection conservation and make the key power flow error within a preset threshold; if the error is out of limit, the allocation weight is adjusted or local correction is adopted and correction information is recorded; on the equivalent lossless network, the carbon flow rate of each power generator is obtained by multiplying the equivalent injection of each power generator by the carbon intensity of the power generator. The method for scheduling a power grid comprises the following steps: for each discrete period t, AC power flow is solved by taking the node injection vector, network topology and line parameters as inputs to obtain active and reactive injection distribution of each node and active and reactive power flow and line loss of each branch. The power generation side carbon flow rate of each generator at time period t is calculated as a carbon flow source input based on the equivalent lossless injection of each generator at time period t and the real-time carbon emission intensity of the unit.

4. The low-carbon scheduling method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that: The branch carbon flow density is calculated based on the node carbon potential, and the node-branch carbon flow mapping relationship is established. The carbon flow rate of each branch and the node carbon emission flow are calculated according to the carbon emission intensity of the unit, including: The node carbon potential of each node in the network within a given time period is determined by sequentially calculating the upstream branch carbon flow information and the local equivalent power generation injection carbon flow of the node with the power flow direction as the propagation direction. For each branch in the network flowing from node i to node j, the carbon flow density of the branch is set to the node carbon potential of the start node i, and a mapping relationship between the node-branch carbon flow is established; the node carbon potential is represented as: ； wherein, is the set of branches connected to node t with power injection, p s=t is the carbon emission flow density of the upstream branch of node t, is the active power flow through node t, is the carbon flow rate through the corresponding node t, s = t is the branch number; The carbon flow rate and cumulative carbon flow of each branch within the time period are calculated based on the branch carbon flow density and the active power flow of the corresponding branch. All generators are converted into power generation side carbon flow rates according to their equivalent lossless injection and carbon intensity, and combined with the branch carbon flow distribution to obtain the node carbon emission flow distribution by time period.

5. The low-carbon scheduling method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that, The carbon emission flow matrix is formed by combining the carbon emission flow rates of each node at each time of the system. For new energy units, the equivalent carbon reduction load is used to maintain the total carbon flow rate of the system, and the node time-varying carbon potential distribution is obtained by combining the node power, including: The carbon emission flow matrix is formed by organizing the carbon emission flow rate results of each node at all discrete time points of the system; each row of the matrix corresponds to a node, each column corresponds to a time, and the matrix element represents the carbon flow rate of the node at that time; For new energy units connected to the system, the equivalent carbon reduction load is used to offset part of the carbon emission flow generated by upstream thermal power units by reducing the carbon flow injection of the equivalent node; The actual load of the system node is composed of ordinary load and carbon reduction load, the ordinary load represents the actual electricity demand, and the carbon reduction load represents the emission reduction contribution of new energy; by combining the two types of loads, the equivalent combined power of the node is obtained, and the new node carbon potential distribution of the system is calculated accordingly.

6. The low-carbon scheduling method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that: The carbon marginal coefficient of each node is obtained by introducing a small active power disturbance and performing a difference approximation calculation on the node injection power. Combined with the external carbon price benchmark, the time-varying carbon price signal of each node is generated, including: Under the premise of keeping the system topology and line parameters unchanged, a small active power disturbance is injected into the selected node, and the total carbon emission before and after the disturbance is obtained by recalculating the power flow distribution and the corresponding carbon emission flow matrix; according to the difference relationship between the total carbon emission change and the node energy increment, the carbon marginal coefficient of each node is calculated; Combined with the externally published carbon price benchmark, the carbon marginal coefficient of the node is multiplied by the carbon price benchmark to generate the time-varying carbon price signal corresponding to the node.

7. The low-carbon dispatching method for multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that: Based on the node time-varying carbon price signal, a weighted combination objective function including economic cost, carbon emission cost and service cost is constructed in the discrete time domain, including: According to the obtained node time-varying carbon price signal, the dispatching period is divided into multiple discrete time periods based on the operation state and load prediction results of the distribution system; For each time period, the output power and carbon emission of each unit, energy storage unit and load node in the system are calculated, and the comprehensive optimization objective function of the system is established in the discrete time domain. 8.The low-carbon dispatching method of multi-type flexible resources of power distribution network based on carbon emission flow according to claim 1, characterized in that: The construction of the tide constraint and the resource timing constraint, the introduction of carbon price signal and the coupling mechanism of Lagrange multiplier synchronous response cost and carbon target; Adopting distributed multi-agent collaborative optimization strategy, solving the weighted combination objective function and constraint conditions, including: Coupling the node carbon price signal with the Lagrange multiplier to form the node-level carbon cost adjustment factor; Dividing the distribution network into multiple functional agents, including distributed power supply agent, energy storage agent, adjustable load agent and network agent; each agent independently makes optimization decision of economic cost and carbon cost in local range, and interacts and coordinates price based on the Lagrange multiplier information; Through multiple rounds of iteration update, realizing the collaborative consistent solution under the global constraint, and achieving the optimal system-level weighted combination objective function. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the low-carbon scheduling method of the multi-type flexible resource of the distribution network based on the carbon emission flow according to any one of claims 1-8.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the low-carbon scheduling method of the multi-type flexible resource of the distribution network based on the carbon emission flow according to any one of claims 1-8.

Images