A distribution network dispatching method based on carbon potential-load data bidirectional feedback mechanism

Abstract

The present invention discloses a distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism, which is used to coordinate and realize low-carbon economic operation at the distribution network level. Based on the carbon emission flow theory, the method constructs a load and energy storage carbon emission flow model at the distribution network level, establishes a demand response priority function to evaluate the demand response capability of various loads, and designs a corresponding demand response strategy based on the priority. In combination with the economy and low carbon nature of the distribution network, a bidirectional feedback mechanism of upper-layer carbon potential data and lower-layer load data is designed, and a double-layer low-carbon dispatching model of the distribution network is established and solved based on the mechanism. The present invention can provide a reference for realizing a low-carbon economic operation strategy at the distribution network level, thereby providing support for the construction of a new power system under the "dual carbon" goal, and has certain engineering use value.

Description

Technical Field

[0001] The invention relates to a distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism, belonging to the technical field of distribution system dispatching. Background Art

[0002] Carbon emissions from the power system account for more than 40% of total carbon emissions. Therefore, it has become an inevitable trend for power system dispatch to develop in the direction of low carbonization. In this context, the demand for low carbonization of power system dispatching and operation has become more prominent.

[0003] The carbon emissions of the power system mainly come from the units that rely on traditional fossil energy on the power generation side, so the responsibility for carbon emissions is generally attributed to the power generation side. However, electricity is transported along the power network and ultimately consumed on the load side. Therefore, the load side that participates in electricity consumption should also be responsible for carbon emissions. When evaluating carbon emissions in the power system, the carbon emission flow theory can comprehensively consider the carbon emissions of power generation companies in various links such as production, transportation and consumption. However, traditional carbon emission research mainly focuses on the transmission network, while the distribution network, as a carrier connecting the transmission network and end users, is responsible for transmitting electricity to the terminal load, and the power transmission loss in the distribution network cannot be ignored. Therefore, research on carbon emissions in the distribution network is also crucial.

[0004] As demand-side resources in the power system continue to expand, in-depth research on demand response technology has attracted widespread attention, and making full use of demand-side resources to reduce carbon emissions has become the focus of current research. However, traditional demand response models do not consider the differences in the response capabilities of various loads, and do not guide users to respond in a more orderly manner through differentiated compensation and incentive policies. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism, which can effectively improve the photovoltaic absorption situation of the distribution network, reduce the operating cost of the distribution network, and reduce carbon emissions, and can provide a reference for realizing a low-carbon economic operation strategy at the distribution network level.

[0006] The present invention adopts the following technical solutions to solve the above technical problems:

[0007] A distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism comprises the following steps:

[0008] Step 1: Based on the basic theory of carbon emission flow, the carbon flow density index is introduced to trace the carbon emission responsibility on the power generation side to the user side, and further establish the carbon emission flow model of load and energy storage;

[0009] Step 2: Establish a load demand response priority function to evaluate the demand response capabilities of various loads, and design corresponding demand response strategies from both price and load aspects based on the priority.

[0010] Step 3: Based on steps 1 and 2, a two-way feedback mechanism for the upper-level carbon potential data and the lower-level load data is designed. Based on the two-way feedback mechanism, a two-layer low-carbon dispatching model for the distribution network is established and solved to obtain the final dispatching result.

[0011] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0012] 1. In order to improve the economy of distribution network operation, reduce carbon emissions, and enhance the rationality and orderliness of load demand response, the present invention is based on the carbon emission flow theory in the distribution network, transfers carbon emissions from the source side to the load side, and effectively reduces the carbon emissions of distribution network operation by adjusting the electricity consumption behavior on the load side and the working state of energy storage.

[0013] 2. The present invention designs a load demand response priority, and establishes a demand response model based on the priority, guiding users to respond in a more orderly manner through differentiated compensation and incentive policies.

[0014] 3. The low-carbon dispatching method for distribution networks based on a two-way data feedback mechanism established by the present invention can take into account both the economy and low-carbon nature of distribution network operation. By alternately feeding back and updating the calculation of the system carbon potential and load power consumption behavior, it effectively improves the photovoltaic absorption rate in the distribution network and reduces the carbon emissions of the distribution network, which can provide a reference for the low-carbon operation of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of the distribution network dispatching method based on the carbon potential-load data bidirectional feedback mechanism of the present invention;

[0016] Figure 2 It is an improved IEEE-33 node distribution network topology diagram;

[0017] Figure 3 It is the predicted power diagram of photovoltaic and load;

[0018] Figure 4 is the per unit value of load power;

[0019] Figure 5 These are the scheduling results of four scenarios, where (a) is scenario 1, (b) is scenario 2, (c) is scenario 3, and (d) is scenario 4;

[0020] Figure 6 are the three types of load dispatch results under different scenarios, where (a) is the residential load demand response result, (b) is the commercial load demand response result, and (c) is the industrial load demand response result;

[0021] Figure 7 These are the scheduling results of high-priority large users in different scenarios, where (a) is the DU scheduling result of scenario three, and (b) is the DU scheduling result of scenario four. DETAILED DESCRIPTION

[0022] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0023] The present invention proposes a distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism, and the specific steps are as follows:

[0024] Step 1: Based on the basic theory of carbon emission flow, the basic indicators of carbon flow rate and carbon flow density are introduced, the calculation method and characteristics of the carbon emission flow of the distribution network are explained, and the carbon emission flow model of load and energy storage is further established.

[0025] Step 101: Based on the basic theory of carbon emission flow, the basic indicators of carbon flow rate and carbon flow density are introduced, and the calculation method and characteristics of the carbon emission flow of the distribution network are explained, as follows:

[0026] Through the carbon emission flow theory, the carbon emission responsibility on the power generation side can be traced back to the user side, thus realizing "source-load interaction" in the power system.

[0027] 1) Carbon flow rate

[0028] Carbon flow rate is defined as: Carbon flow rate refers to the amount of carbon emissions transmitted through a branch or node with the power flow per unit time. It is represented by the symbol R and the unit is tCO2 / h.

[0029]

[0030] Where F represents the cumulative carbon flow through a branch or node, the unit is tCO2, and t is time.

[0031] 2) Carbon flow density

[0032] Carbon flow density is defined as the amount of carbon emissions per unit of electricity in the power grid. It is represented by the symbol ρ, with the unit being tCO2 / MWh.

[0033]

[0034] Where P represents the active power flow in the power system.

[0035] According to the active power flow of the power system, the carbon flow density can be divided into two categories: node carbon flow density and branch carbon flow density. Node carbon flow density is further divided into generator node carbon flow density and load node carbon flow density. Generator node carbon flow density, also known as generator carbon emission intensity, is denoted by E. G The carbon emission intensity vector of the system’s generator set is E G =[e G1 ,e G2 ,…,e Gk ], reflecting the carbon emissions corresponding to the unit of electricity produced by the generator set; and the load node carbon flow density, also known as the load node carbon potential, is symbolized by e i The system node carbon potential vector is E N =[e N1 ,e N2 ,…,e NN ], which represents the carbon emissions corresponding to the unit electricity consumed by the load.

[0036] Compared with the transmission network, most distribution networks show obvious radial characteristics. When the distribution network operates in an open loop, there is no circulation in the network. Therefore, when there is no energy storage system and distributed resources in the distribution network, the carbon potential of all nodes is determined only by the only node upstream. In this case, the carbon potential of all nodes is equal to the carbon potential of the main network. When there are energy storage elements and distributed resources in the system, the carbon potential of the load nodes connected to them will inevitably be affected, thereby continuing to affect the carbon potential of the nodes downstream.

[0037] Step 102: Establish a carbon emission flow model for load and energy storage, as follows:

[0038] The carbon potential of a load node is equivalent to the weighted average of the carbon flow density of all branches flowing into the node and the carbon emission intensity of the generators connected to the node with respect to the power flow, that is:

[0039]

[0040] Where j is the load node number, P i,j is the active power flowing from node i to node j, P G,j represents the active power injected into node j by the generator G connected to node j, ρ i,j is the carbon flow density of the branch connecting node i and node j, e G,i is the carbon emission intensity of the generator G connected to node j, and N is the number of load nodes.

[0041] As can be seen from the above formula, the carbon potential of the load node actually reflects the weighted average of the carbon emission intensity of all relevant generators with respect to the output power when the power flows to the load. Therefore, when the output contribution of the high carbon emission intensity unit to the load node is greater, the carbon emission intensity of the load node is higher, and the carbon emission value equivalent to the power generation side caused by the unit electricity consumption at the node is greater.

[0042] Energy storage elements play a vital role in peak load shaving and valley filling and smoothing load fluctuations, and therefore are widely used in distribution networks. Based on the carbon flow tracking model, the charging phase of the energy storage element can be regarded as an accumulation process of carbon flow, at which time its behavior is similar to that of a load; while in the discharge phase, the energy storage element releases the carbon flow accumulated in the previous charging process back into the system, at which time its behavior is similar to that of a distributed power source.

[0043] Therefore, the present invention introduces the concept of energy storage "Carbon-to-Charge Ratio" (CTCR), which represents the ratio between the carbon emissions absorbed by the energy storage element and its stored electricity, that is, the carbon emissions associated with each kilowatt-hour of electricity in the energy storage system. This indicator can reflect the impact of the energy storage system on carbon emissions in actual operation.

[0044] 1) Carbon emissions during energy storage charging F t ESS,ch

[0045]

[0046] Where P t ESS,ch is the charging power of the energy storage system in period t; is the carbon emission density on the charging branch during period t, where energy storage can be regarded as a load node; Δt is the scheduling time interval.

[0047] 2) Carbon emissions during energy storage discharge F t ESS,dis

[0048]

[0049] Where P t ESS,dis is the discharge power of the energy storage system in period t, which represents the power flowing out of the external port of the energy storage; is the carbon flow density at the external port of the energy storage discharge during period t, then Indicates the power that the energy storage device needs to release; η ESS,dis is the discharge power of energy storage; is the CTCR of energy storage in period t-1.

[0050] Given that the charging and discharging efficiency of energy storage is usually less than 1, the carbon flux density of the external characteristics of energy storage is greater than the internal CTCR. In other words, the carbon emissions caused by energy loss are clearly reflected in the difference in carbon emission density inside and outside the energy storage.

[0051] After the above derivation, the CTCR of the energy storage device in period t is:

[0052]

[0053] In the formula, is the amount of electrical energy stored in the energy storage system during period t. As can be seen from the above formula, through in-depth analysis based on this model, the operating status of energy storage components in the distribution system can be accurately evaluated, laying the foundation for achieving low-carbon optimized scheduling of the distribution network.

[0054] Step 2: A demand response priority function is established to evaluate the demand response capabilities of various load types, and based on this priority, corresponding demand response strategies are designed from two aspects: IDR and PDR.

[0055] Step 201: Taking into account the peak-valley power ratio and the daily load dispersion, the load characteristics of different loads are deeply analyzed, the dispatch priority of various loads is evaluated, and a demand response priority function is established to evaluate the demand response capability of various loads, as follows:

[0056] In order to ensure the rationality and orderliness of scheduling, the present invention comprehensively considers multiple factors such as user scale, direct power purchase, and power transmission distance, and divides industrial users into two main categories, namely high-priority large users (Direct User, DU) and ordinary users (Indirect User, IU). Industrial large users can directly sign contracts with power producers or power generation companies to purchase electricity through direct power purchase, which makes them highly controllable in terms of user demand response. Therefore, high-priority large users DU are placed at the highest level of scheduling priority (Grade A) to ensure the efficiency and flexibility of scheduling.

[0057] IU includes residential, commercial and other industrial users that do not belong to DU. In order to comprehensively evaluate the load curve's ability to smooth out peaks and fill valleys, a series of key parameters including peak power proportion, valley power proportion and load fluctuation rate were selected based on the basic properties of the load curve during peak and valley periods to characterize the load's demand response capability and provide a basis for formulating the demand response priority function and determining the three different priorities of B, C and D.

[0058] 1) Peak power ratio F

[0059]

[0060] In the formula, the duration of peak, flat and valley periods of time-of-use electricity price are T f =[t1, t2], T p =[t3, t4], T g =[t5, t6]; is the load power of node i in period t; L f , L a They are the load integral during the peak period and the load integral for the whole day of the daily average load curve respectively; the ratio indicates the proportion of the load power consumption during the peak period to the total power consumption in a day. The smaller the value, the more significant the contribution of the load in peak shaving and valley filling.

[0061] 2) Valley power consumption ratio G

[0062]

[0063] Where, L g It is the load integral of the trough period of the daily average load curve; this ratio indicates the proportion of electricity consumption during the trough period to the total electricity consumption in a day. The larger the value, the greater the contribution of the load in peak shaving and valley filling.

[0064] 3) Load fluctuation rate λ

[0065]

[0066]

[0067] In the formula, represents the average daily load power consumption; χ represents the degree of dispersion of the daily load curve, and λ represents the fluctuation of the user load curve. The larger its value is, the greater the volatility of user power consumption is and the more adverse factors it causes to the power grid.

[0068] Taking the above indicators into consideration, the load demand response priority evaluation function Z is established. i , in order to accurately characterize the effect of load in peak shaving and valley filling in the power grid, and then carefully divide the scheduling priorities of various types of loads.

[0069] Z i =(1-F i )+G i -λ i

[0070] This indicator shows that the greater the elasticity of the load curve, the higher its dispatch priority. For each type of load, the calculated priority function values ​​are sorted and divided into B, C, and D levels, and corresponding policies are formulated. This helps to motivate users to improve capacity utilization and optimize electricity consumption behavior, thereby making positive contributions to the low-carbon economic development of the power system.

[0071] Step 202: Based on the above priorities, corresponding demand response strategies are designed from two aspects of IDR and PDR.

[0072] The elasticity coefficient of electricity demand refers to the sensitivity of electricity demand to changes in electricity prices or other influencing factors. It is usually expressed as the ratio between the relative change in electricity demand and the relative change in the factors that cause such changes:

[0073]

[0074] Where ε is the demand elasticity coefficient of electricity; p and Δp are the initial load and the load change based on price changes; q and Δq are the initial electricity price and the change in electricity price.

[0075] Furthermore, the elasticity coefficient between the load of user L in period t1 and the price in period t2 is It can be expressed as:

[0076]

[0077]

[0078]

[0079] In the formula, The change rate of electricity consumption of user L in period t1 before and after the implementation of time-of-use electricity price; is the electricity price floating ratio; and The power consumption of user L in time period t1 before and after the electricity price adjustment; and is the electricity price corresponding to the period t2 before and after the electricity price adjustment. When t1 = t2, is the elastic coefficient; otherwise is the mutual elastic coefficient.

[0080] Based on the self-elasticity coefficient, the calculation formula for the rate of change of electricity consumption considering the reduction of users is:

[0081]

[0082] Based on the mutual elasticity coefficient, the calculation formula for the rate of change of electricity consumption considering user transfer is:

[0083]

[0084] In the formula, λ ff and λ gg Represents the effect of load peak shaving and valley filling during peak and valley periods; fp , fg and λ pg Represents the effect of load transfer.

[0085] The load response model based on the time-of-use electricity price mechanism is described as:

[0086]

[0087] In the formula, p f 、p p and p g They represent the average power consumption of users during peak hours, normal hours and valley hours respectively.

[0088] Based on the existing time-of-use electricity price system, the present invention sets different discount prices for users of four different priorities, aiming to motivate high-priority users to respond to demand more actively and frequently. To achieve this goal, by making adjustments based on the original time-of-use electricity price, the peak period electricity price remains unchanged, while the normal period electricity price is reduced by 15%, 10%, 5% and 5% respectively; the valley period electricity price is reduced by 15%, 10%, 10% and 5% respectively. Through such a strategy, high-priority users are encouraged to participate in demand response more actively, thereby achieving more efficient power system scheduling.

[0089] By adopting the discounted time-of-use electricity price, it can be observed that during peak and valley hours, as k f 'Increase and k g 'Decrease, the corresponding calculated |λ ff | and |λ gg | all showed an increasing trend, and

[0090]

[0091] In the formula, and They are the differences in the peak-shaving and valley-filling effects before and after the discounted time-of-use electricity price is implemented for the loads with priority levels A, B, C, and D during the peak hours; and They are the differences in the peak shaving and valley filling effects before and after the discounted time-of-use electricity price is implemented for the loads with priority levels A, B, C, and D during the valley period.

[0092] This shows that by reducing the load's own power consumption, the load curve can be more effectively smoothed. It is worth noting that the higher the scheduling priority of the load, the more significant its effect. At the same time, fp , fg and λ pg The increase in λ indicates that the peak shaving and valley filling effect is also more significant through the role of load transfer. And the higher the load scheduling priority, the corresponding λ fp , fg and λ pgThe greater the increase, the better the effect. This result emphasizes that the discount time-of-use electricity price tiered demand response mechanism has a positive impact on the benefits of peak shaving and valley filling and load transfer while prompting users to respond to electricity demand.

[0093] The incentive-based demand response mechanism provides incentives through the power supplier to encourage end users to actively adjust their electricity consumption behavior when the system needs it. The IDR model constructed by the present invention takes into account transferable load (TL) and curtailable load (CL).

[0094] The cost of incentive-based demand response is DR Including the scheduling cost f DR,dis and incentive cost f DR,com Two parts:

[0095] f DR =f DR,dis +f DR,com

[0096]

[0097]

[0098] In the formula, ΔP i L,tran The transferred power for TL to respond to demand; ΔP i L,cut Power reduction for CL demand response; c DR,dis,tran and c DR,dis,cut are the cost coefficients of TL and CL in the scheduling cost respectively; c DR,com,tran and c DR,com,cut are the cost coefficients of TL and CL in the incentive cost respectively.

[0099] The traditional IDR model uses a fixed cost coefficient, on which a step compensation strategy is introduced to formulate different compensation strategies for users with four different priorities. Through this strategy, power suppliers can provide higher compensation costs to users with higher scheduling priorities, and relatively lower compensation costs to users with lower scheduling priorities. That is:

[0100]

[0101] In the formula, and are the cost coefficients in the incentive cost of TL at priority levels A, B, C, and D, respectively; and They are the cost coefficients in the incentive cost of CL when it is at priority level A, B, C, and D respectively.

[0102] Through flexible pricing strategies, high-priority users can be more effectively incentivized to participate in the response, and power suppliers can provide differentiated incentives to different user groups.

[0103] Among them, for TL and CL involved in IDR, the expressions of power transfer and reduction are as follows:

[0104]

[0105]

[0106] In the formula, and are the load power of TL before and after demand response, respectively; and They represent the load power of CL before and after demand response respectively.

[0107] Taking into account the transfer power constraints of TL and CL, the mathematical model is expressed as:

[0108]

[0109]

[0110]

[0111] Step 3: A two-way feedback mechanism between the upper-level carbon potential data and the lower-level load data was designed. Based on this mechanism, a two-layer low-carbon dispatch model of the distribution network was established and solved.

[0112] Step 301: Establish an upper carbon potential feedback model, as follows:

[0113] The upper model aims to minimize the total cost of power generation in the distribution network, including the main grid power purchase cost f MG , network loss cost f loss , fuel cost of gas turbine DG and the penalty cost of abandoning light f PV , characterized as follows:

[0114] F up =f MG +f loss +f DG +f PV

[0115]

[0116] In the formula, F up is the objective function of the distribution network in the upper model. is the price of electricity purchased from the main grid during period t, P t MGis the power purchased from the main grid during period t; c loss is the cost coefficient of network active power loss, r ij is the resistance of branch ij, I t,ij is the current flowing through branch ij at time t; N DG is the node connected to the gas turbine, a i 、b i 、c i are the quadratic coefficient, linear coefficient and constant coefficient of the gas turbine power generation cost connected to the i-th node, is the active power output of the gas turbine connected to the i-th node in period t; N PV For nodes connected to distributed photovoltaics, c PV is the abandonment penalty cost coefficient, and They are the predicted output and actual output of the distributed photovoltaic unit connected to the i-th node in period t.

[0117] The constraints of the upper model include main grid power purchase constraints, unit operation constraints and distribution network flow constraints.

[0118] 1) Main network electricity purchase restrictions

[0119]

[0120] In the formula, and are the powers transmitted from the distribution network to the upper grid at time t.

[0121] 2) Unit operation constraints

[0122]

[0123]

[0124]

[0125] In the formula, and are the maximum and minimum active outputs of the gas turbine connected to the i-th node respectively; The start and stop status of the unit. When the value is 0, it means shutdown, and when the value is 1, it means startup; and are respectively the maximum climbing rate and maximum sliding rate of the unit connected to the i-th node when it is in the startup state.

[0126] 3) Distribution network flow constraints

[0127]

[0128]

[0129] V i min ≤V i,t ≤V i max

[0130]

[0131]

[0132] Where B is the set of system branches; F(i) is the set of end nodes of the branch with node i as the first end node; T(i) is the set of head nodes of the branch with node i as the last end node; P i,t / Q i,t is the active / reactive power injected into node i during period t; P ij,t / Q ij,t is the active / reactive power flowing through branch ij during period t; l ij is the square of the current amplitude on branch ij; r ij / x ij is the resistance / reactance of branch ij; V i,t is the square of the voltage amplitude at node i; V i min / V i max is the minimum and maximum value of the square of the voltage amplitude at node i; is the active / reactive power purchased from the upper grid at time t; are node i connected to the upper power grid, distributed gas turbine, distributed photovoltaic and energy storage equipment respectively; tanψ is the tangent value of the load power factor.

[0133] Step 302: Establish a lower-layer load feedback model, as follows:

[0134] The objective function F of the lower model is higher than the objective function F of the upper model. up Based on this, the energy storage system operation cost f is further considered ESS , load demand response cost f DR And the various carbon emission costs of the system Characterization is as follows:

[0135]

[0136]

[0137] In the formula, c ESS is the energy storage operation cost coefficient, N ESS A collection of nodes connected to the energy storage system in the system; represents the carbon emission cost coefficient, and are the carbon emission intensity of power purchase from the upper power grid and operation of energy storage equipment during period t, is the carbon emission intensity of the gas turbine connected to the i-th node.

[0138] The constraints of the lower-level load feedback model include not only the constraints in the upper-level model, but also the energy storage system operation constraints, demand response constraints, and carbon emission constraints. This hierarchical relationship ensures that the lower-level model incorporates the relevant constraints of the low-carbon goal while considering the economic goal.

[0139] 1) Energy storage system operation constraints

[0140]

[0141]

[0142]

[0143]

[0144] In the formula, The maximum value of energy storage charging and discharging power; and are 0 / 1 variables indicating the charge and discharge state of energy storage; E ESS is the rated capacity of the energy storage system; and They are the minimum and maximum capacity of the energy storage system; SOC min With SOC max are the minimum and maximum state of charge respectively; η ESS,ch The charging efficiency of energy storage.

[0145] 2) Demand response constraints

[0146] See step 202 for details.

[0147] 3) Carbon emission constraints

[0148]

[0149]

[0150]

[0151] In the formula, F i,t is the carbon emission of node i in period t; is the load power of node i at time t after demand response; is the node carbon potential of node i at time t; is the branch carbon flow density of branch l with i as the first segment; Ωi+ and Ω i- are the branch sets that inject power into node i and flow out of node i respectively; t dis is the set of time periods when the energy storage is in discharge; is the carbon emission intensity of the distributed photovoltaic connected to the i-th node; is the active power flow of branch l.

[0152] Step 303: The scheduling model solution process based on bidirectional data feedback is introduced, which is as follows:

[0153] In the optimization and solution process of the lower load feedback model, the load distribution is adjusted by demand response and substituted into the upper model, and then the output of each unit needs to be recalculated in the upper model. This iterative solution is repeated until the optimal solution is obtained. The specific process is as follows: Figure 1 The specific steps can be summarized as follows:

[0154] 1) Input distribution network parameters, load and photovoltaic predicted output data, solve the upper carbon potential feedback model, and obtain the power flow results after scheduling.

[0155] 2) Based on the flow results obtained after upper-level scheduling, the carbon emission flow model is used to calculate the carbon potential of load nodes and the carbon potential of energy storage charging and discharging.

[0156] 3) Feedback the calculated node carbon potential to the lower model, solve the lower load feedback model and calculate the load output after demand response.

[0157] 4) The adjusted load output after the lower-level scheduling is re-substituted into the upper-level model, the upper-level carbon potential feedback model is re-solved, the system flow distribution and the scheduling strategies of each subject are updated, and the optimization and iteration are repeated until the optimal scheduling result is obtained.

[0158] 5) Output the final dispatch results, including total cost, system carbon emissions, load and energy storage output status, etc. These data will clearly show the performance indicators and status of the system after optimized dispatch.

[0159] Example

[0160] Improve the system topology of IEEE-33 node distribution network Figure 2As shown. The distribution network is connected to the upper power grid at node 0, where gas turbines with an installed capacity of 200kW are installed at nodes 17, 18 and 33; distributed photovoltaic units with an installed capacity of 600kW are installed at nodes 15, 17 and 20, and distributed photovoltaic units with an installed capacity of 800kW are installed at nodes 16, 18 and 32. A 200kWh energy storage system is connected to nodes 18 and 32, with a maximum charge and discharge power of 200kW, a charging efficiency of 0.953, and a discharge efficiency of 0.926. Set the scheduling period to 24h and the step size to 1h. The predicted power of photovoltaics and loads is shown in Figure 3 shown.

[0161] The loads are divided into three categories: residential, commercial and industrial. Representative load curves are selected, and the load power per unit value is as follows: Figure 4 As shown. Based on the demand response priority evaluation parameters, the priority function values ​​of the three loads are calculated, and the order of arrangement is: residential load < commercial load < industrial load. Therefore, high-priority large users are classified as Class A, ordinary industrial users are Class B, commercial users are Class C, and residential users are Class D.

[0162] 1) Analysis of the results of the low-carbon dispatch model of distribution network

[0163] In order to verify the effectiveness of the proposed low-carbon dispatch model for distribution network based on the carbon potential-load data bidirectional feedback mechanism considering carbon emission flow and demand response priority, four scenarios are set up for comparative analysis:

[0164] Scenario 1: Traditional economic dispatch model, without considering demand response and energy storage.

[0165] Scenario 2: A low-carbon dispatching method for distribution networks based on a two-way data feedback mechanism is used, taking energy storage into account but not demand response.

[0166] Scenario 3: Adopt a low-carbon dispatching method for distribution networks based on a two-way data feedback mechanism, taking into account energy storage and traditional demand response.

[0167] Scenario 4: A low-carbon dispatching method for distribution networks based on a two-way data feedback mechanism is adopted, taking into account demand response with energy storage and dispatch priority.

[0168] Solve the four scenarios and obtain the scheduling results as follows: Figure 5 As shown in (a)-(d).

[0169] Depend on Figure 5 From (a), we can see that in scenario 1, during the peak period of photovoltaic output, the distribution network can meet the power demand in the system by relying solely on distributed photovoltaic power generation. However, photovoltaic energy is not fully utilized, its absorption rate is low, and the carbon emission level of the system is high. Figure 5(b) shows that, compared with scenario 1, after introducing energy storage, scenario 2 reduces the penalty cost of abandoned light by 33.66 yuan (11.17%), the total system operation cost by 423.54 yuan (2.51%), and the carbon emissions by 0.041t (0.15%). This shows that energy storage can increase the photovoltaic absorption rate of the distribution network, thereby reducing the power that the distribution network purchases from the upper level and reducing the carbon emissions of the system. Figure 5 (c) shows that compared with scenario 2, scenario 3 demand response has played a role in shaving peaks and filling valleys of the system carbon potential and improving the photovoltaic absorption rate, which further reduced the total operating cost by 2225.04 yuan (20.13%) and the system carbon emissions by 3.795t (22.13%). By comparing with scenarios 1 and 2, it can be found that the synergy of demand response and energy storage can further improve the economy and low carbon nature of the distribution network. Figure 5 (d) It can be observed that compared with scenario three, the introduction of scheduling priority in scenario four enables large users with low scheduling costs to respond first, reducing the demand response cost by 51.08 yuan (6.59%), and the total system operation cost is further reduced by 1380.49 yuan (9.71%). In addition, the system carbon emissions are further reduced by 1.95t (8.40%). Therefore, considering the carbon emission flow and demand response priority, the low-carbon scheduling model of the distribution network based on the data two-way feedback mechanism can not only improve the absorption rate of wind power, but also reduce the total system operation cost, reduce carbon emissions, and achieve the dual goals of low carbon and economy.

[0170] 2) Analysis of the impact of demand response priority on demand response results

[0171] In order to more clearly reflect the impact of the demand response priority proposed in the present invention on the load change of the three types of loads before and after they participate in demand response, Figure 6 (a)-(c) show the power demand of residential load, commercial load and industrial load under scenario 2, scenario 3 and scenario 4.

[0172] exist Figure 6 It can be observed that the change in residential load before and after the participation in demand response is small. The reason is that the residential load accounts for a small proportion of the total load, has relatively low controllability, and its electricity consumption period is relatively fixed, resulting in a relatively small scale of its participation in demand response. In addition, due to the low priority of demand response for residential loads and commercial loads, the system is more inclined to give priority to industrial users with higher priority and lower response cost under the conditions of meeting low carbon and economic requirements. Therefore, under scenario four, the change in residential load and commercial load is smaller than that in scenario three, while the change in industrial load is greater than that in scenario three.

[0173] Figure 7(a)-(b) show the dispatch results of DU in scenarios 3 and 4. By comparison, after considering the priority of demand response, the transfer amount of load in DU increases significantly, and the transferable load is adjusted from the period of high carbon emission of the system to the period of low carbon emission, further absorbing the photovoltaic output, thereby reducing the total carbon emission of the system.

[0174] Based on the same inventive concept, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the aforementioned distribution network scheduling method based on the carbon potential-load data bidirectional feedback mechanism are implemented.

[0175] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the aforementioned distribution network scheduling method based on the carbon potential-load data bidirectional feedback mechanism.

[0176] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented 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.

[0177] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0178] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0180] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A distribution network dispatching method based on a carbon potential-load data bidirectional feedback mechanism, characterized in that: The steps include: Step 1: Based on the basic theory of carbon emission flow, the carbon flow density index is introduced to trace the carbon emission responsibility on the power generation side to the user side, and further establish the carbon emission flow model of load and energy storage; Step 2: Establish a load demand response priority function to evaluate the demand response capabilities of various loads. Based on the priority, design corresponding demand response strategies from both price and load aspects. The specific process is as follows: Step 201, various types of loads include large industrial users, residents, commercial and other industrial users. Large industrial users are placed at the highest level of scheduling priority, i.e., level A. Peak power proportion, valley power proportion and load fluctuation index are selected to characterize the load demand response capability. The load demand response priority function is established as follows: Z=(1-F)+G-λ Where Z represents the load demand response priority evaluation function; F represents the peak power ratio, which is: In the formula, the duration of peak, flat and valley periods of time-of-use electricity price are T f =[t1, t2], T p =[t3, t4], T g =[t5, t6]; is the load power of node i in period t; L f , L a They are the peak load integral and the whole day load integral of the daily average load curve respectively; G represents the proportion of valley power consumption, which is: Where, L g The load integral of the trough period of the daily average load curve; λ represents the load fluctuation rate, which is: In the formula, represents the average daily load power consumption, and χ represents the degree of dispersion of the daily load curve; Calculate the priority function values ​​of residential, commercial and other industrial users, and sort them from large to small. According to the sorting, the dispatching priorities are divided into B, C and D levels in order of sorting; Step 202, designing a demand response strategy from the price aspect, setting different discount prices for users of four different priorities in the time-of-use electricity price system, that is: making adjustments based on the original time-of-use electricity price, keeping the peak period electricity price unchanged, reducing the electricity prices for users of levels A, B, C and D in the normal period by 15%, 10%, 5% and 5% respectively; reducing the electricity prices for users of levels A, B, C and D in the valley period by 15%, 10%, 10% and 5% respectively; With the adjustment of time-of-use electricity price, the peak period electricity price fluctuation ratio k f ' and the price fluctuation ratio k during the off-peak period g ' decreases, |λ ff | and |λ gg | all showed an increasing trend, and In the formula, λ ff and λ gg They represent the effects of load shaving and valley filling during peak and valley periods respectively. and They are the differences in the peak-shaving and valley-filling effects of loads at levels A, B, C, and D before and after the discounted time-of-use electricity price is implemented during peak hours; and They are the differences in the peak-shaving and valley-filling effects before and after the discounted time-of-use electricity price is implemented for loads at levels A, B, C, and D during the valley period; Design a demand response strategy from the load aspect, introduce a step compensation strategy, and formulate different compensation strategies for users with four different priorities. The power supplier provides higher compensation costs to users with higher scheduling priorities and lower compensation costs to users with lower scheduling priorities through the step compensation strategy, namely: In the formula, and are the cost coefficients in the incentive cost when the transferable load is at levels A, B, C and D respectively; and are the cost coefficients in the incentive cost when the load that can be reduced is at levels A, B, C and D respectively; The expressions for power transfer and reduction corresponding to the transferable load and the curtailable load are as follows: In the formula, and are the load powers that can be reduced before and after demand response, and are the load powers of the transferable load before and after demand response, and are the load powers that can be reduced before and after demand response, respectively, and T is a scheduling cycle; The transfer power constraints of the transferable load and the curtailable load are as follows: Step 3: Based on steps 1 and 2, a two-way feedback mechanism for the upper-level carbon potential data and the lower-level load data is designed. Based on the two-way feedback mechanism, a two-layer low-carbon dispatching model for the distribution network is established and solved to obtain the final dispatching result.

2. The distribution network dispatching method based on the carbon potential-load data bidirectional feedback mechanism according to claim 1 is characterized in that: The specific process of step 1 is as follows: According to the active power flow of the power system, the carbon flow density is divided into two categories: node carbon flow density and branch carbon flow density. The node carbon flow density is further divided into generator node carbon flow density and load node carbon flow density. The generator node carbon flow density is also called the generator carbon emission intensity, and the load node carbon flow density is also called the load node carbon potential. The size of the load node carbon potential is equivalent to the weighted average sum of the carbon flow density of all branches flowing into the load node and the carbon emission intensity of the generators connected to the load node with respect to the power flow, that is: Where j is the load node number, e j is the load node carbon potential of node j, P i,j is the active power flowing from node i to node j, ρ i,j is the carbon flow density of the branch connected by node i and node j, P G,j represents the active power injected into node j by the generator connected to node j, e G,i is the carbon emission intensity of the generator connected to node j, and N is the number of load nodes; The energy storage carbon rate is introduced to reflect the impact of the energy storage system on carbon emissions during actual operation. The carbon emissions during energy storage charging and discharging are expressed as: F t ESS,ch =P t ESS,ch ρ t ESS,ch Δt In the formula, F t ESS,ch 、F t ESS,dis are the carbon emissions during energy storage charging and discharging, P t ESS,ch , P t ESS,dis are the charging power and discharging power of the energy storage system in period t, respectively, t ESS,ch is the carbon emission density on the charging branch during period t, ρ t ESS,dis is the carbon flow density of the energy storage discharge port during period t, Δt is the scheduling time interval, η ESS,dis is the discharge power of the energy storage, is the carbon charging rate of energy storage in period t-1; The carbon charge rate of the energy storage device in period t is expressed as: In the formula, ρ t ESS is the carbon charging rate of energy storage during period t, E t ESS are the electrical energy stored in the energy storage system in time periods t-1 and t respectively.

3. The distribution network dispatching method based on the carbon potential-load data bidirectional feedback mechanism according to claim 1 is characterized in that: The specific process of step 3 is as follows: Step 301, with the goal of minimizing the total cost of power generation in the distribution network, establish an upper carbon potential feedback model, specifically: F up =f MG +f loss +f DG +f PV In the formula, F up is the objective function of the upper carbon potential feedback model of the distribution network, f MG 、f loss 、f DG and f PV They are the main grid electricity purchase cost, network loss cost, gas turbine fuel cost and abandoned solar penalty cost. is the price of electricity purchased from the main grid during period t, P t MG is the power purchased from the main grid during period t, c loss is the cost coefficient of network active power loss, r ij is the resistance of branch ij, I t,ij is the current flowing through branch ij during period t, Ω branch is the set of all transmission lines in the distribution network, N DG is the node connected to the gas turbine, a i 、b i 、c i are the quadratic coefficient, linear coefficient and constant coefficient of the gas turbine power generation cost at access node i, is the active power output of the gas turbine connected to node i during period t, c PV is the penalty cost coefficient for abandoned light, N PV To access the nodes of distributed photovoltaic, and are the predicted output and actual output of the distributed photovoltaic unit connected to node i in period t, and Δt is the scheduling time interval; The constraints of the upper carbon potential feedback model are: 1) Main network electricity purchase restrictions In the formula, and are the minimum and maximum values ​​of power transmitted from the distribution network to the upper power grid during period t, respectively; 2) Unit operation constraints In the formula, and are the maximum and minimum active output of the gas turbine connected to node i, respectively; They are the start and stop status of the unit in the time periods t and t-1, respectively. A value of 0 indicates shutdown, and a value of 1 indicates startup; is the active power output of the gas turbine connected to node i in period t-1, R i up and R i down are the maximum climbing rate and maximum sliding rate of the unit connected to node i when it is in the startup state; 3) Distribution network flow constraints In i min ≤V i,t ≤V i max Where P i,t , Q i,t are the active and reactive power injected into node i during period t; P ij,t , Q ij,t are the active and reactive power flowing through branch ij in period t respectively; F(i) is the set of end nodes of the branch with node i as the first endpoint; T(i) is the set of head nodes of the branch with node i as the terminal node; P ki,t , Q ki,t are respectively the active and reactive power flowing through branch ki during period t; l ki,t , l ij,t are the squares of the currents flowing through branches ki and ij during period t; r ki 、x ki are the resistance and reactance of branch ki respectively; N is the number of load nodes; V i,t 、V j,t are the squares of the voltage amplitudes at nodes i and j respectively; r ij 、x ij are the resistance and reactance of branch ij respectively; B is the set of system branches; V i min 、V i max are the minimum and maximum values ​​of the square of the voltage amplitude at node i, respectively; The node i is connected to the upper power grid, distributed gas turbine, distributed photovoltaic and energy storage equipment set respectively; are the active and reactive power purchased from the superior power grid at time t respectively; are the discharge power and charging power of the energy storage system connected to node i in period t respectively; is the load power of node i in period t; tanψ is the tangent value of the load power factor; Step 302, establish a lower layer load feedback model, specifically: F=F up +f ESS +f DR +f CO2 Where F is the objective function of the lower load feedback model, f ESS 、f DR and are respectively the energy storage system operation cost, load demand response cost and various carbon emission costs of the system, c ESS is the energy storage operation cost coefficient, P t ESS,ch , P t ESS,dis are the charging power and discharging power of the energy storage system in period t, is the carbon emission cost coefficient, and are the carbon emission intensity of power purchase from the upper power grid and operation of energy storage equipment during period t, is the carbon emission intensity of the gas turbine connected to node i; The constraints of the lower load feedback model are: 1) Energy storage system operation constraints In the formula, is the maximum value of energy storage charging and discharging power, and are 0 / 1 variables representing the energy storage charging and discharging states, E ESS is the rated capacity of the energy storage system, and are the minimum and maximum capacity of the energy storage system, SOC min and SOC max are the minimum and maximum state of charge, respectively. are the electrical energy stored by the energy storage system in time periods t+1 and t, respectively, η ESS,ch , η ESS,dis are the charging and discharging power of energy storage respectively; 2) Demand response constraints, i.e. step 202 3) Carbon emission constraints In the formula, F i,t is the carbon emission of node i in period t, is the node carbon potential of node i in period t, is the load power of node i in period t after demand response, is the branch carbon flow density of branch l with i as the first segment, Ω i+ and Ω i- are the branch sets that inject power into node i and flow out of node i, respectively. is the carbon charging rate of energy storage during period t, is the carbon emission intensity of distributed photovoltaic access node i, is the active power flow of branch l, is the carbon flow density of branch l, t dis is the set of time periods when the energy storage is in discharge; Step 303, solving the double-layer low-carbon dispatch model of the distribution network based on the carbon potential-load data bidirectional feedback mechanism, specifically: Step 3031, input distribution network parameters, load and photovoltaic predicted output data, solve the upper carbon potential feedback model, and obtain the power flow results after scheduling; Step 3032, using the power flow results obtained in step 3031, the carbon emission flow model is used to calculate the carbon potential of the load node and the carbon potential of the energy storage charge and discharge; Step 3033, feeding back the calculated node carbon potential to the lower load feedback model, solving the lower load feedback model and calculating the load output after demand response; Step 3034, re-substitute the post-demand response load output calculated in step 3033 into the upper carbon potential feedback model, re-solve the upper carbon potential feedback model, update the system flow distribution and the scheduling strategy of each user, and repeatedly optimize and iterate until the optimal scheduling result is obtained and output.

4. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the computer program, the steps of the distribution network dispatching method based on the carbon potential-load data bidirectional feedback mechanism as described in any one of claims 1 to 3 are implemented.

5. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the distribution network dispatching method based on the carbon potential-load data bidirectional feedback mechanism as described in any one of claims 1 to 3 are implemented.

Images