Park low-carbon electricity consumption operation method considering carbon flow demand response

Abstract

The invention relates to a park low-carbon power consumption operation method considering carbon flow demand response, and the method comprises the following steps: firstly, carrying out the load modeling according to the characteristics of various types of loads in a park, calculating the 24-hour carbon flow distribution of a distribution network in combination with a carbon flow model and a dispatching mode of an upper-layer distribution network, and obtaining a 24-hour carbon potential curve of a node where the park is located; secondly, guiding various types of flexible loads and electric vehicle clusters in the park to perform low-carbon electricity demand response by integrating time-of-use electricity price and node carbon potential; simulation research shows that the method can give full play to the flexibility of the electric vehicle cluster in the park, saves the energy storage investment cost, and effectively improves the power utilization low-carbon benefit of the park.

Description

Technical Field

[0001] The present invention relates to the field of optimal dispatching of power system distribution networks, and more specifically, to a method for low-carbon power consumption operation of a park considering carbon flow demand response. Background Art

[0002] As the core driving force for the development of electricity consumption in China, industrial parks concentrate more than 70% of industrial energy consumption. Therefore, the park is the main battlefield for the power system to carry out energy conservation and carbon reduction transformation, and promoting the development of low-carbon power technologies is an inevitable trend for the future development of the power industry. The quantification and allocation of carbon emissions in the power system are the premise for the development of low-carbon power technologies. The carbon flow theory realizes the transfer of power carbon emission responsibility from the source side to the load side from the "carbon perspective". Current research has considered the calculation and application of the carbon flow theory in various operating scenarios, and promotes the low-carbon operation of multi-agent interaction in the power system by combining the indirect carbon emissions from the power generation side to the power consumption side with mechanisms such as source-load interaction and carbon exchange. However, most of the current carbon flow-related research starts from the large power transmission network side, and there is little research on low-carbon dispatching between the distribution network side and the lower-level park. Conducting research on guiding the demand response of the lower-level park by the carbon flow on the distribution network side, so as to improve the low-carbon power consumption benefit of the new distribution system, has become a hot issue in current research. Summary of the Invention

[0003] Object of the Invention. To solve the deficiencies in the prior art, the object of the present invention is to effectively improve the low-carbon power consumption benefit of the park, so a method for low-carbon power consumption operation of a park considering carbon flow demand response is provided.

[0004] Technical Solution. To solve the above technical problems, the present invention proposes a method for low-carbon power consumption operation of a park considering carbon flow demand response, and the method includes the following steps:

[0005] Step 1: Obtain the operation parameters of the upper-level distribution network and the lower-level park, including the parameters of each unit, each load prediction curve, the time-of-use electricity price between the upper and lower levels, and the predicted data of renewable energy output, and model each type of load in the lower-level park;

[0006] Step 2: The upper-level distribution network conducts power consumption dispatching according to the load prediction power consumption curve to obtain the steady-state power flow distribution of the distribution network;

[0007] Step 3: Calculate the carbon potential curve of the node where the lower-level park is located according to the steady-state power flow distribution of the upper-level distribution network, allocate the initial carbon emission right quota to the park, and generate a carbon exchange mechanism model;

[0008] Step 4: The lower-level park comprehensively considers the changes in electricity price and carbon potential at different times, conducts low-carbon power consumption dispatching, and generates a linear integer programming problem with the sum of power consumption cost and carbon exchange cost as the objective according to the park operation constraints and the constraints of various load models therein;

[0009] Step 5: Power distribution network conducts power consumption low-carbon scheduling for the lower-level park according to the power consumption curve and completes carbon exchange in the park on the same day. The linear integer programming problem is solved using the GUROBI solver on the GAMS optimization platform.

[0010] Furthermore, in Step 1, the modeling of various types of loads in the lower-level park is as follows:

[0011] 1) Electric vehicle load

[0012] The constraints for electric vehicle load when accessing the park are as follows:

[0013]

[0014] In the formula, the subscript i is the serial number of the electric vehicle, the subscript t is the scheduling period, and the superscripts C and D represent the charging and discharging states respectively. and are the charging / discharging power of electric vehicle i at time t respectively. and are the charging / discharging efficiency of electric vehicle i at time t respectively. E i,t is the electricity quantity of electric vehicle i at time t. E i,leave is the electricity quantity that electric vehicle i needs to meet when leaving the charging pile. and are the upper and lower limits of the charging power of electric vehicle i respectively. and are the upper and lower limits of the discharging power of electric vehicle i respectively. and are the upper and lower limits of the electricity quantity of electric vehicle i respectively.

[0015] At the same time, corresponding compensation is carried out for vehicle discharging:

[0016]

[0017] In the formula, C EV is the discharging compensation cost of the electric vehicle, n is the number of electric vehicles, is the unit power compensation coefficient for electric vehicle discharging.

[0018] 2) Curtailable load

[0019] The scheduling constraints for curtailable load are as follows:

[0020]

[0021] In the formula, T = 24 is the length of the time interval of a day, u t is a 0-1 state variable, u t = 1 indicates that the load is curtailed at time t. and They are the minimum and maximum durations that the load needs to satisfy during response curtailment respectively.

[0022] Compensate for the load curtailment simultaneously:

[0023]

[0024] In the formula, C cut is the load curtailment compensation cost, is the curtailed power consumption of the load at time t, is the subsidy obtained for the load response curtailment per unit power, while is the fixed subsidy obtained for curtailment in a certain period.

[0025] 3) Shiftable load

[0026] The continuous constraint of the shiftable load scheduling working time is as follows:

[0027]

[0028] In the formula, τ is the starting time after the load is shifted, t s is the continuous working time; y t is a 0-1 state variable, y t = 1 indicates that there is a shifted load at time t;

[0029] Compensate for the load shift simultaneously:

[0030]

[0031] In the formula, C shift is the load shift compensation cost, is the subsidy obtained by the user after shifting the load per unit power, t sh- and t sh+ are the acceptable initial and final working periods of the shiftable load in a day respectively, P t shift is the shifted load at time t.

[0032] 4) Transferable load

[0033] The transferable load scheduling constraints are as follows:

[0034]

[0035] In the formula, and P t trans are the transferable load powers at time t before and after scheduling respectively, and are the upper and lower limits of the power after the load is transferred respectively, υ tA 0-1 state variable υ for judging whether the load has shifted t = 1 indicates that the load has shifted during period t, t tr- and t tr+ are respectively the initial and final working periods that the shiftable load can accept in a day, is the minimum continuous operation time;

[0036] At the same time, compensate for the load shift:

[0037]

[0038] In the formula, C trans is the load shift compensation cost, is the compensation price for unit power load shift.

[0039] Furthermore, the upper-level distribution network operator scheduling model in step 2:

[0040] 1) Micro diesel generator operation constraints:

[0041]

[0042] |P DE,t+1 -P DE,t |≤ΔP DE,max (18)

[0043] In the formula, P DE,t and Q DE,t are respectively the active and reactive powers generated by the diesel generator, P DE,max and P DE,min are the upper and lower limits of the active power generated by the diesel generator, and the corresponding Q DE,max and Q DE,min are the upper and lower limits of the reactive power, and ΔP DE,max is the upper limit of the diesel generator's ramp-up / down power;

[0044] 2) Distribution network energy storage operation constraints:

[0045] The energy storage operation constraints are similar to those of the electric vehicle load and satisfy formulas (1)-(5).

[0046] 3) Distribution network operation constraints:

[0047]

[0048] V i,min ≤V i,t ≤V i,max (22)

[0049] l ij,t ≤l ij,max (23)

[0050] Wherein, P ij,t and Q ij,t are respectively the active and reactive powers on branch ij during period t, r ij and x ij are respectively the resistance and reactance of branch ij, F(i) is the set of end nodes of the branches with node i as the starting end point, T(i) is the set of starting end nodes of the branches with node i as the end node, l ij,t is the square of the current amplitude on branch ij during period t, V i,t is the square of the voltage amplitude at node i during period t, V i,max and V i,min are the upper and lower limits of the node voltage amplitude, l ij,max is the upper limit of the square of the branch current.

[0051] Dispatch objective function of the upper-level distribution network operator:

[0052]

[0053] Wherein, C D is the electricity consumption cost of distribution network operation, c t is the time-of-use electricity price of the superior power grid, P sub,t represents the power purchased from the superior power grid during period t, c ES is the unit power cost of energy storage charge and discharge, c DE is the power generation cost coefficient of the diesel engine, P ch,t and P dis,t are respectively the charge and discharge powers of the energy storage in the distribution network during period t, P DE,t is the output power of the diesel engine.

[0054] Furthermore, in step 3, the carbon potential calculation method of the node where the park is located is as follows:

[0055] The carbon potential of node i in the power system is calculated through the proportional sharing principle:

[0056]

[0057] Wherein, e i is the carbon potential of node i, and T i + respectively represent the set of branches and the set of units that are connected to node i and inject active power into it; G s represents the active power output of generator set s; e s represents the carbon emission intensity of generator set s; P j→i represents the active power flowing from branch ji to node i; e j represents the carbon potential of node j on branch ji.

[0058] Further, in step 3, the initial carbon emission rights quota of the lower-level park is allocated by means of free quota, and carbon exchange is carried out according to the quota, the actual indirect carbon emissions and the upper-level power grid connection:

[0059]

[0060] In the formula, is the initial quota; ε e is the carbon quota coefficient corresponding to the unit power consumption, is the predicted value of the active power of the net power load of the park at time t, C c is the carbon exchange cost of the park, c is the basic carbon price in the carbon exchange market, E c is the indirect carbon emission caused by the actual purchase of non-clean electricity by the park, e i,t is the carbon potential of node i where the park is located at time t, P 0,t is the actual electricity purchase volume of the park at time t, that is, the tie-line power.

[0061] Further, the specific method of step 4 is as follows:

[0062] Optimization scheduling model for lower-level park operators:

[0063] minC all =C E +C c +C com (29)

[0064]

[0065] C com =C cut +C shift +C trans +C EV (31)

[0066] In the formula, C all is the comprehensive cost of electricity and carbon for park operation, C E is the electricity purchase cost of the park. Among them, c t ' is the time-of-use electricity price for the park to purchase electricity from the power grid, C c is the carbon exchange cost of the park, C com is the compensation cost required during the park scheduling process;

[0067] Meet the power balance constraint conditions of the park:

[0068]

[0069] P 0,t +P PV,t =P L,t (33)

[0070] Wherein, P L,t is the total load of the park at time period t, P zy,t is the load that does not participate in scheduling, P shift,t is the load that can be shifted at time period t, P cut,t is the load that can be curtailed at time period t, is the load of the electric vehicle cluster at time period t, P PV,t is the power generation power of the small photovoltaic units in the park at time period t.

[0071] Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0072] Replacing the energy storage unit with an EV cluster for demand response scheduling can save the investment and construction cost of energy storage. At the same time, it can ensure the flexibility of scheduling at a very low scheduling cost and has a quite high carbon emission reduction potential. This method can control the flexible load in the park to interact according to the carbon potential signal and time-of-use electricity price, which can not only achieve a certain "peak shaving and valley filling" effect, but also effectively reduce the indirect carbon emissions in the park, control the carbon exchange cost in the park, and ensure the low-carbon and power consumption operation of the system. Description of the drawings

[0073] Figure 1 is the conceptual framework diagram of EV intelligent charging;

[0074] Figure 2 is the topological diagram of the IEEE33 distribution system adopted in the present invention;

[0075] Figure 3 is the heat map of time-of-use carbon potential data;

[0076] Figure 4 is the framework diagram of the two-layer scheduling model of the low-carbon economic operation strategy of the park considering carbon flow demand response in the present invention;

[0077] Figure 5 is the data diagram of photovoltaic output and initial load in the park;

[0078] Figure 6 is the 24-hour carbon potential broken line diagram of nodes 3&15 in the example;

[0079] Figure 7 is the comparison schematic diagram of the power of the connecting line of Park 1;

[0080] Figure 8 is the comparison schematic diagram of the EV intelligent charging power in Park 1. Detailed implementation manners

[0081] To better understand the present invention, the technical solution of the present invention will be further described below in conjunction with the drawings and embodiments.

[0082] As Figure 4As shown in the figure, the present invention proposes a low-carbon electricity operation method for a park considering carbon flow demand response, and the method includes the following steps:

[0083] Step 1: Obtain the upper-level distribution network and lower-level park operation parameters, including various unit parameters, various load prediction curves, time-of-use electricity prices between the upper and lower levels, and renewable energy output prediction data, and model various types of loads in the lower-level park;

[0084] Step 2: The upper-level distribution network performs power consumption scheduling according to the load prediction power consumption curve to obtain the steady-state power flow distribution of the distribution network;

[0085] Step 3: Calculate the carbon potential curve of the node where the lower-level park is located according to the steady-state power flow distribution of the upper-level distribution network, allocate initial carbon emission allowances to the park, and generate a carbon exchange mechanism model;

[0086] Step 4: The lower-level park comprehensively considers the electricity price and carbon potential changes in different periods, conducts low-carbon power consumption scheduling, and generates a linear integer programming problem with the sum of power consumption costs and carbon exchange costs as the objective according to the park operation constraints and various load model constraints therein;

[0087] Step 5: The distribution network conducts carbon exchange with the lower-level park according to the low-carbon power consumption scheduling power consumption curve of the lower-level park on the same day, and solves this linear integer programming problem using the GUROBI solver on the GAMS optimization platform.

[0088] Further, in Step 1, various types of loads in the lower-level park are modeled as follows:

[0089] 1) Electric vehicle load

[0090] When the electric vehicle load is connected to the park, it satisfies the following constraints:

[0091]

[0092] In the formula, the subscript i is the serial number of the electric vehicle, the subscript t is the scheduling period, and the superscripts C and D represent the charging and discharging states respectively, and are the charging / discharging power of electric vehicle i in period t respectively, and are the charging / discharging efficiencies of electric vehicle i in period t respectively, E i,t is the electric quantity of electric vehicle i in period t, E i,leave is the electric quantity that needs to be satisfied when electric vehicle i leaves the charging pile, and are the upper and lower limits of the charging power of electric vehicle i respectively, and are the upper and lower limits of the discharging power of electric vehicle i respectively, and are the upper and lower limits of the power of electric vehicle i respectively;

[0093] At the same time, corresponding compensation is carried out for the vehicle discharging:

[0094]

[0095] In the formula, C EV is the discharging compensation cost of the electric vehicle, n is the number of electric vehicles, is the unit power compensation coefficient for the electric vehicle discharging.

[0096] 2) Curtailable load

[0097] The scheduling constraints of the curtailailable load are as follows:

[0098]

[0099] In the formula, T = 24 is the length of the time interval of a day, u t is a 0-1 state variable, u t = 1 indicates that the load is curtailed in time period t, and are the minimum and maximum durations that the load needs to satisfy when responding to curtailment respectively.

[0100] At the same time, compensation is carried out for the load curtailment:

[0101]

[0102] In the formula, C cut is the load curtailment compensation cost, is the curtailed power consumption of the load in time period t, is the subsidy obtained by the unit power load responding to curtailment, while is the fixed subsidy obtained when curtailment occurs in a certain time period.

[0103] 3) Shiftable load

[0104] The continuous constraint of the scheduling working time of the shiftable load is as follows:

[0105]

[0106] In the formula, τ is the starting time after the load is shifted, t s is the continuous working time; y t is a 0-1 state variable, y t = 1 indicates that there is a load after shifting in time period t;

[0107] At the same time, compensation is carried out for the load shifting:

[0108]

[0109] Where C shift is the load shifting compensation cost, is the subsidy obtained by the user after shifting the load of unit power, and are respectively the acceptable initial and final working periods of the shiftable load in a day, and P t shift is the shifted load at time period t.

[0110] 4) Shiftable load

[0111] The shiftable load scheduling constraints are as follows:

[0112]

[0113] Where and P t trans are respectively the shiftable load powers before and after scheduling at time period t, and are respectively the upper and lower limits of the power after load transfer, and υ t is a 0-1 state variable for judging whether the load has been transferred. υ t =1 indicates that the load is transferred at time period t, and t tr- and t tr+ are respectively the acceptable initial and final working periods of the shiftable load in a day, is the minimum continuous operation time;

[0114] Meanwhile, compensation is made for the load transfer:

[0115]

[0116] Where C trans is the load transfer compensation cost, is the compensation price for the load transfer of unit power.

[0117] Furthermore, the upper-level distribution network operator scheduling model in step 2:

[0118] 1) Micro diesel generator operation constraints:

[0119]

[0120] |P DE,t+1 -P DE,t ≤ΔP DE,max (18)

[0121] Where P DE,t and Q DE,t are respectively the active and reactive powers generated by the diesel generator, and P DE,max and PDE,min are the upper and lower limits of the active power generated by the diesel generator, and the corresponding Q DE,max and Q DE,min are the upper and lower limits of the reactive power, and ΔP DE,max is the upper limit of the ramp-up / ramp-down power of the diesel generator;

[0122] 2) Distribution network energy storage operation constraints:

[0123] The energy storage operation constraints are similar to those of the electric vehicle load and satisfy equations (1)-(5).

[0124] 3) Distribution network operation constraints:

[0125]

[0126] V i,min ≤V i,t ≤V i,max (22)

[0127] l ij,t ≤l ij,max (23)

[0128] In the formula, P ij,t and Q ij,t are the active and reactive powers on the branch ij at time t, r ij and x ij are the resistance and reactance of the branch ij respectively, F(i) is the set of the end nodes of the branches with node i as the starting end point, T(i) is the set of the starting end nodes of the branches with node i as the end node, l ij,t is the square of the current amplitude on the branch ij at time t, V i,t is the square of the voltage amplitude at node i at time t, V i,max and V i,min are the upper and lower limits of the node voltage amplitude, and l ij,max is the upper limit of the square of the branch current.

[0129] Upper-level distribution network operator scheduling objective function:

[0130]

[0131] In the formula, C D is the power consumption cost of the distribution network operation, c t is the time-of-use electricity price of the superior power grid, P sub,t represents the power purchase from the superior power grid at time t, c ES is the unit power cost of the energy storage charge and discharge, c DE is the power generation cost coefficient of the diesel engine, P ch,t and P dis,t are the charge and discharge powers of the energy storage in the distribution network at time t respectively, and P DE,tIt is the output power of the diesel engine.

[0132] Further, in step 3, the carbon potential calculation method of the node where the park is located is as follows:

[0133] The carbon potential of node i in the power system is calculated through the proportional sharing principle:

[0134]

[0135] In the formula, e i is the carbon potential of node i, and T i + respectively represent the set of branches and the set of units that are connected to node i and inject active power into it; G s represents the active power output of generator set s; e s represents the carbon emission intensity of generator set s; P j→i represents the active power flowing into node i on branch ji; e j represents the carbon potential of node j on branch ji.

[0136] Further, in step 3, the initial carbon emission right quota is allocated to the lower-level park in the form of free quotas, and carbon exchange is carried out with the upper-level distribution network according to the quota and the actual indirect carbon emissions:

[0137]

[0138] In the formula, is the initial quota; ε e is the carbon quota coefficient corresponding to unit power consumption, is the predicted value of the active power of the net electrical load of the park at time t, C c is the carbon exchange cost of the park, c is the basic carbon price in the carbon exchange market, E c is the indirect carbon emission caused by the actual purchase of non-clean electricity by the park, e i,t is the carbon potential of node i where the park is located at time t, P 0,t is the actual electricity purchase volume of the park at time t, i.e., the tie-line power.

[0139] Further, the specific method of step 4 is as follows:

[0140] Lower-level park operator's optimal dispatch model:

[0141] minC all =C E +C c +C com (29)

[0142]

[0143] C com = C cut + C shift + C trans + C EV (31)

[0144] In the formula, C all is the comprehensive cost of electricity and carbon for the operation of the park, C E is the electricity purchase cost of the park. Among them, c t ' is the time-of-use electricity price for the park to purchase electricity from the distribution network, C c is the carbon exchange cost of the park, C com is the compensation cost required during the dispatching process of the park;

[0145] Satisfy the power balance constraint condition of the park:

[0146]

[0147] P 0,t + P PV,t = P L,t (33)

[0148] In the formula, P L,t is the total load of the park at time t, P zy,t is the load not participating in dispatching, P shift,t is the load that can be shifted at time t, P cut,t is the load that can be reduced at time t, is the load of the electric vehicle cluster at time t, P PV,t is the power generation power of the small photovoltaic units in the park at time t.

[0149] Such as Figure 1As shown in the figure, the low-carbon economic operation strategy of the park considering carbon flow demand response is achieved by guiding various types of flexible variable loads based on carbon potential. Thanks to the rapid development of EVs, the EVs of vehicle owners are connected to the EV aggregator at fixed times every day. As a load, it has certain dispatchable potential, and its role can be regarded as a generalized energy storage unit in the park, participating in the economic and low-carbon interaction of the park together. After the individual EV users sign an incentive compensation agreement with the park, under certain time and conditions, the park has the right to control the charging and discharging behaviors of each EV at each time period through the EV aggregator and freely dispatch them. Considering the daily schedules of EV owners in different scenarios, the intelligent charging behavior of EVs can be summarized into two categories: home charging and public charging. In the home scenario, the vehicle owner starts charging after getting home from work and ends charging when getting ready to go to work; in the public scenario, the vehicle owner starts charging when arriving at the workplace in the morning and then completes charging after work. For the convenience of model calculation, it is assumed that the work and rest schedules of each vehicle owner are regular and they go to and from work on time, and the arrival / departure times of each EV are set as fixed values according to the scenario. At the same time, according to the needs of vehicle owners, EVs usually can choose the fast charging mode or the slow charging mode. The slow charging mode can slow down the aging speed of the EV battery to a certain extent.

[0150] The topology of the present invention is as Figure 2 shown. Taking the improved IEEE33 distribution system and the typical park model as examples, micro diesel generators are introduced at nodes 15 and 22 respectively, and photovoltaic units are introduced at nodes 18 and 31. Finally, they are connected to the park at nodes 3 and 15 respectively. The basic carbon price is set at 100 yuan / ton, the carbon quota coefficient is taken as 0.4 tons / MWh, the time step is taken as 1h, and the optimization results of 24h a day are combined to study the low-carbon economic dispatch model of the park proposed in this paper. The capacities of the micro diesel generators DE1 and DE2 in the system are 300kW, the capacities of the photovoltaic units PV1 and PV2 are 500kW and 300kW respectively, and the capacities of the energy storage devices ESS1 and ESS2 are 100kW and 60kW respectively. The linear integer programming problem proposed is solved using the GUROBI solver on the GAMS optimization platform.

[0151] The carbon potential changes of each node in the distribution network are as Figure 3 shown. In practice, the carbon emission form of the power system is mainly the carbon dioxide gas generated by coal and gas at the generator side. It is virtualized and equivalently regarded as the carbon emission flow flowing from the power source side to the load side along with the power flow. By calculating the carbon potential of each node in the system, the actual carbon emissions generated by the power source side corresponding to the electricity consumption at the load side of this node can be known. According to the carbon emission flow theory, the carbon potential of node i in the power system can be calculated through the proportional sharing principle:

[0152]

[0153] In the formula: and Ti + respectively represent the set of branches and the set of units that are all connected to node i and inject active power into it; G s represents the active power output of generator set s; e s represents the carbon emission intensity of generator set s; P j→i represents the active power flowing from branch ji to node i; e j represents the carbon potential of node j on branch ji.

[0154] Since the distribution network is radial and has no circulating current, the network loss will not affect the transmission loss and its distribution of the carbon flow. If there is no discharge equipment connected to the system, the carbon potential of all nodes is equal to the node carbon potential of the root node; if there is discharge equipment such as energy storage or distributed generators connected to the system, the carbon potential of the downstream nodes in the distribution network will be affected. Therefore, for the distribution network, the carbon emission flow can be directly calculated by recursion from the upstream nodes to the downstream nodes, omitting the complex matrix inversion operation and making the calculation more convenient.

[0155] There are two forms of introducing carbon emission flow into the distribution network: one is the power purchase behavior of the root node from the main grid, and the other is the non-clean distributed power sources connected to the downstream nodes. First, the distribution network is economically dispatched according to the predicted data of the 24-hour load, and then the distribution of the carbon emission flow in the distribution network is calculated. Figure 3 The calculation results in [reference] show that the carbon potential of the nodes connected with distributed power sources will change accordingly, and at the same time affect the carbon potential of adjacent nodes. There are significant differences in the carbon potential changes of different nodes during a day. The carbon potential of the nodes connected with photovoltaic units and their adjacent nodes has decreased significantly due to their proximity to clean energy; while the carbon potential of the nodes connected with diesel engines and their adjacent nodes maintains a relatively high level during the day.

[0156] Such as Figure 4 shown, considering the feasibility of the "source-load interaction" of the carbon flow theory, the low-carbon economic dispatch strategy of the park proposed by this method requires the interaction between the upper-layer distribution network and the lower-layer park to drive the energy conservation and emission reduction of the park. The carbon flow is similar to demand response and time-of-use electricity price and demand response. The difference is that one is for low-carbon dispatch and the other is for economic dispatch. This model framework includes two major dispatch entities: the upper-layer 10kV distribution network and the lower-layer 380V park. The dispatchable objects of the former include the micro-diesel generator sets and energy storage connected to the distribution network, which belong to the directly controlled resources on the grid side to achieve the optimal operation economy.

[0157] The upper-layer distribution network operator first conducts the optimal economic dispatch of each unit according to the predicted electricity demand information of each user and the park in the lower layer, calculates the carbon flow distribution through the power flow data after dispatch, and obtains the carbon potential of each node at each time period; the lower-layer park can dispatch EVs and flexible loads for demand response according to the carbon potential change and time-of-use electricity price information of the node where it is located to achieve the purpose of economic low-carbon. The dispatch objective functions of the two are respectively:

[0158]

[0159] Wherein, c t is the time-of-use electricity price of the superior power grid; P sub,t represents the power purchase from the superior power grid at time t; c DE is the power generation cost coefficient of the diesel engine; P DE,t is the output power of the diesel engine.

[0160] minC all = C E + C c + C com

[0161] Wherein, C E is the electricity purchase cost of the park, where c t ' is the time-of-use electricity price for the park to purchase electricity from the distribution network; C c is the carbon exchange cost of the park; C com is the compensation cost required during the park dispatching process, including the compensation for the load that can be curtailed, shifted, and EV cluster users.

[0162] C com = C cut + C shift + C trans + C EV

[0163] In the park, there are photovoltaic units with a capacity of 100 kW and a certain number of EVs. The predicted photovoltaic output value and the typical daily load data of the park are as Figure 5 shown. Considering the long duration of the EV charging behavior and in order to slow down the battery aging as much as possible, a slow charging mode is uniformly adopted, and the upper limit of its charging and discharging power is 7 kW. Next, the load settings of the two parks will be specifically described:

[0164] Park 1: Mainly for industrial production. Its load types are relatively comprehensive, and the EV load is connected to the aggregator to participate in intelligent charging during the working hours of employees.

[0165] Park 2: Mainly for residential life. Since the residents go to work outside during the day, there is almost no dispatchable load in the park except for some important loads. At night, the users go home to rest, and the load flexibility is low. Only the shiftable load that can be shifted as a whole and the EV load that participates in intelligent charging at night are considered.

[0166] To verify that the method in this paper has a certain universality, considering the differences in carbon potential presented by each node in the distribution network at different times, the carbon potential change curves of the nodes where Park 1 and Park 2 are located are as Figure 6As shown. For these two nodes, the periods with relatively high carbon potential during a day are also different. For Node 3, the carbon potential is relatively high from 10:00 to 19:00, while for Node 15, the carbon potential is relatively high at 8:00, from 18:00 to 23:00.

[0167] Three scheduling modes are set to verify the effectiveness of the model. Mode (a): The low-carbon economic scheduling of the park based on carbon flow demand response proposed in this paper; Mode (b): The low-carbon scheduling of the park that only considers carbon emission reduction while ignoring the operating cost; Mode (c): The traditional economic scheduling of the park without considering carbon emission reduction.

[0168] For the above scenarios, the total economic cost and indirect carbon emissions after the park scheduling are calculated respectively. The data of the two after scheduling refer to Table 1, where the total economic cost includes the scheduling operation cost and the carbon exchange cost.

[0169] Table 1 Scheduling results under each mode

[0170]

[0171] The costs generated by the scheduling of each type of load and the carbon emission reduction benefits brought by the proposed method refer to Table 2, where the power change amount refers to the sum of the absolute values of the power change amounts of each type of load at each moment of the day due to scheduling.

[0172] Table 2 Flexible load scheduling cost and emission reduction benefits

[0173]

[0174] As can be seen from Table 2, in terms of scheduling cost, since the total power consumption of shiftable and transferable loads is actually not affected, and the aging rate of the battery during the charging and discharging process of EVs is also very slow, their scheduling costs are slightly lower. While for curtailable loads, certain production cuts are inevitably brought during the scheduling process, and their subsidy costs are relatively high; in terms of carbon emission reduction benefits, if the shiftable and transferable loads are originally in periods with low carbon potential, the carbon emission reduction benefits they bring will not be very significant. While for the intelligent charging of EVs, due to its long time span and more flexible charging and discharging, it has excellent emission reduction effects.

[0175] The tie-line power can directly reflect the consumption situation of non-clean electric energy in the park, and at the same time represents the net load result after the park scheduling. In order to reflect the influence of carbon potential on the park scheduling, the power purchased from the distribution network through the tie-line of Park 1 at different modes from 10:00 to 18:00 is analyzed. As Figure 7 shown, it can be found that the flexible adjustable loads in the park in the low-carbon economic scheduling and low-carbon scheduling modes are both guided by the carbon potential signal, and the carbon cost constraint is considered in the objective function. Park 1 from 12:00 to 18:00 (E NThe electricity purchase behavior of ≥0.7 kgCO2·(kW·h)-1) has been reduced to a certain extent; since low-carbon dispatching does not consider economic costs, its electricity purchase power directly reflects the real-time change of the carbon potential. During the period of low carbon potential but high electricity price, the electricity purchase volume increases instead of decreasing, resulting in unnecessary economic losses.

[0176] Figure 8 It shows the changes in the EV intelligent charging power in Park 1 under three different dispatching modes, fully demonstrating the high flexibility of EV intelligent charging in participating in demand response. The discharging periods of the EV clusters under the three modes have been shifted to different extents between 9:00 - 17:00, which is related to the low-carbon incentive benefits brought by the reduction in electricity purchase. At the same time, both (a) and (b) are charged during the period with the lowest carbon potential, achieving a certain emission reduction effect.

[0177] The applicant of the present invention has made a detailed description and illustration of the embodiments of the present invention in combination with the accompanying drawings. However, those skilled in the art should understand that the above embodiments are only the preferred implementation schemes of the present invention. The detailed description is only to help readers better understand the spirit of the present invention, rather than a limitation on the protection scope of the present invention. On the contrary, any improvement or modification based on the spirit of the present invention should fall within the protection scope of the present invention.

Claims

1. A low-carbon electricity operation method for a park considering carbon flow demand response, characterized in that: The method comprises the following steps: Step 1: Obtain the operating parameters of the upper distribution network and the lower park, including the parameters of each unit, each load forecast curve, the time-of-use electricity price between the upper and lower layers, and the renewable energy output forecast data, and model each type of load in the lower park; Step 2: The upper distribution network performs power dispatching according to the load forecast power consumption curve to obtain the steady-state power flow distribution of the distribution network; Step 3: Calculate the carbon potential curve of the node where the lower-level park is located according to the steady-state power flow distribution of the upper-level distribution network, allocate the initial carbon emission quota to the park and generate a carbon exchange mechanism model; Step 4: The lower-level park comprehensively considers the changes in electricity prices and carbon potential in different periods, conducts low-carbon electricity scheduling, and generates a linear integer programming problem with the sum of electricity cost and carbon exchange cost as the target based on the park operation constraints and various load model constraints; Step 5: The distribution network completes carbon exchange with the park based on the low-carbon electricity consumption curve of the lower-level park on the same day, and uses the GUROBI solver on the GAMS optimization platform to solve this linear integer programming problem.

2. According to the low-carbon electricity operation method for a park considering carbon flow demand response as described in claim 1, it is characterized in that: In step 1, the load types of the lower park are modeled as follows: 1) Electric vehicle load Electric vehicle loads must meet the following constraints when connected to the park: In the formula, subscript i is the serial number of the electric vehicle, subscript t is the scheduling period, and superscripts C and D represent the charging and discharging states, respectively. and are the charging / discharging power of electric vehicle i in period t, and are the charging / discharging efficiency of electric vehicle i in period t, E i,t is the power of electric vehicle i at time t, E i,leave is the amount of electricity required for electric vehicle i to leave the charging station, and are the upper and lower limits of the charging power of electric vehicle i, and are the upper and lower limits of the discharge power of electric vehicle i, and are the upper and lower limits of the electric power of electric vehicle i respectively; At the same time, corresponding compensation is made for vehicle discharge: In the formula, C EV is the electric vehicle discharge compensation cost, n is the number of electric vehicles, It is the compensation coefficient of unit power of electric vehicle discharge. 2) Load reduction possible The load scheduling constraints that can be reduced are as follows: In the formula, T = 24 is the length of the time interval of a day, u t is a 0-1 state variable, u t =1 means that the load is reduced in period t, and are the minimum and maximum durations that the load must meet when responding to shedding. At the same time, load shedding is compensated: In the formula, C cut Compensation costs for load reduction, is the load power reduction in time period t, is the subsidy received for load response reduction per unit power, and A fixed subsidy for cuts that occur over a period of time. 3)Transferable load The working time continuity constraints of the shiftable load scheduling are as follows: Where, τ is the starting time after load translation, t s y is the continuous working time; t is a 0-1 state variable, y t =1 indicates that there is a translational afterload in time period t; At the same time, load translation is compensated: In the formula, C shift Compensate for the load shift cost, is the subsidy that the user receives after shifting the load of unit power, t sh - and t sh+ are the initial and final working periods of the translatable load in a day, respectively, t shift is the load after translation in time period t. 4) Transferable load The transferable load scheduling constraints are as follows: In the formula, and are the transferable load powers before and after the dispatch period t, and are the upper and lower limits of the power after load transfer, υ t is a 0-1 state variable to determine whether load transfer occurs, υ t =1 means that the load is transferred in time period t, t tr- and t tr+ are the initial and final working periods of the transferable load in a day, is the minimum continuous running time; At the same time, load transfer is compensated: In the formula, C trans Compensate for load shifting costs, The compensation price for unit power load transfer.

3. According to the low-carbon electricity operation method for a park considering carbon flow demand response as described in claim 2, it is characterized in that: The upper-layer distribution network operator scheduling model in step 2: 1) Micro diesel generator operation constraints: |P DE,t+1 -P DE,t |≤ΔP DE,max (18) Where P DE,t and Q DE,t are the active and reactive power generated by the diesel generator, P DE,max and P DE,min The upper and lower limits of the active power emitted by the diesel generator, corresponding to Q DE,max and Q DE,min is the upper and lower limits of reactive power, ΔP DE,max The upper limit of the diesel generator's climbing / sliding power; 2) Distribution network energy storage operation constraints: The energy storage operation constraints are similar to those of electric vehicle loads and satisfy equations (1)-(5). 3) Distribution network operation constraints: In i,min ≤V i,t ≤V i,max (22) l ij,t ≤l ij,max (23) Where P ij,t and Q ij,t are the active and reactive power on branch ij during period t, r ij and x ij are the resistance and reactance on branch ij, respectively; 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 first end nodes of the branch with node i as the last end node; l ij,t is the square of the current amplitude on branch ij during period t, V i,t is the square of the voltage amplitude at node i during period t, V i,max and V i,min is the upper and lower limits of the node voltage amplitude, l ij,max is the upper limit of the square of branch current. The upper-layer distribution network operator’s dispatch objective function: In the formula, C D is the electricity cost of the distribution network operation, c t is the time-of-use electricity price of the upper power grid, P sub,t represents the power purchased from the upper grid during period t, c ES is the unit power cost of energy storage charging and discharging, c DE is the diesel engine power generation cost coefficient, P ch,t and P dis,t are the charging and discharging power of the energy storage in the distribution network during period t, P DE,t is the output power of the diesel engine.

4. According to the method of low-carbon electricity operation in a park considering carbon flow demand response as described in claim 3, it is characterized in that: In step 3, the carbon potential of the node where the park is located is calculated as follows: The carbon potential of node i in the power system is calculated by the proportional sharing principle: In the formula, e i is the carbon potential of node i, and They represent the set of branches and units connected to node i and injecting active power into it; G s Indicates the active output of generator set s; e s represents the carbon emission intensity of generator set s; P j→i represents the active power flowing to node i on branch ji; e j Represents the carbon potential of node j on branch ji.

5. A low-carbon electricity operation method for a park considering carbon flow demand response according to claim 4, characterized in that: In step 3, the initial carbon emission quota is allocated to the lower-level parks in the form of free quotas, and carbon exchange is carried out based on the quotas, actual indirect carbon emissions and the upper-level distribution network: In the formula, is the initial quota; ε e The carbon quota coefficient corresponding to the unit electricity consumption, is the predicted value of the net load active power of the park in period t, C c is the carbon exchange cost of the park, c is the basic carbon price in the carbon exchange market, E c The indirect carbon emissions caused by the actual purchase of non-clean electricity in the park, e i,t is the carbon potential of the node i in the park at time t, P 0,t It is the actual power purchase amount of the park during period t, i.e. the interconnection line power.

6. A low-carbon electricity operation method for a park considering carbon flow demand response according to claim 5, characterized in that: The specific method of step 4 is as follows: Lower-level park operator optimization scheduling model: minC all =C E +C c +C com (29) C com =C cut +C shift +C trans +C EV (31) In the formula, C all is the comprehensive cost of electricity and carbon in the park operation, C E is the electricity purchase cost of the park, where c t ' is the time-of-use electricity price of the park when purchasing electricity from the distribution network, C c is the carbon exchange cost of the park, C com The compensation costs required in the park scheduling process; Meeting the power balance constraints of the park: P 0,t +P PV,t =P L,t (33) Where P L,t is the total load of the park during period t, P zy,t is the load not involved in the dispatch, P shift,t is the load that can be translated during period t, P cut,t is the load that can be reduced during period t, is the electric vehicle cluster load during period t, P PV,t is the power generation power of the small photovoltaic unit in the park during period t.

Images