Power-heat integrated energy system station network joint planning method based on planning decision characteristics

Abstract

The invention relates to an electric-thermal integrated energy system station network joint planning method based on planning decision characteristics, which comprises the following steps: substituting a prediction day set into a single-day planning model to obtain a planning decision set, generating a representative planning decision set by adopting a clustering algorithm, selecting a representative day according to the representative planning decision set, and establishing a representative day planning model; solving the representative day planning model, obtaining a preliminary planning decision without considering the extreme day, executing the preliminary planning decision on each prediction day, calculating the total loss load amount of each prediction day, and taking the prediction day with the maximum total loss load amount as the extreme day; and establishing an electric-thermal integrated energy system station network joint planning model by taking minimization of the total cost as a target, substituting the representative day and the extreme day into the electric-thermal integrated energy system station network joint planning model, and solving to obtain a final planning decision. Compared with the prior art, the reliability of station network joint planning of the electric-thermal integrated energy system can be further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated energy system planning and design, and in particular relates to a station-network joint planning method for an electric-thermal integrated energy system based on planning decision characteristics. Background Art

[0002] As the only way to achieve energy transformation, the integrated energy system (IES) has developed rapidly in recent years. IES is a multi-type energy supply system composed of a variety of energy supply equipment and various energy distribution networks that are highly coupled. Compared with traditional energy systems, it improves energy utilization efficiency, reduces carbon emissions, and enhances the reliability of energy supply. Therefore, studying the joint planning of IES stations and networks is of great significance to the construction and development of energy systems.

[0003] The equipment installed capacity and the architecture of the distribution network within the IES are usually designed with the goal of minimizing costs or carbon emissions. Modeling the IES station network joint planning problem as a mixed integer linear programming (MILP) problem is more in line with actual engineering needs. However, since planning requires reference to a large amount of forecast data, solving the MILP model takes a lot of time or is even unsolvable. Therefore, it is necessary to select some representative days or representative time periods to represent the entire forecast data set for planning.

[0004] However, the current research on the selection of representative days and extreme days in the joint planning of IES station networks still has the following problems: 1) The selection method of representative days and extreme days is overly dependent on the numerical characteristics of the forecast data, lacks consideration of the nonlinear characteristics of the planning model, and the selected representative days may be relatively unrepresentative; 2) The selection process of representative days and extreme days is relatively independent, and there is no reasonable method to combine the two in the planning process. The selection of extreme days ignores the actual operation performance of the system. In summary, it is necessary to design a new method for the joint planning of the station network of the electric-thermal integrated energy system to further improve the reliability of the joint planning of the station network of the electric-thermal integrated energy system. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics, fully considering the interdependent characteristics and nonlinear characteristics of the planning model, and further improving the reliability of joint planning of power-heat integrated energy system stations and networks.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] The present invention provides a method for joint planning of a power-heat integrated energy system station network based on planning decision characteristics, comprising the following steps:

[0008] Establish a single-day planning model, obtain a set of predicted days, substitute the set into the single-day planning model to obtain a planning decision set, use a clustering algorithm to generate a representative planning decision set, select a representative day and establish a representative day planning model based on the set, the representative day contains nonlinear information of the single-day planning model;

[0009] Solving the representative day planning model to obtain a preliminary planning decision without considering extreme days, constructing a single-day operation model, executing the preliminary planning decision for each forecast day, calculating the total amount of load loss for each forecast day, and taking the forecast day with the largest total amount of load loss as the extreme day;

[0010] A station-network joint planning model for an electric-thermal integrated energy system is established with the goal of minimizing the total cost, wherein the total cost includes the investment cost and the operating cost. The representative day and the extreme day are substituted into the station-network joint planning model for the electric-thermal integrated energy system to obtain the final planning decision.

[0011] Furthermore, the expression of the single-day planning model is as follows:

[0012]

[0013] Among them, C represents the total cost, C inv represents the investment cost, represents the operating cost at time t, M T Represents the total number of forecast data.

[0014] Furthermore, the representative day selection process is as follows:

[0015] S101, splitting the forecast day set to obtain a number of independent forecast days, each forecast day including digital features of renewable energy power generation and various load demand levels on that day;

[0016] S102, substituting each forecast day into the single-day planning model separately, and solving to obtain the planning decision for each forecast day, including the installed capacity of various energy supply equipment and the construction plan of the interconnection line and the pipeline network;

[0017] S103, normalizing the planning decisions of all forecast days, and performing K-mediods clustering to obtain a number of representative planning decisions and a weight corresponding to each representative planning decision;

[0018] S104. Select the forecast day corresponding to each representative planning decision as a representative day, and the weight of each representative day is the weight of the corresponding representative planning decision.

[0019] Furthermore, the expression of the representative daily planning model is specifically as follows:

[0020]

[0021] Among them, C represents the total cost, C inv represents the investment cost, represents the operating cost at time t on the nth representative day, N represents the number of representative days, ω n represents the weight of the nth representative day, M T Represents the total number of forecast data.

[0022] Furthermore, the expression of the single-day operation model is as follows:

[0023]

[0024] Where C represents the total cost, represents the operating cost at time t.

[0025] Furthermore, in the power-heat integrated energy system station-network joint planning model, the expression of the investment cost is specifically as follows:

[0026]

[0027] Among them, C inv represents the investment cost, D g Indicates the installed capacity of device g, M g represents the unit capacity installation cost of device g, Γ represents the device set, N l Indicates the number of line construction, M l represents the construction cost of line l;

[0028] The operating cost includes the representative day planning operating cost and the single day planning operating cost, and the expressions are as follows:

[0029]

[0030] in, represents the daily planned operating cost, and They represent the energy purchase cost, load loss cost and wind and solar curtailment cost at the tth moment on the nth representative day respectively; Indicates the single-day planning and operation cost, and They represent the energy purchase cost, load loss cost and wind and solar power abandonment cost at the tth moment respectively.

[0031] Furthermore, the constraints of the station-network joint planning model of the electric-thermal integrated energy system include equipment constraints, distribution system constraints, and thermal system constraints.

[0032] Furthermore, the equipment constraints include equipment output constraints, equipment ramp constraints, equipment investment constraints, cogeneration unit constraints and energy storage equipment constraints.

[0033] Furthermore, the distribution system constraints include DistFlow flow constraints, line flow constraints, node voltage constraints, power failure load constraints and distribution system radial constraints.

[0034] Furthermore, the thermal system constraints include node heat exchange constraints, node temperature mixing constraints, node temperature constraints, heat loss load constraints and heat abandonment constraints.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. The present invention proposes a joint planning method for a power-heat integrated energy system station network based on planning decision characteristics. On the one hand, the planning decision contains the nonlinear information of a single-day planning model. The present invention selects representative days based on the planning decision characteristics, which can fully consider the nonlinear characteristics of the planning model and make the obtained planning decisions closer to the optimal solution; on the other hand, the present invention selects extreme days based on operating results. Compared with selecting extreme days directly based on predicted data, the selected extreme days can be more objective, the load loss can be greatly reduced, and the generated planning decisions can be more reliable.

[0037] 2. The present invention normalizes the planning decisions of all forecast days, which can eliminate the influence of the large difference in magnitude between planning decision data on the clustering results, and then performs K-mediods clustering to obtain several representative planning decisions, which can improve the accuracy of clustering and make the selected representative days more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Schematic diagram of the selection method of representative days and extreme days in the present invention;

[0039] Figure 2 A schematic diagram of an electric-thermal integrated energy test system used in an example of the embodiment;

[0040] Figure 3 Schematic diagram of mean absolute percentage error (MAPE) of different representative day selection methods;

[0041] Figure 4 To add a schematic diagram of the impact of extreme days on the running results;

[0042] Figure 5 The power balance diagram on extreme days for planning decisions that do not consider extreme days;

[0043] Figure 6 Power balance diagram on extreme days for planning decisions considering extreme days. DETAILED DESCRIPTION

[0044] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0045] Example:

[0046] This embodiment provides a method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics. The innovations are: (1) A method for selecting representative days based on planning decision characteristics is proposed, and the clustering objects are converted from prediction data characteristics to planning decision characteristics to obtain representative days containing nonlinear information of the planning model; (2) The preliminary planning decisions are used to execute the operation problems of each prediction day, and extreme days are selected according to the amount of load loss and incorporated into the operation constraints of the planning model to obtain planning decisions that take extreme days into consideration. The above method can obtain more accurate planning decisions using only a small number of representative days. At the same time, planning decisions that take extreme days into consideration can significantly reduce the amount of load loss. The specific steps of this method are as follows:

[0047] S1. Selection of representative days based on planning decision characteristics.

[0048] The planning method based on representative days reduces the computational burden of the mixed integer linear programming (MILP) problem by selecting several representative days that can represent the entire prediction data set based on relevant features and assigning them weights. It can achieve a certain planning accuracy while improving planning efficiency.

[0049] like Figure 1 As shown in the figure, first obtain the set of forecast days, then substitute it into the single-day planning model to obtain the planning decision set, use the clustering algorithm to generate the representative planning decision set, and then select the representative day and establish the representative day planning model. The representative day selection process is as follows:

[0050] S101, M T The prediction data is split to obtain M T / 24 independent forecast days, each containing numerical characteristics of renewable energy generation and various load demand levels for that day.

[0051] S102. Substitute each forecast day into the single-day planning model to obtain the planning decision for each forecast day, including the installed capacity of various energy supply equipment and the construction plan of the interconnection line and pipeline network.

[0052] The expression of the single-day planning model is as follows:

[0053]

[0054] Where C represents the total cost; C inv represents the investment cost; represents the operating cost at time t; and represents the variables related to investment cost; A and B represent the coefficients of the variables related to investment cost; and represents the variables related to operating cost; E and F represent the coefficients of the variables related to operating cost; p and q represent the thresholds of the variables related to operating cost and investment cost, respectively; V inv and V ope They respectively represent the sets to which investment cost related variables and operating cost related variables belong.

[0055] S103, normalizing the planning decisions of all forecast days to eliminate the influence of the large difference in magnitude between planning decision data on the clustering results, and then performing K-mediods clustering to obtain N representative planning decisions and the weight corresponding to each representative planning decision.

[0056] S104. Select the forecast day corresponding to each representative planning decision as the representative day.

[0057] The representative day planning model can be regarded as a mapping relationship between representative days and planning decisions. Due to the complex equipment and network modeling inside the planning model, this mapping relationship has a strong nonlinear characteristic. In order to select a suitable representative day, the nonlinear characteristics of this planning model cannot be ignored. The representative day selected according to the above method contains the nonlinear information of the single-day planning model, so that the subsequently generated planning decisions can contain the nonlinear information of the planning model. The weight of each representative day is the weight of its corresponding representative planning decision, and the following representative day planning model is constructed:

[0058]

[0059] in, represents the operating cost at time t on the nth representative day, N represents the number of representative days, ω n Represents the weight of the nth representative day.

[0060] S2. Operation result-oriented extreme day selection.

[0061] Considering extreme days helps ensure that the final decision can make the system run safely, stably and efficiently. In order to objectively select extreme days, such as Figure 1As shown, this embodiment first solves the representative day planning model to obtain a preliminary planning decision without considering extreme days. Then, a single-day operation model such as formula (3) is constructed to execute preliminary planning decisions for each forecast day. Since the amount of load loss is a comprehensive reflection of the IES operation process, it is possible to locate the weak links of the IES planning decision when each forecast day is running under the existing planning decision. Therefore, this embodiment calculates the total amount of load loss for each forecast day, and takes the forecast day with the largest total amount of load loss as the extreme day.

[0062]

[0063] Since the extreme days selected by this extreme day selection method do not have weights, it is only necessary to include the selected extreme days in the operating constraints of the single-day operation model. During planning, the investment cost is considered without considering the operating cost to meet the energy supply on the extreme days, and finally a planning decision that takes extreme days into consideration is obtained.

[0064] S3. A joint planning model for the power-heat integrated energy system is established with the goal of minimizing the total cost. Representative days and extreme days are substituted to obtain the final planning decision.

[0065] Total cost C includes investment cost C inv and operating cost C ope , the objective function of the joint planning model of the power-heat integrated energy system station network is as follows:

[0066] minC=C inv +C ope (4)

[0067] Investment cost inv The specific expression is as follows:

[0068]

[0069] Where D g Indicates the installed capacity of device g, M g represents the unit capacity installation cost of device g, Γ represents the device set, N l Indicates the number of line construction, M l Represents the construction cost of line l.

[0070] Operating cost ope Including the representative daily planning and operation cost and the single-day planning and operation cost, the expressions are as follows:

[0071]

[0072] In the formula, and They represent the energy purchase cost, load loss cost and wind and solar curtailment cost at the tth moment of the nth representative day, and They represent the energy purchase cost, load loss cost and wind and solar power abandonment cost at the tth moment respectively.

[0073] The constraints of the joint planning model of the power-heat integrated energy system station network include equipment constraints, distribution system constraints, and thermal system constraints, as follows:

[0074] Ⅰ. Equipment Constraints

[0075] Equipment output constraints:

[0076]

[0077] In the formula, They represent the upper and lower output limits of device g, respectively. n,t,g Represents the output of device g at time t on day n.

[0078] Equipment climbing constraints:

[0079] -Ramp g ≤P n,t,g -P n,t-1,g ≤Ramp g (9)

[0080] Where Ramp g Indicates the climbing rate of device g.

[0081] Equipment investment constraints:

[0082]

[0083] In the formula, Indicates the maximum installed capacity of device g.

[0084] Cogeneration unit constraints:

[0085]

[0086] Where η CHP represents the heat-to-electricity ratio of the cogeneration unit, They respectively represent the heat supply and power supply of the cogeneration unit at node j at time t on the nth day.

[0087] Energy storage equipment constraints:

[0088]

[0089] In the formula, E n,j (t) represents the energy storage device power at node j at time t on day n, η ESIt represents the charging and discharging efficiency of the energy storage device, and γ represents the self-discharge rate of the energy storage device.

[0090]

[0091] In the formula, They represent the charging and discharging amount of the energy storage device at node j at time t on day n, It is a 0-1 variable, indicating the charge and discharge state of the energy storage device, and τ indicates the capacity-to-charge ratio of the energy storage device.

[0092] E n,j (0) = E n,j (24) (16)

[0093] Formula (16) is the head-to-tail balance constraint of the energy storage device during the day.

[0094] II. Distribution System Constraints

[0095] DistFlow power flow constraints:

[0096]

[0097]

[0098] In the formula, r ij 、x ij Represents the resistance and reactance of line ij, P n,t,ij , Q n,t,ij ,I n,t,ij They represent the active power, reactive power and current flowing through line ij at time t on day n. n,t,j , Q n,t,j They represent the active power and reactive power injected into node j at time t on day n, respectively. n,t,j represents the voltage of node j at time t on day n. f(j) represents the set with node j as the parent node. Equations (17)-(19) are non-convex and nonlinear and need to be converted into a second-order cone optimization model that can be solved by commercial solvers.

[0099] Line flow constraints:

[0100]

[0101] In the formula, α ij It is a 0-1 variable, indicating the construction flag of line ij. “0” indicates that the line is not under construction, and “1” indicates that the line is under construction. They respectively represent the upper limits of active load, reactive load and current transmitted on the line.

[0102] Node voltage constraints:

[0103] Vmin ≤V n,t,j ≤V max (twenty three)

[0104] Where, V n,t,j represents the voltage of node j at time t on day n, V min 、V max Indicates the minimum and maximum values ​​allowed for the voltage.

[0105] Power failure load constraint:

[0106]

[0107] In the formula, They represent the active load loss and active load of node j at time t on the nth day respectively.

[0108] Constraints on curtailment of wind and solar power:

[0109]

[0110] In the formula, It represents the amount of wind and solar power curtailment at node j at time t on day n.

[0111] Radial constraints of power distribution system:

[0112]

[0113] Where n e is the number of nodes in the power distribution system, f(i) represents the set of all nodes with node i as the parent node, is a 0-1 variable. The distribution system needs to be radial when modeled using the DistFlow power flow model. Equations (26)-(29) can keep the distribution system architecture radial.

[0114] III. Thermal System Constraints

[0115] Node heat exchange constraints:

[0116]

[0117] In the formula, C w Indicates the specific heat capacity of water, take 4.2kJ / (kg*℃). j Represents the flow rate of heat exchange node j. φ n,t,j , They represent the heat exchange capacity, water supply node temperature and return water node temperature of node j at time t on day n respectively.

[0118] Nodal temperature mixing constraints:

[0119]

[0120] In the formula, Represents the outlet water temperature of pipe p at time t on the nth day.

[0121] Node temperature constraints:

[0122] T min ≤T n,t,j(p) ≤T max (32)

[0123] Where, T min 、T max They represent the design maximum and minimum values ​​of the node or pipe temperature respectively.

[0124] Heat loss load constraint:

[0125]

[0126] Heat rejection constraint:

[0127]

[0128] To verify the effectiveness of the above method, this embodiment uses IES consisting of IEEE 6-node power distribution system and 6-node thermal system as the test system. Figure 2 The planned projects include the installed capacity of gas turbines (GT), the installed capacity of combined heat and power units (CHP), the installed capacity of energy storage equipment (ES), and the construction plan of the interconnection line of the distribution system. The unit installation cost of each equipment is shown in Table 1, and the construction cost of the distribution network interconnection line is shown in Table 2.

[0129] Table 1 Equipment installation cost

[0130]

[0131] Table 2 Tie line construction cost

[0132]

[0133] In order to reduce the computational burden of the example and obtain the optimal planning scheme, this example processes 2880 hours of historical data recorded in a heating season in a certain area in the northwest, including historical data of wind turbines and photovoltaic power generation and historical data of two loads: electric and thermal. After processing with prediction technology, 2880 hours of prediction data is obtained, and then the station-network joint planning of the electric-thermal integrated energy system is carried out.

[0134] In order to compare the performance of the two representative day selection methods based on the predicted data characteristics and the planning decision characteristics, this example designed 12 groups of examples to compare the planning decisions obtained by selecting different numbers of representative days and extreme days for planning in different selection methods, and compared the impact of considering extreme days and not considering extreme days on planning decisions. The planning decisions obtained in each group of experiments are recorded in Table 3.

[0135] Table 3 Comparison of planning decisions in different examples

[0136]

[0137]

[0138] From Table 3, we can find that compared with the optimal planning decision based on the full-time forecast data, the representative day planning method will lead to the underestimation of the installed capacity of energy supply equipment such as gas turbines and cogeneration units, while the installed capacity of energy storage equipment will be overestimated. This is because the planning only refers to the representative day obtained by clustering, and does not take into account all scenarios in the full time period. In order to be more economical and meet the load demand of each representative day, the representative day planning method will tend to use cheaper energy storage equipment to meet the load supply, thereby reducing the installed capacity of expensive energy supply equipment and reducing investment costs.

[0139] Comparing the two representative day selection methods, there is no obvious difference in planning costs between the two representative day selection methods. When the same number of representative days is used, the installed capacity of energy supply equipment obtained by the selection method based on planning decision characteristics is larger, which shows that the planning decision obtained by this selection method is more reliable. At the same time, the installed capacity of energy storage equipment is smaller, which also avoids the installation of too much energy storage equipment and reduces the waste of investment costs. In order to quantify the accuracy of planning decisions obtained by the two representative day selection methods, this example uses the mean absolute percentage error (MAPE) to measure the planning decision errors obtained by different selection methods. The expression of this indicator is shown in formula (35):

[0140]

[0141] In the formula, G represents the number of devices, D g represents the installed capacity of device g, Represents the optimal installed capacity of equipment g. MAPE estimates the average deviation of the installed capacity of the equipment. The smaller its value is, the closer the planning decision is to the optimal solution.

[0142] Compare the two representative day selection methods under the premise of using the same number of representative days, such as Figure 3As shown, it can be found that the MAPE based on the planning decision feature selection method is lower than the MAPE based on the forecast data feature selection method, indicating that the representative day selected by this representative day selection method that takes into account the nonlinear characteristics of the planning model itself is more representative, and the obtained planning decision is more accurate than the planning decision based on the forecast data feature selection method.

[0143] The results of the example also show that when 15 representative days are used, the planning decision error obtained based on the planning decision feature selection method is about 6%. As the number of selected representative days increases, the error is basically maintained within 6%, which means that for this planning model, the optimal number of representative days selected based on the planning decision feature selection method is 15. However, as the number of selected representative days increases, the planning decision error continues to decrease based on the selection method based on the predicted data features, which means that more representative days need to be selected to achieve a smaller error using this selection method, which also increases the solution time in disguise.

[0144] Different representative day selection methods will also affect the interconnection line construction plan of the IES distribution system. Table 4 records the distribution system construction plans of different examples, where "1" means line construction and "0" means line non-construction.

[0145] Table 4 Tie line construction schemes for different examples

[0146]

[0147] It can be found that when more than 10 representative days are selected based on planning decision characteristics, the connection line construction plan obtained is always consistent with the connection line construction plan of the optimal planning decision. However, the connection line construction plan obtained by the representative day selection method based on the predicted data characteristics keeps changing with the increase in the number of selected representative days, which makes it difficult for planners to select the best connection line construction plan.

[0148] Therefore, the representative day selection method based on planning decision characteristics can use as few representative days as possible to obtain equipment installed capacity that is closer to the optimal planning decision and a distribution system interconnection line construction plan that is the same as the optimal planning decision. This selection method can improve planning efficiency while obtaining better planning decisions than conventional selection methods.

[0149] When planning IES, considering extreme days can improve the reliability of planning decisions, so that IES can guarantee the energy supply of various loads during operation and reduce the occurrence of load loss events. Figure 4The load loss during the whole heating season of the planning decisions of 15 representative days and 15 representative days + 1 extreme day was compared. It can be seen that under the premise of selecting 15 representative days, the planning decisions considering extreme days have a much smaller number of load loss days than those not considering extreme days, and the maximum and average values ​​of load loss are reduced, which shows that considering extreme days has a significant improvement in the reliability of planning decisions.

[0150] This example also compares the performance of the two planning decisions on extreme days. Figure 5 The operation results of the planning decisions obtained without considering extreme days on extreme days show that there is a large shortage of electricity load. Figure 6 The operating results of the planning decisions made to consider extreme days on extreme days show that there is no load shortfall when operating on extreme days. This is because incorporating extreme days into constraints in the representative day planning can make the planned equipment installed capacity meet the load demand on extreme days, and the load shortfall can also be filled.

[0151] In summary, adding extreme days when planning IES can improve the operational reliability of IES on the full time scale, which will lead to increased investment costs, but its planning decisions can ensure that the system meets load demand and operates safely.

[0152] The above description of the embodiments is to facilitate the understanding and use of the invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. A method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics, characterized in that: The following steps are involved: Establish a single-day planning model, obtain a set of predicted days, substitute the set into the single-day planning model to obtain a planning decision set, use a clustering algorithm to generate a representative planning decision set, select a representative day and establish a representative day planning model based on the set, the representative day contains nonlinear information of the single-day planning model; Solving the representative day planning model to obtain a preliminary planning decision without considering extreme days, constructing a single-day operation model, executing the preliminary planning decision for each forecast day, calculating the total amount of load loss for each forecast day, and taking the forecast day with the largest total amount of load loss as the extreme day; A station-network joint planning model for an electric-thermal integrated energy system is established with the goal of minimizing the total cost, wherein the total cost includes the investment cost and the operating cost. The representative day and the extreme day are substituted into the station-network joint planning model for the electric-thermal integrated energy system to obtain the final planning decision.

2. According to claim 1, a method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics is characterized in that: The expression of the single-day planning model is as follows: Among them, C represents the total cost, C inv represents the investment cost, represents the operating cost at time t, M T Indicates the total number of predicted data.

3. According to claim 1, a method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics is characterized in that: The selection process of the representative day is as follows: S101, splitting the forecast day set to obtain a number of independent forecast days, each forecast day including digital features of renewable energy power generation and various load demand levels on that day; S102, substituting each forecast day into the single-day planning model separately, and solving to obtain the planning decision for each forecast day, including the installed capacity of various energy supply equipment and the construction plan of the interconnection line and the pipeline network; S103, normalizing the planning decisions of all forecast days, and performing K-mediods clustering to obtain a number of representative planning decisions and a weight corresponding to each representative planning decision; S104. Select the forecast day corresponding to each representative planning decision as a representative day, and the weight of each representative day is the weight of the corresponding representative planning decision.

4. According to claim 1, a method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics is characterized in that: The expression of the representative daily planning model is as follows: Among them, C represents the total cost, C inv represents the investment cost, represents the operating cost at time t on the nth representative day, N represents the number of representative days, ω n represents the weight of the nth representative day, M T Indicates the total number of predicted data.

5. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 1 is characterized in that: The expression of the single-day operation model is as follows: Where C represents the total cost, Represents the operating cost at time t.

6. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 1 is characterized in that: In the station-network joint planning model of the electric-thermal integrated energy system, the expression of the investment cost is specifically as follows: Among them, C inv represents the investment cost, D g Indicates the installed capacity of device g, M g represents the unit capacity installation cost of device g, Γ represents the device set, N l Indicates the number of line construction, M l represents the construction cost of line l; The operating cost includes the representative day planning operating cost and the single day planning operating cost, and the expressions are as follows: in, represents the daily planned operating cost, and They represent the energy purchase cost, load loss cost and wind and solar curtailment cost at the tth moment on the nth representative day respectively; Indicates the single-day planning and operation cost, and They represent the energy purchase cost, load loss cost and wind and solar power abandonment cost at the tth moment respectively.

7. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 1 is characterized in that: The constraints of the station-network joint planning model of the electric-thermal integrated energy system include equipment constraints, distribution system constraints, and thermal system constraints.

8. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 7 is characterized in that: The equipment constraints include equipment output constraints, equipment ramp constraints, equipment investment constraints, cogeneration unit constraints and energy storage equipment constraints.

9. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 7 is characterized in that: The distribution system constraints include DistFlow power flow constraints, line power flow constraints, node voltage constraints, power failure load constraints and distribution system radial constraints.

10. The method for joint planning of power-heat integrated energy system stations and networks based on planning decision characteristics according to claim 7, characterized in that: The thermal system constraints include node heat exchange constraints, node temperature mixing constraints, node temperature constraints, heat loss load constraints and heat abandonment constraints.

Images