Carbon accounting method for optimal allocation of carbon flow in power system based on decision tree C4.5

Abstract

The present invention relates to a carbon metering method for optimizing the allocation of carbon flow in a power system based on a decision tree C4.5, and belongs to the field of carbon metering in a power system. The method is: combining a carbon source layer C, a hub layer S, a branch layer F, and a terminal layer Z, and then optimizing the allocation of carbon flow in a power system based on a decision tree C4.5, to determine the optimization path of the carbon flow; specifically as follows: defining power plants in a specified area as a carbon source layer C, a hub layer S, a branch layer F, and a terminal layer Z; calculating the information entropy H(C) of the carbon source layer C; calculating the information entropy H(S) of the hub layer S; calculating the conditional entropy H(C|S) of the attribute splitting at the substation; calculating the information gain g(C,S) of the hub layer to the carbon source layer; calculating the intermediate information entropy H(C) of the carbon source layer to the hub layer S (C); Calculate the information gain ratio g from the carbon source layer to each hub layer node R (C,S), select the smallest one as the preferred direction of carbon flow from the carbon source layer to the hub layer.

Description

Technical Field

[0001] The invention relates to a carbon metering method for optimizing carbon flow allocation in a power system based on a decision tree C4.5, and belongs to the field of carbon metering in a power system. Background Art

[0002] The power industry is the single industry with the highest carbon emissions in my country, among which thermal power generation accounts for a large proportion, accounting for about 40% of the total carbon emissions in the society. In the future, as the degree of electrification continues to increase, its carbon emissions will continue to increase. At present, there is no effective method to control the carbon flow distribution of the power system, and the carbon flow is affected by the geographical distribution. It is currently in a state of extensive and disorderly management. It is a problem worthy of study in the future of carbon trading, carbon allocation and carbon measurement. It is necessary to comprehensively allocate the carbon flow direction according to the carbon emission factors, carbon emissions, electricity consumption and other factors of each power node to achieve the optimal allocation of carbon flow.

[0003] Among the existing technical solutions, "Integrated Energy Optimal Dispatch of Industrial Parks Considering Extended Carbon Emission Flow and Carbon Trading Bargaining Model" in "Automation of Electric Power Systems" 2023, 9 proposes a comprehensive energy optimal dispatch model for industrial parks based on dual game on both sides of supply and demand. First, the concept of "carbon load rate" of energy storage and the principle of conservation of carbon emissions during the dispatching cycle were introduced to establish an extended carbon emission flow model. Based on this model, a carbon-dominated pricing mechanism is implemented within the industrial park, and a double-layer game model between the industrial park and users is established. The Nash bargaining model is used between multiple industrial parks to simulate the game behavior of various entities participating in the carbon trading market, thereby forming a dual game mechanism.

[0004] The article "Multi-objective and multi-time scale optimization scheduling based on VPP carbon flow calculation" published in "Smart Power" 2024,1 proposed a study on the low-carbon economic operation strategy of virtual power plants, and implemented multi-objective and multi-time scale optimization scheduling based on the energy flow and carbon flow models of virtual power plants. First, a nonlinear multi-objective planning model for the low-carbon economic operation of virtual power plants was established based on the carbon flow trend, and the normalized normal constraint method was used to solve it; then, a finite time domain optimization model based on the dynamic matrix control algorithm was established to track and optimize the day-ahead results and feedback correction to reduce the impact of uncertainty factors; finally, the simulation comparison experiment verified that the proposed scheduling strategy can achieve low-carbon economic operation of virtual power plants under the premise of ensuring that the carbon flow calculation accuracy exceeds 95%.

[0005] The above existing various solutions still have shortcomings, so it is necessary to design a new solution to meet actual needs. Summary of the invention

[0006] In order to overcome the problems of disorder, wide spatial distribution and many branch nodes in the prior art of carbon emissions from electricity, the present invention uses the decision tree C4.5 carbon emission flow optimization allocation calculation model to guide users to change their energy consumption behavior. It is a carbon-dominated energy system optimization operation mode, which realizes the path optimization iteration of the carbon source layer, hub layer, branch layer and terminal layer, and improves the economy and intelligence level of carbon flow allocation in the context of large spatial scale and multiple branch nodes.

[0007] The carbon metering method for optimizing carbon flow allocation in the power system based on the decision tree C4.5 of the present invention is:

[0008] Combining the carbon source layer C, hub layer S, branch layer F and terminal layer Z, the optimal allocation of carbon flow in the power system based on the decision tree C4.5 is used to determine the optimal path of carbon flow; the details are as follows:

[0009] (1) Define the power plants in the specified area as the carbon source layer C, hub layer S, branch layer F and terminal layer Z;

[0010] (2) Calculate the information entropy H(C) of the carbon source layer C;

[0011] (3) Calculate the information entropy H(S) of the hub layer S;

[0012] (4) Calculate the conditional entropy H(C|S) of the substation attribute split;

[0013] (5) Calculate the information gain g(C,S) of the hub layer to the carbon source layer;

[0014] (6) Calculate the intermediate information entropy H of the carbon source layer to the hub layer S (C);

[0015] (7) Calculate the information gain ratio g from the carbon source layer to each hub layer node R (C, S), select the smallest one as the preferred direction of carbon flow from the carbon source layer to the hub layer.

[0016] Further, (8) repeat the above steps (1) to (7) to calculate the information gain ratio g from the hub layer to each branch layer respectively. R (S, F), select the smallest one as the preferred direction of carbon flow from the hub layer to the branch layer.

[0017] Furthermore, (9) repeat the above steps (1) to (8) to calculate the information gain ratio g from the branch layer to each terminal layer respectively. R (F, Z), select the first three electricity meters corresponding to the three smallest information gains as the final direction of the carbon flow from the branch layer to the terminal layer.

[0018] Specifically, in step (1), the step-down substation above 10 kV is the hub layer S; the 10 kV distribution transformer is the branch layer F; the collection equipment including the dedicated transformer collection terminal and concentrator and the electric energy meter is the terminal layer Z.

[0019] Specifically, in step (2), the information entropy H(C) of the carbon source layer C is calculated, and the sunny rate of the area is defined as p1, the cloudy rate is p2, the overcast rate is p3, and the rainy rate is p4.

[0020]

[0021] Specifically, in step (3), the information entropy H(S) of the hub layer S is calculated, and the probability that the average value of the load rate of a certain transformer in the hub layer S is greater than 100% is defined as q1, the probability that the average value of the load rate of the transformer is greater than 75% and less than or equal to 100% is defined as q2, and the probability that the average value of the load rate of the transformer is less than or equal to 75% is defined as q3.

[0022]

[0023] Specifically, in step (4), the conditional entropy H(C|S) of the substation attribute split is calculated, and the information entropy of n substations is averaged.

[0024]

[0025] (3) Specifically, in step (5), the information gain g(C,S) of the hub layer to the carbon source layer is calculated

[0026] g(C,S) = H(C)- H(C|S) (4).

[0027] (3) Specifically, in step (6), the intermediate information entropy H of the carbon source layer to the hub layer is calculated S (C)

[0028]

[0029] Specifically, in step (7), the information gain ratio g from the carbon source layer to each hub layer node is calculated respectively. R (C, S), select the smallest one as the preferred direction of carbon flow.

[0030]

[0031] Wherein GF is the temperature correction coefficient, which is 1 when the temperature is 25 degrees Celsius, 1.1 when the temperature is greater than 25 degrees Celsius and less than or equal to 32 degrees Celsius, 1.2 when the temperature is greater than 32 degrees Celsius, 0.9 when the temperature is greater than 5 degrees Celsius and less than or equal to 35 degrees Celsius, and 0.8 when the temperature is less than 5 degrees Celsius.

[0032] An electronic device includes one or more processors for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors execute the aforementioned carbon metering method for optimizing the allocation of carbon flows in a power system based on the decision tree C4.5.

[0033] A storage medium having a computer program stored therein, wherein the computer program is configured to execute the aforementioned carbon metering method for optimizing carbon flow allocation in a power system based on the decision tree C4.5 when running.

[0034] Compared with the prior art, the carbon metering method for optimizing carbon flow allocation in a power system based on the decision tree C4.5 provided by the present invention has the following characteristics:

[0035] Combining the carbon source layer C, hub layer S, branch layer F, and terminal layer Z, the power system carbon flow optimization allocation based on decision tree C4.5 determines the optimal path of carbon flow. By optimizing the carbon flow allocation, unnecessary carbon emissions can be reduced, thereby reducing the operating costs of enterprises. It can reduce energy consumption and waste gas emissions and improve resource utilization efficiency.

[0036] Considering factors such as information entropy, conditional entropy, information gain at different levels, and then considering the temperature correction coefficient, the node information gain ratio is corrected to determine the optimal flow direction of carbon flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the workflow of the carbon metering method for optimizing carbon flow allocation in the power system based on the decision tree C4.5 of the present invention.

[0038] Figure 2 This is a schematic diagram of the connection status of the carbon source layer C, the hub layer S, the branch layer F and the terminal layer Z in the carbon metering method for optimizing the allocation of carbon flows in the power system based on the decision tree C4.5 of the present invention. DETAILED DESCRIPTION

[0039] The specific implementation plan of this program is as follows:

[0040] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following reference is made to the attached Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the protection scope of the present invention.

[0041] The carbon metering method for optimizing carbon flow allocation in the power system based on the decision tree C4.5 of the present invention is:

[0042] Combining the carbon source layer C, hub layer S, branch layer F and terminal layer Z, the optimal allocation of carbon flow in the power system based on the decision tree C4.5 is used to determine the optimal path of carbon flow; the details are as follows:

[0043] (1) Define the power plants in the specified area as the carbon source layer C, hub layer S, branch layer F and terminal layer Z;

[0044] (2) Calculate the information entropy H(C) of the carbon source layer C;

[0045] (3) Calculate the information entropy H(S) of the hub layer S;

[0046] (4) Calculate the conditional entropy H(C|S) of the substation attribute split;

[0047] (5) Calculate the information gain g(C,S) of the hub layer to the carbon source layer;

[0048] (6) Calculate the intermediate information entropy H of the carbon source layer to the hub layer S (C);

[0049] (7) Calculate the information gain ratio g from the carbon source layer to each hub layer node R (C, S), select the smallest one as the preferred direction of carbon flow from the carbon source layer to the hub layer.

[0050] Further, (8) repeat the above steps (1) to (7) to calculate the information gain ratio g from the hub layer to each branch layer respectively. R (S, F), select the smallest one as the preferred direction of carbon flow from the hub layer to the branch layer.

[0051] Furthermore, (9) repeat the above steps (1) to (8) to calculate the information gain ratio g from the branch layer to each terminal layer respectively. R (F, Z), select the first three electricity meters corresponding to the three smallest information gains as the final direction of the carbon flow from the branch layer to the terminal layer.

[0052] Specifically, in step (1), the step-down substation above 10 kV is the hub layer S; the 10 kV distribution transformer is the branch layer F; the collection equipment including the dedicated transformer collection terminal and concentrator and the electric energy meter is the terminal layer Z.

[0053] Specifically, in step (2), the information entropy H(C) of the carbon source layer C is calculated, and the sunny rate of the area is defined as p1, the cloudy rate is p2, the overcast rate is p3, and the rainy rate is p4.

[0054]

[0055] Specifically, in step (3), the information entropy H(S) of the hub layer S is calculated, and the probability that the average value of the load rate of a certain transformer in the hub layer S is greater than 100% is defined as q1, the probability that the average value of the load rate of the transformer is greater than 75% and less than or equal to 100% is defined as q2, and the probability that the average value of the load rate of the transformer is less than or equal to 75% is defined as q3.

[0056]

[0057] Specifically, in step (4), the conditional entropy H(C|S) of the substation attribute split is calculated, and the information entropy of n substations is averaged.

[0058]

[0059] (4) Specifically, in step (5), the information gain g(C,S) of the hub layer to the carbon source layer is calculated

[0060] g(C,S) = H(C)- H(C|S) (4).

[0061] (4) Specifically, in step (6), the intermediate information entropy H of the carbon source layer to the hub layer is calculated S (C)

[0062]

[0063] Specifically, in step (7), the information gain ratio g from the carbon source layer to each hub layer node is calculated respectively. R (C, S), select the smallest one as the preferred direction of carbon flow.

[0064]

[0065] Wherein GF is the temperature correction coefficient, which is 1 when the temperature is 25 degrees Celsius, 1.1 when the temperature is greater than 25 degrees Celsius and less than or equal to 32 degrees Celsius, 1.2 when the temperature is greater than 32 degrees Celsius, 0.9 when the temperature is greater than 5 degrees Celsius and less than or equal to 35 degrees Celsius, and 0.8 when the temperature is less than 5 degrees Celsius.

[0066] An electronic device includes one or more processors for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors execute the aforementioned carbon metering method for optimizing the allocation of carbon flows in a power system based on the decision tree C4.5.

[0067] A storage medium having a computer program stored therein, wherein the computer program is configured to execute the aforementioned carbon metering method for optimizing carbon flow allocation in a power system based on the decision tree C4.5 when running.

[0068] Example

[0069] 1. The carbon metering method for optimizing the carbon flow allocation of the power system based on the decision tree C4.5 is combined with the carbon source layer C, the hub layer S, the branch layer F, and the terminal layer Z to determine the optimal path of the carbon flow based on the optimal allocation of the carbon flow of the power system based on the decision tree C4.5, specifically:

[0070] (1) Define power plant 1 in area A as the carbon source layer C, step-down substation 1 and step-down substation 2 above 10 kV as the hub layer S, 10 kV distribution transformer 1, distribution transformer 2, distribution transformer 3, and distribution transformer 4 as the branch layer F, and data collection equipment (including dedicated data collection terminals and concentrators) and electric energy meters as the terminal layer Z.

[0071] (2) The information entropy H(C) of the carbon source layer C is calculated to be 1.6.

[0072] (3) The information entropy H(S) of the hub layer S is calculated to be 1.2.

[0073] (4) The conditional entropy H(C|S) of the substation attribute splitting is calculated to be 1.1.

[0074] (5) The information gain g(C,S) of the hub layer to the carbon source layer is calculated to be 0.5 and 0.7 respectively.

[0075] (6) Calculate the intermediate information entropy H of the carbon source layer to the hub layer S (C) are 0.5 and 0.4 respectively.

[0076] (7) Calculate the information gain ratio g from the carbon source layer to each hub layer node R (C, S) are 1.1 and 1.925 respectively, and the information gain ratio g of step-down substation 1 R (C, S) is the smallest, and step-down substation 1 is selected as the preferred direction of carbon flow.

[0077] (8) Repeat the above steps to calculate the information gain ratio g from the hub layer to each branch layer respectively. R (S, F), the ratio from step-down substation 1 to 10kV distribution transformer 1 is 1.2, which is the smallest information gain ratio from the hub layer to each branch layer. Step-down substation 1 to distribution transformer 1 is the preferred direction of carbon flow.

[0078] (5) Repeat the above steps to calculate the information gain ratio g from the branch layer to each terminal layer respectively. R (F, Z), select distribution transformer 1 to user 1, user 3, and user 4 as the three smallest information gains corresponding to the first three terminals, and power plant 1-step-down substation 1-distribution transformer 1-user 1, user 3, and user 4 as the final flow direction of the carbon flow.

Claims

1. A carbon measurement method for optimizing carbon flow allocation in power systems based on decision tree C4.5, characterized by: Combine the carbon source layer C, hub layer S, branch layer F and terminal layer Z, and then determine the optimal path of carbon flow based on the optimal allocation of carbon flow in the power system of decision tree C4.5; The specific steps are as follows: (1) Define the power plants in the specified area as the carbon source layer C, hub layer S, branch layer F and terminal layer Z; (2) Calculate the information entropy H(C) of the carbon source layer C; (3) Calculate the information entropy H(S) of the hub layer S; (4) Calculate the conditional entropy H(C|S) of the substation attribute split; (5) Calculate the information gain g(C,S) of the hub layer to the carbon source layer; (6) Calculate the intermediate information entropy H of the carbon source layer to the hub layer S (C); (7) Calculate the information gain ratio g from the carbon source layer to each hub layer node R (C, S), select the smallest one as the preferred direction of carbon flow from the carbon source layer to the hub layer.

2. According to claim 1, a carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 is characterized by: (8) Repeat the above steps (1) to (7) to calculate the information gain ratio g from the hub layer to each branch layer respectively. R (S, F), select the smallest one as the preferred direction of carbon flow from the hub layer to the branch layer.

3. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 2 is characterized by: (9) Repeat the above steps (1) to (8) to calculate the information gain ratio g from the branch layer to each terminal layer respectively. R (F, Z), select the first three electricity meters corresponding to the three smallest information gains as the final direction of the carbon flow from the branch layer to the terminal layer.

4. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (1), the step-down substation above 10 kV is the hub layer S; the 10 kV distribution transformer is the branch layer F; the collection equipment including the dedicated transformer collection terminal and concentrator and the electric energy meter is the terminal layer Z.

5. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (2), the information entropy H(C) of the carbon source layer C is calculated, and the sunny rate of the area is defined as p1, the cloudy rate is p2, the overcast rate is p3, and the rainy rate is p4; 6. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (3), the information entropy H(S) of the hub layer S is calculated, and the probability that the average value of the load rate of a certain transformer in the hub layer S is greater than 100% is defined as q1, the probability that the average value of the load rate of the transformer is greater than 75% and less than or equal to 100% is defined as q2, and the probability that the average value of the load rate of the transformer is less than or equal to 75% is defined as q3; 7. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (4), the conditional entropy H(C|S) of the substation attribute split is calculated, and the information entropy of n substations is averaged; 8. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (5), calculate the information gain g(C,S) of the hub layer to the carbon source layer g(C,S) = H(C)- H(C|S) (4).

9. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (6), calculate the intermediate information entropy H of the carbon source layer to the hub layer S (C) 10. The carbon metering method for optimizing carbon flow allocation in a power system based on decision tree C4.5 according to claim 1 is characterized by: In step (7), the information gain ratio g from the carbon source layer to each hub layer node is calculated respectively. R (C, S), select the smallest one as the preferred direction of carbon flow; Wherein GF is the temperature correction coefficient, which is 1 when the temperature is 25 degrees Celsius, 1.1 when the temperature is greater than 25 degrees Celsius and less than or equal to 32 degrees Celsius, 1.2 when the temperature is greater than 32 degrees Celsius, 0.9 when the temperature is greater than 5 degrees Celsius and less than or equal to 35 degrees Celsius, and 0.8 when the temperature is less than 5 degrees Celsius.

Images