A power distribution network gridding loss reduction management method

Abstract

The application discloses a power distribution network gridding loss reduction management method, solves the technical problem of loss reduction optimization of the near and long-term power distribution network in the grid, and realizes the network frame optimization of the lines in the grid from the planning management level, so that the data scale is constrained from the source, the problem of "combination explosion" caused by the huge number of power distribution lines and the exhaustive search is avoided, the search ability and efficiency of the algorithm are improved, and the power distribution network structure, load type and equipment condition in the grid are different, the loss reduction scheme taking the grid as a management unit is proposed, and the differentiated loss reduction management is realized, and finally, in order to realize the coordination and unity with the grid planning, the method taking the long-term target network frame as the guide is constructed, and the optimization and reconstruction of the near and long-term power distribution network in the grid are realized.

Description

Technical Field

[0001] This invention belongs to the field of distribution network loss reduction planning technology, specifically relating to a grid-based loss reduction management method for distribution networks, and particularly a grid-based loss reduction method for distribution networks that can take into account both near-term and long-term load development. Background Technology

[0002] The main research content of power grid loss reduction planning is to rationally adopt various practical and effective energy-saving and loss-reduction measures to minimize line losses and reduce overall power consumption, while ensuring a robust power grid structure, coordinated power grid development, optimized structure, and sufficient power supply. Power grid loss reduction planning often involves analyzing the current operation of the distribution network and, based on the line loss of distribution equipment, focusing on optimizing and adjusting high-loss equipment. Power grid losses are mainly caused by factors such as outdated equipment, insufficient conductor cross-section, low utilization rate of energy-saving equipment, and unreasonable power supply radius.

[0003] In recent years, with the increased investment in distribution network construction, the reasons for high losses in distribution networks have been significantly improved. However, existing technologies, which address line loss management at the distribution line and single-equipment level in transformer substations, are insufficient to address issues such as load imbalance, inadequate reactive power compensation, and voltage quality problems caused by large-scale distributed network access, severely impacting power grid loss reduction and efficiency improvement. Furthermore, single-equipment line loss management often lacks consideration of the effectiveness of loss reduction, resulting in inaccurate assessments of the benefits. Summary of the Invention

[0004] To address the technical challenges of loss reduction optimization in distribution networks within grids in both the near and long term, this invention proposes a grid-based loss reduction management method for distribution networks. This method uses grids as units to perform statistical analysis on the magnitude, spatial location, and temporal variations of load demand from point to surface, improving the adaptability and feasibility of planning schemes, achieving refined investment targets, and ensuring effective integration with regional development plans. Grid-based loss reduction planning, based on grid-based planning, strengthens economic constraints on planning schemes, constructs an economical network structure that considers both near and long-term loads, and achieves efficient utilization of near-term distribution network equipment.

[0005] The technical solution adopted in this invention is a grid-based loss reduction management method for power distribution networks. This method is implemented by establishing a power distribution network model based on data information from power distribution network lines, performing calculations, and ultimately achieving the optimal solution for loss reduction and upgrading of power distribution lines. The specific steps of the loss reduction management method are as follows:

[0006] a. Based on grid data, equipment file data, topology data, and operational data, data fusion technology is used, along with supporting big data software tools, to construct a grid-based model of the distribution network and calculate the power flow of the grid and wiring.

[0007] b. Based on regional differences, set high-loss lines and high-loss grid thresholds. For high-loss equipment that exceeds the threshold for a long time, initiate a grid-based loss reduction simulation calculation for the distribution network that takes into account both near-term and long-term load development.

[0008] c. For the inner layer logic of the grid, the target grid structure is determined. Based on the basic data of the number of line cycles, grid structure, line length, line type and maximum operating current in the grid, the standardized wiring structure and the load-bearing capacity of different standardized wiring are set according to different power supply zones. The standard wiring structure in the grid is judged, and the target grid structure of the grid distribution line is constructed. With saturated load as a constraint, the long-term target grid structure is constructed through the TS optimization search algorithm.

[0009] d. For the outer layer logic of the grid, with the goal of minimizing the cost of the load optimization scheme, summarize the set of all measures to reduce line losses within the grid;

[0010] e. Using power flow, feeder current carrying capacity, and voltage drop as constraints, the renovation costs for implementing various loss reduction measures are determined.

[0011] f. Based on the two dimensions of load development and grid compatibility, a compatibility calculation model is constructed. The grid load development level is divided into three levels according to 30%, 60%, and 90% of the saturated load. The compatibility of the line topology structure between the near-term and long-term schemes is verified and output according to the load saturation situation, with the topology line compatibility selected at a ratio of 40%, 70%, and 100%, as the best scheme for loss reduction and transformation of the grid distribution lines.

[0012] In step d, the specific steps for summarizing the set of line loss reduction measures within all grids are as follows:

[0013] d1. Given that the near-term load nodes and power supply nodes are both determined, the objective function for minimizing the payback period and maximizing the cost-efficiency ratio is established as follows:

[0014] minF=C i -C0 Formula 1,

[0015] Where F is the final cost, C i Cost of grid-based loss reduction retrofit, and the loss reduction benefits of C0 grid retrofit;

[0016] d2, Cost of grid-based loss reduction retrofit C i The function is:

[0017]

[0018] In Equation 2, m represents the number of substations, n represents the number of load nodes, A represents the annual return on investment coefficient, and f(s) i ) represents the capacity s i Construction costs, u(s)i ) represents the transformer capacity s i The operation and maintenance costs, g(r) i Let α be the construction cost of the i-th load node, α be the number of line loops, and β1 be the construction cost per unit length of line. Let k be the length of the kth possible power supply path ending at load node i. Let i be the k-th possible path ending at load node i. Let k be the k-th possible power supply path ending at load node i. There are L possible power supply paths for load node i.

[0019] in,

[0020] In Equation 3, r0 is the discount rate and T is the depreciation period;

[0021] d3. The C0 function for the loss reduction benefit of grid modification is:

[0022]

[0023] In Equation 4: β2 is the electricity price, T2 is the electricity calculation time, B represents the total number of distribution network branches; R j Let P be the equivalent resistance of branch j. L,i Q represents the active power load demand of node i. L,i Let i represent the reactive load demand of node i, where i∈j, and let i represent the set of nodes i including node j and the nodes downstream of node j.

[0024] In step e,

[0025] e1. The power flow constraint function of the distribution network is: AP = D (Equation 5).

[0026] In Equation 5, A is the node association matrix; P is the network power flow; and D is the load demand.

[0027] e2, The feeder current carrying capacity constraint function is:

[0028] In Equation 6, S j,max This indicates the maximum current carrying capacity of branch j;

[0029] e3. Voltage drop constraint:

[0030] In a power grid, the voltage drop from a substation to the i-th (1≤i≤n) load point out of all n load points is ΔU. i Then the voltage drop from the substation to all n load points should be less than the maximum allowable voltage drop ΔU. max Its value can be determined according to the specific operating requirements of the power grid, generally the maximum allowable voltage drop ΔU maxThe value is 10% of the rated voltage, therefore there is

[0031] max{ΔU1,ΔU2,...,ΔU i ,...,ΔU n}≤ΔU max Equation 7,

[0032] The fitness function is as follows:

[0033] The objective functions for energy-saving retrofitting of power distribution lines are to minimize network losses and total investment costs, which is a minimum optimization problem. The adjusted fitness function value is as follows:

[0034]

[0035] In Equation 8, F(i) is the objective function value corresponding to the chromosome. min t represents the minimum objective function value in the current generation of the evolutionary population, and t is the temperature parameter.

[0036] In step a, the accompanying big data software tool is the Big Data Kettle tool.

[0037] In step b, the threshold for high-loss lines and high-loss grids is set to 10 days.

[0038] The beneficial effects of this invention are as follows: Based on the economic considerations of the distribution network, in the near term, the load is optimized and adjusted by considering the power flow, feeder current carrying capacity, and voltage drop of the distribution network; in the long term, a standard grid structure is aimed at and saturated load is used as a constraint to achieve the construction of the long-term target grid; through comparative analysis of the near-term load optimization scheme and the long-term target grid structure, and based on the calculation results of the edge grid line loss, a distribution network grid-based loss reduction planning scheme that takes into account the near-term and long-term load development is formed; through effective decision-making on grid-based loss reduction, redundant investment is avoided, and effective management of grid-based loss reduction is achieved. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method of the present invention.

[0040] Figure 2 It is a topology diagram for power flow calculation of cross-grid interconnection lines.

[0041] Figure 3 Based on Figure 2 The topology diagram after the switch is activated.

[0042] Figure 4 This is a topology diagram of the grid load before the modification.

[0043] Figure 5 This is a topology diagram of the grid load after the modification. Detailed Implementation

[0044] See Figure 1 The specific steps of the loss reduction management method of the present invention are described in detail.

[0045] This paper proposes a grid-based management model for distribution networks to achieve comprehensive optimization of line losses from single-equipment level to regional level. Based on the grid's future load development, the target grid structure is prioritized for solidification. Following this, simulation analysis of near-term loss reduction optimization for high-loss equipment within the grid is conducted to obtain the optimized grid structure. Based on the grid's development attributes, the compatibility with the target grid structure is compared, achieving efficient integration of loss reduction and planning.

[0046] Definitions:

[0047] TS Optimization Search Algorithm: Taking the existing grid structure as the optimization object, based on the target grid principle of the power supply zone to which the grid belongs, the algorithm calculates the theoretical line loss of the grid by adding new lines, forming the current optimal solution for the grid. By moving the added lines, a grid structure satisfying the grid principle is constructed, generating a set of solutions. Among the generated solutions, the solution that best satisfies the constraints is selected as the new current optimal solution. A storage structure is constructed to record the grid structure after the movement. When a new current optimal solution is generated, it is recorded in the storage structure as a taboo "object". After each movement, the generated calculation results are compared with the optimal solution of the taboo "object", and the best solution among the optimal solutions is retained.

[0048] The technical concept of this invention is:

[0049] (1) Based on information such as grid data, equipment file data, topology data, and operation data, a data fusion technology is used to construct a model based on the grid-based distribution network.

[0050] (2) Taking long-term high loss as the trigger condition, construct the grid distribution line target grid based on the target grid of different power supply zones, and realize the construction of the long-term target grid with saturated load as the constraint condition.

[0051] (3) Construct a model that converts the investment costs of substations, intermediate nodes and lines into an average annual investment cost model using the equal annual value method, and uses the power flow of the distribution network, feeder current carrying capacity and voltage drop as constraints to realize the near-term load optimization scheme.

[0052] (4) By dividing the grid into grids, the line loss calculation method is used. Considering the existence of cross-grid line connections, the power supply grid of the original equipment is changed by opening and closing the switch. Considering the actual situation, the theoretical line loss of the grid is calculated based on the connection topology relationship and then assigned to the original grid.

[0053] (5) By constructing a near-term load optimization scheme, a comparative analysis is conducted with the long-term target grid structure. The scheme with the highest degree of fit between the near-term scheme and the target grid structure is selected as the best scheme for reducing losses in the grid distribution lines.

[0054] In specific implementation of the present invention,

[0055] 1) Based on grid data, equipment file data, topology data, and operational data, data fusion technology is used, and the Kettle big data tool is employed to construct a grid-based model of the distribution network, enabling power flow calculation of the grid and wiring.

[0056] 2) Based on regional differences, set high-loss lines and high-loss grid thresholds. Long-term high loss is defined as lines and grids that exceed the high loss threshold for 10 consecutive days. For long-term high-loss equipment, initiate grid-based loss reduction simulation calculations for distribution networks that take into account both near-term and long-term load development.

[0057] 3) Within the grid, prioritize determining the target grid structure. Based on fundamental data such as the number of line cycles, grid structure, line length, line type, and maximum operating current within the grid, set standardized wiring structures and the load-bearing capacity of different standardized wirings according to different power supply zones. Conduct standard wiring structure assessments within the grid, construct the target grid structure for the grid distribution lines, and use saturated load as a constraint. Utilize the TS optimization search algorithm to achieve the construction of the long-term target grid structure.

[0058] 4) The outer layer logic of the grid determines the transformation plan by judging whether the selected line needs to be transformed. The purpose is to find the line transformation situation that minimizes line loss and investment while meeting certain reliability conditions.

[0059] 4.1) Function Determination

[0060] An economic model for optimizing the network load in the near term is constructed. With distribution network power flow, feeder current carrying capacity, and voltage drop as constraints, the topology relationship is solved under the premise that the near-term load nodes and power supply nodes are determined, so as to achieve the shortest payback period and the highest cost-efficiency ratio.

[0061] The objective function is as follows:

[0062] minF=C i -C0 (1)

[0063] Where F is the final cost, C i Cost of grid-based loss reduction and transformation, and the benefits of C0 grid-based loss reduction and transformation.

[0064] 4.2) Cost of grid-based loss reduction retrofit C i

[0065]

[0066] Where: m is the number of substations; n is the number of load nodes; A is the annual return on investment coefficient; f(s) i ) represents the capacity s iConstruction costs; u(s) i ) represents the transformer capacity s i Maintenance costs; g(r) i ) represents the construction cost of the i-th load node; α represents the number of line loops; and β1 represents the construction cost per unit length of line. Let k be the length of the kth possible power supply path ending at load node i. Let i be the kth possible path ending at load node i. Let k be the k-th possible power supply path ending at load node i; set There are L possible power supply paths for load node i.

[0067]

[0068] r0 is the discount rate, and T is the depreciation period.

[0069] 4.3) Grid transformation loss reduction benefits C0

[0070]

[0071] In the formula: β2 is the electricity price; T2 is the electricity calculation time; B represents the total number of distribution network branches; R j P is the equivalent resistance of branch j. L,i Q represents the active power load demand of node i; L,i Let i represent the reactive load demand of node i; i∈j means that the set of nodes i includes node j and the nodes downstream of node j.

[0072] 5) Using power flow, feeder current carrying capacity, and voltage drop as constraints, the renovation costs for implementing various loss reduction measures are determined.

[0073] The constraint functions are as follows:

[0074] 5.1) Current Constraints

[0075] AP = D (5)

[0076] In the formula: A is the node association matrix; P is the network power flow; D is the load demand.

[0077] 5. Power flow calculation method for cross-grid interconnection lines. Based on archival data, a topology structure is formed as follows: Figure 2 As shown, the root node is the power point, 10 is the tie switch, and the tie switch is the dividing line, with grid A on the left and grid B on the right.

[0078] When grid A is heavily overloaded, sectionalizing switch 8 and tie switch 10 operate, transferring the load of sections L8 and L9 from grid A to grid B. Based on the edge-based partitioning method, theoretical line loss calculations are performed on the newly formed grid according to its topology. The calculation results are then reassigned to the original grid, enabling a comparative analysis before and after grid optimization. (See [link to relevant documentation]). Figure 3 .

[0079] 5.2) Feeder current carrying capacity constraint

[0080]

[0081] In the formula S j,max This represents the maximum current carrying capacity of branch j.

[0082] 5.3) Voltage drop constraint

[0083] In a power grid, the voltage drop from a substation to the i-th (1≤i≤n) load point out of all n load points is ΔU. i Then the voltage drop from the substation to all n load points should be less than the maximum allowable voltage drop ΔU. max Its value can be determined according to the specific operating requirements of the power grid, generally the maximum allowable voltage drop ΔU max The value is 10% of the rated voltage. Therefore, there is

[0084] max{ΔU1,ΔU2,...,ΔU i ,...,ΔU n}≤ΔU max (7)

[0085] The fitness function is as follows:

[0086] The objective functions for energy-saving retrofitting of power distribution lines are to minimize network losses and total investment costs, which is a minimum optimization problem. The adjusted fitness function value is as follows:

[0087]

[0088] In the formula: F(i) is the objective function value corresponding to the chromosome; F min t represents the minimum objective function value in the current generation of the evolutionary population; t is the temperature parameter.

[0089] 5.4) Calculation steps

[0090] 5.4.1) Based on the long-term high loss of lines within the grid, initiate external grid calculations. Based on the current weak points within the grid, determine the reasons for the long-term high loss of lines, including uneven load distribution on the lines within the grid, a large proportion of heavily overloaded lines, and unreasonable connection of distributed power sources.

[0091] 5.4.2) Scenario 1: No cross-grid connections, no distributed power sources within the grid, resulting in uneven line load. Initiate grid-based line load rate determination to identify heavily overloaded (≥80%) and lightly overloaded (<20%) lines. Based on existing switches, perform segmented load analysis on heavily overloaded lines to ensure the load after resizing is between 40% and 60%, creating a list of users to be resized. Select lightly overloaded lines and connect them to the nearest users on the resizing list via newly created lines. Initiate line loss calculation for the optimized distribution lines within the grid.

[0092] 5.4.3) Scenario 2: No cross-grid connections, distributed power sources within the grid, and no line load imbalance issues. Initiate a check for 10kV line backfeed to the distributed power source. If present, mark the distributed power source. Locate the distribution area within a 300m radius of the distributed power source point, analyze the area's capacity and load, rank the areas' absorption capacity, conduct simulation analysis of the identified distributed power source connection, calculate the line loss of the area. If the line loss is within the economic range, perform line loss calculations for the grid-optimized distribution lines; if it is outside the economic range, repeat the simulation calculations for areas with low absorption capacity to ensure the line loss is within the economic range.

[0093] 5.4.4) Scenario 3: No cross-mesh communication, distributed power sources within the mesh, and unbalanced line load. The process begins by determining the line load rate within the grid, identifying lines that are heavily overloaded (≥80%) or lightly overloaded (<20%). Next, for lightly overloaded lines, a distributed power source backfeed problem is assessed. If present, the distributed power source is identified, and transformer substations are located within a 300m radius centered on the source. Substation capacity and load conditions are analyzed, and substations are ranked according to their absorption capacity. Simulation analysis of the substations connected to the identified distributed power sources is conducted, calculating the line loss. If the line loss falls within the economic range, the line loss calculation for the grid-optimized distribution lines is performed. If it falls outside the economic range, the simulation calculation for substations with low absorption capacity is repeated to ensure the line loss falls within the economic range. If no line loss occurs, the heavily overloaded line assessment is initiated. Based on existing switches, segmented load analysis is performed on the heavily overloaded lines to ensure the load after reshoring is between 40% and 60%, creating a list of users to be reshored. Lightly overloaded lines are selected, and new lines are created to connect the users on the reshoring list to the nearest available line. The line loss calculation for the grid-optimized distribution lines is then initiated.

[0094] 5.4.5) Scenario 4: High loss problem exists in cross-grid interconnection lines. Initiate cross-grid line segmented load analysis, analyze the effects and impacts of different switching operations by opening and closing switches, calculate the loss reduction result of cross-grid lines after each switching operation, record the sequence of each switching action, and form the optimal switching state.

[0095] 6) Compare the suitability of alternative loss reduction space frame schemes with the target space frame scheme to determine the final scheme.

[0096] A compatibility calculation model is constructed based on two dimensions: load development and grid compatibility. Grid load development levels are divided into three levels: 30%, 60%, and 90% of saturated load. The compatibility of the line topology between the near-term and long-term solutions is verified and output based on load saturation, with topology compatibility ratios of 40%, 70%, and 100%, as the optimal solution for loss reduction and upgrading of the grid distribution lines. Figure 4 , 5 As shown, the grid load before the modification was 92.4% of the saturation load. Through gridding loss reduction measures, the modified topology was formed, and the topologies highly overlapped.

[0097] 6.1) Calculate the proportion of the current load to the saturation load and classify the grid development level.

[0098] 6.2) Construct and display the current topology, saturated topology, and topology after recent modifications. Compare the topology structures, mainly based on indicators such as connection methods, changes in connection objects, and length of newly added lines. Set weights and obtain the fit results.

[0099] 6.3) Based on the topology line fit, combined with the fit gradient determined by the grid development level, output the grid evaluation results.

[0100] This invention constrains the network structure of distribution lines within a grid from a planning and management perspective, thereby optimizing the network structure. It constrains the data scale from the source, avoiding the "combinatorial explosion" problem caused by an exhaustive search due to the large number of distribution lines, thus improving the algorithm's search capability and efficiency. Secondly, considering that the same loss reduction scheme is not universally applicable across different medium-voltage circuits in distribution network grid loss reduction research, this invention proposes a loss reduction scheme based on the grid as the management unit, taking into account the differences in distribution network structure, load type, and equipment conditions within the grid, to achieve differentiated loss reduction management. Finally, to achieve coordination and unity with grid-based planning, a method guided by the long-term target network structure is constructed to optimize and transform the distribution network within the grid in both the near and long term.

Claims

1. A grid-based loss reduction management method for power distribution networks, wherein the loss reduction management method is based on establishing a power distribution network model from power distribution network line data information, performing calculations and processing, and thereby realizing the optimal scheme for loss reduction transformation of power distribution lines, characterized in that: The specific steps of the loss reduction management method are as follows: a. Based on grid data, equipment file data, topology data, and operational data, data fusion technology is used, along with supporting big data software tools, to construct a grid-based model of the distribution network and calculate the power flow of the grid and wiring. b. Based on regional differences, set high-loss lines and high-loss grid thresholds. For high-loss equipment that exceeds the threshold for a long time, initiate a grid-based loss reduction simulation calculation for the distribution network that takes into account both near-term and long-term load development. c. For the inner layer logic of the grid, the target grid structure is determined. Based on the basic data of the number of line cycles, grid structure, line length, line type and maximum operating current in the grid, the standardized wiring structure and the load-bearing capacity of different standardized wiring are set according to different power supply zones. The standard wiring structure in the grid is judged, and the target grid structure of the grid distribution line is constructed. With saturated load as a constraint, the long-term target grid structure is constructed through the TS optimization search algorithm. d. For the outer layer logic of the grid, with the goal of minimizing the cost of the load optimization scheme, summarize the set of all measures to reduce line losses within the grid; e. Using power flow, feeder current carrying capacity, and voltage drop as constraints, the renovation costs for implementing various loss reduction measures are determined. f. Based on two dimensions—load development and grid compatibility—a compatibility calculation model is constructed. Grid load development levels are divided into three grades: 30%, 60%, and 90% of saturated load. The compatibility of the line topology between the near-term and long-term solutions is verified and output at ratios of 40%, 70%, and 100% based on load saturation. This output serves as the optimal solution for loss reduction and upgrading of the grid distribution lines. In step d, the specific steps for summarizing the set of line loss reduction measures within all grids are as follows: d1. Given that the near-term load nodes and power supply nodes are both determined, the objective function for minimizing the payback period and maximizing the cost-efficiency ratio is established as follows: minF=C i -C0(Equation 1), Where F is the final cost, C i Cost of grid-based loss reduction retrofit, and the loss reduction benefits of C0 grid retrofit; d2, Cost of grid-based loss reduction retrofit C i The function is: In Equation 2, m represents the number of substations, n represents the number of load nodes, A represents the annual return on investment coefficient, and f(s) i ) represents the capacity s i Construction costs, u(s) i ) represents the transformer capacity s i The operation and maintenance costs, g(r) i Let α be the construction cost of the i-th load node, α be the number of line loops, and β1 be the construction cost per unit length of line. Let k be the length of the kth possible power supply path ending at load node i. Let i be the k-th possible path ending at load node i. Let k be the k-th possible power supply path ending at load node i. There are L possible power supply paths for load node i. in, In Equation 3, r0 is the discount rate and T is the depreciation period; d3. The C0 function for the loss reduction benefit of grid modification is: In Equation 4: β2 is the electricity price, T2 is the electricity calculation time, B represents the total number of distribution network branches; R j Let P be the equivalent resistance of branch j. L,i Q represents the active power load demand of node i. L,i Let i represent the reactive load demand of node i, where i∈j, and let i represent the set of nodes i including node j and the nodes downstream of node j.

2. The distribution network grid-based loss reduction management method according to claim 1, characterized in that: In step e, e1. The power flow constraint function of the distribution network is: AP = D (Equation 5). In Equation 5, A is the node association matrix; P is the network power flow; and D is the load demand. e2, The feeder current carrying capacity constraint function is: In Equation 6, S j,max This indicates the maximum current carrying capacity of branch j; e3. Voltage drop constraint: In a power grid, the voltage drop from a substation to the i-th (1≤i≤n) load point out of all n load points is ΔU. i Then the voltage drop from the substation to all n load points should be less than the maximum allowable voltage drop ΔU. max Its value can be determined according to the specific operating requirements of the power grid, the maximum allowable voltage drop ΔU max The value is 10% of the rated voltage, therefore there is max{ΔU1,ΔU2,...,ΔU i ,...,ΔU n}≤ΔU max (Equation 7), The fitness function is as follows: The objective functions for energy-saving retrofitting of power distribution lines are to minimize network losses and total investment costs, which is a minimum optimization problem. The adjusted fitness function value is as follows: In Equation 8, F(i) is the objective function value corresponding to the chromosome. min t represents the minimum objective function value in the current generation of the evolutionary population, and t is the temperature parameter.

3. The distribution network grid-based loss reduction management method according to claim 1, characterized in that: In step a, the accompanying big data software tool is the Big Data Kettle tool.

4. The distribution network grid-based loss reduction management method according to claim 1, characterized in that: In step b, the threshold for high-loss lines and high-loss grids is set to 10 days.

Images