Power grid ice melting strategy optimization method based on range icing

Abstract

The invention discloses a power grid ice melting strategy optimization method based on range icing, and particularly relates to the field of power grid ice melting, which comprises the following steps: identifying strategy adjustment behavior differences formed between regions based on actual ice melting equipment states, extracting key boundary paths with significant adjustment direction offset, and constructing an initial identification structure for cross-region consistency control; constructing a fusion type feedback path chain to cope with a boundary strategy direction confrontation problem, and embedding the fusion type feedback path chain into a dynamic weight function to realize structure self-adaptive updating of a boundary weight adjustment mechanism; the adjustment effect after the boundary strategy callback is evaluated, unbalanced nodes are judged and identified through a combined threshold value, and a structure rollback mechanism is triggered to maintain the stability and convergence of boundary scheduling; weight adaptive evolution is realized by constructing a region adjustment state vector, a boundary offset recognition mechanism and a dynamic weight adjustment function and fusing historical behavior evolution, so that the problem of lack of global modeling and boundary cooperative control in existing scheduling is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid de-icing, and more particularly to a power grid de-icing strategy optimization method based on range icing. Background Art

[0002] Existing technologies for grid icing response often employ local response mechanisms based on node-level ice thickness thresholds. Strategy scheduling relies on fixed rules or manual experience, lacking systematic modeling of the overall grid topology and the impact range of power outages. This makes it difficult to support optimal allocation of ice-melting resources globally. Even some research has begun to incorporate ice thickness monitoring and equipment control models, but these efforts primarily focus on improving the efficiency of individual lines, and have yet to establish a scheduling system focused on minimizing total outage losses. As this solution is implemented and put into operation in multi-regional power grids, regional dispatch strategies based on the dynamic weight adjustment mechanism will continue to evolve adaptively. Based on the independent optimization of multiple regions, strategy biases and boundary inconsistencies may occur, resulting in a decrease in cross-regional coordination efficiency and affecting the overall dispatch stability. Therefore, the problem that needs to be urgently solved after the actual deployment of this solution is: how to maintain the coordination and boundary consistency of ice-melting dispatch behavior across the entire network based on the long-term evolution of the multi-region dynamic weight adjustment strategy. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for optimizing the power grid de-icing strategy based on range icing. By constructing a regional regulation state vector, a boundary offset identification mechanism and a dynamic weight adjustment function, the adaptive evolution of weights is realized by integrating historical behavior evolution, so as to solve the problem of lack of global modeling and boundary collaborative control in existing scheduling.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing a power grid de-icing strategy based on range icing, comprising: S1. Identify differences in policy adjustment behaviors between regions based on the actual status of ice melting equipment, extract key boundary paths with significant adjustment direction deviations, and construct an initial identification structure for cross-region consistency control; S2. Construct a fusion feedback path chain to deal with the boundary strategy direction confrontation problem and embed it into the dynamic weight function to achieve structural adaptive update of the boundary weight adjustment mechanism; S3. Evaluate the adjustment effect after the boundary policy callback, identify the imbalanced nodes through joint threshold judgment, and trigger the structural fallback mechanism to maintain the stability and convergence of the boundary scheduling; S4. Extract weight paths with periodic stable characteristics in the regulation evolution history, generate differentiated initial structures through path type determination, and realize adaptive guidance of cross-cycle weight regulation.

[0005] In a preferred embodiment, S1 further includes obtaining a set of policy adjustment trajectories for each region through the adjustment behavior data returned by the range ice melting equipment cluster, the adjustment behavior data including the weight adjustment instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response result associated with the ice thickness status of the node, performing structured mapping on the policy adjustment trajectory set, and outputting a regional adjustment state vector group; The interaction similarity evaluation between the regional regulation state vector group and the grid boundary node structure is performed. The regulation behavior data includes the weighted regulation instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response results associated with the node ice thickness status. The boundary interaction response vector set is extracted, and the directional difference analysis is performed on the boundary interaction response vector set to output the boundary strategy difference matrix.

[0006] In a preferred embodiment, S1 further includes determining whether there is a response path in the boundary strategy difference matrix in which the cumulative difference in the adjustment direction of the nodes in the path is greater than a preset strategy deviation threshold. If so, the key nodes in the path are extracted and output as a strategy inconsistent boundary node set; otherwise, the output is a consistency confirmation identifier set. Perform path connectivity expansion on the strategy-inconsistent boundary node set to generate a boundary strategy path set to be corrected;

[0007] in, For the current cycle The set of policy-adjusted response path values ​​for each boundary node in the ; For the Nodes at time The state trajectory vector of For the The structural coupling relationship matrix between a node and its adjacent boundary nodes in the topological structure; For the Nodes at time The strategic direction changes the tension vector; To adjust the start time of the history window; is the set of boundary nodes; Current time Neidi The cumulative adjustment trend response value of each node; is the set of all paths to be analyzed in the area; When the cumulative response value of the adjustment trend exceeds the preset threshold The set of exception strategy paths; The decision threshold for the path cumulative offset response.

[0008] In a preferred embodiment, S2 further includes obtaining an adjustment sensitivity parameter group of each node in the path through the boundary strategy path set to be corrected, performing boundary direction tension modeling on the adjustment sensitivity parameter group, and outputting a boundary control tendency tensor group; Determine whether there is a tension substructure with mutually exclusive directions in the boundary control tendency tensor group. If so, extract the tension substructure and output it as a strategy confrontation chain group. Otherwise, output it as a consistent control chain group.

[0009] In a preferred embodiment, S2 further includes performing a direction decoupling operation on the strategy adversarial chain group and generating a weight callback sequence, which is then merged with the consistent control chain group to output a fused boundary feedback path chain; The fused boundary feedback path chain is embedded into the original dynamic weight adjustment function as a structural term, the structure is updated, and a set of boundary weight adjustment functions is output.

[0010]

[0011] in, For the The fusion adjustment feedback value of each node; For the The set of adjacent nodes of a node; For nodes The adjustment structure response gain coefficient; For nodes and The tension between the spatial gradient tensor; For the Response inertia parameters of each node; For nodes At the moment The regulatory sensitivity function of For nodes and The boundary structure conduction factor between them; is the integral path element; Feedback weight function set for fusion strategy; is the multi-path tension response merging function.

[0012] In a preferred embodiment, S3 further includes applying a boundary weight adjustment function set to a policy-inconsistent boundary node set, executing a weight policy callback operation, and outputting a node state data set after the callback; The difference calculation is performed on the node status data set after the callback and the status data of the same node at the corresponding time in the previous cycle, and the minimization index is output as the weighted callback effect index set.

[0013] In a preferred embodiment, S3 further includes determining whether there is a node in the weighted callback effect indicator set whose power outage loss increment is greater than a preset loss threshold, and whose state change rate exceeds a stability fluctuation threshold at the same time. If the joint condition is met, the output is a policy imbalance node set; otherwise, the output is a regulation stability confirmation mark. Execute control freeze and path rollback operations on the strategy imbalance node set, and output the structure rollback path node group;

[0014]

[0015] in, For the Nodes in the time interval The policy perturbation response value within ; For nodes In time The state execution trajectory vector of For the The second-order derivative of the node state trajectory with respect to time; is the node’s immediate benefit response factor; is the systemic imbalance threshold for responding to policy disturbances; is the set of nodes identified as having policy perturbation imbalance.

[0016] In a preferred embodiment, S4 further includes writing the adjustment stability confirmation mark and the structure fallback path node group into the behavior history body, the behavior history body includes the node response data of each cycle, the adjustment behavior trajectory and the weight configuration record, and outputting the strategy fusion evolution history body; The strategy is integrated with the evolution history execution behavior clustering analysis and path response periodicity extraction to generate a sustainable control path sequence set.

[0017] In a preferred embodiment, S4 further includes determining whether there is a path group in the sustainable control path sequence set that satisfies the similarity greater than a preset similarity threshold in terms of the weight evolution direction, callback amplitude, and node response mode of the adjustment trajectory in multiple consecutive cycles. If so, a path memory guidance structure is constructed; otherwise, an initialization reconstruction guidance structure is constructed. If it is determined that the construction is a path memory guidance structure, it will be used as the initial fitting term of the historical fusion weight and input into the dynamic weight function of the next cycle; if it is an initialization reconstruction guidance structure, it will be used to construct the basic response structure and perform new parameter initialization, and finally the unified output will be the starting weight configuration group of the new cycle;

[0018] in, For the The historical path compression vector of each node; For nodes In time The state execution trajectory vector of is the time decay coefficient of the historical trajectory; is the set of trajectories clustered as stable paths in historical behaviors; Behavior fitting function for path memory nodes; Reconstruct the state structure of the memoryless node; For the The expected trajectory of the target behavior of each node; For the The initial weight value of each node in the next cycle.

[0019] Technical effects and advantages of the present invention: To address the issue of inconsistent boundary strategies in the evolution of multi-region dynamic weight regulation, this solution constructs a regional regulation state vector group and extracts the regulation direction offset path. By aggregating path nodes, it generates a set of inconsistent boundary nodes and expands them into the path to be corrected. This achieves cross-region boundary consistency identification and structural closed-loop control to support global scheduling coordination based on the goal of minimizing total power outage losses. Based on the extraction of path regulation sensitivity parameters, boundary direction tension modeling is performed and a control tendency tensor is constructed. By identifying mutually exclusive tension substructures, a fused feedback path chain is generated and embedded with a dynamic weight function, thus constructing a path-level structural regulation subsystem to coordinate the coupling direction of ice thickness, grid connectivity, and equipment execution capability at the path layer. Based on the time-varying rate of node callback states and the incremental loss due to synchronized outages, a two-dimensional indicator determination mechanism is constructed to identify policy-imbalanced nodes and execute path freezing and rollback operations. This forms a real-time policy fault-tolerance mechanism to limit the interference of highly volatile nodes on the system loss control path, thereby improving the feedback constraint capability and local stability during the regulation process. Establish a regulation evolution history body and perform path response clustering, identify path sequences with periodic similar characteristics, build a path memory guidance structure or initialize the reconstruction structure, and generate a new cycle starting weight configuration according to the difference in structure type, realize the inheritance and differentiated control of policy parameters between cycles, and form an adaptive update logic for cross-cycle dynamic regulation trajectories. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The figure is a flow chart of the method steps of the present invention.

[0021] Figure 2 This is a critical path identification flow chart of the present invention.

[0022] Figure 3 A flow chart is constructed for the feedback path fusion of the present invention.

[0023] Figure 4 This is a flow chart of the strategy callback and imbalance judgment of the present invention.

[0024] Figure 5 This is a flow chart of the memory boot and initialization configuration of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] Refer to the instruction manual Figure 1-5 According to an embodiment of the present invention, a method for optimizing a power grid de-icing strategy based on range icing includes: S1. Identify differences in policy adjustment behaviors between regions based on the actual status of ice melting equipment, extract key boundary paths with significant adjustment direction deviations, and construct an initial identification structure for cross-region consistency control; S2. Construct a fusion feedback path chain to deal with the boundary strategy direction confrontation problem and embed it into the dynamic weight function to achieve structural adaptive update of the boundary weight adjustment mechanism; S3. Evaluate the adjustment effect after the boundary policy callback, identify the imbalanced nodes through joint threshold judgment, and trigger the structural fallback mechanism to maintain the stability and convergence of the boundary scheduling; S4. Extract weight paths with periodic stable characteristics in the regulation evolution history, generate differentiated initial structures through path type determination, and realize adaptive guidance of cross-cycle weight regulation.

[0027] S1 also includes obtaining a set of policy adjustment trajectories for each region through the adjustment behavior data transmitted back by the range ice melting equipment cluster. The adjustment behavior data includes the weight adjustment instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response results associated with the node ice thickness status. The policy adjustment trajectory set is structured mapped and a regional adjustment state vector group is output. The adjustment behavior data is obtained by obtaining the execution status of the range ice melting equipment cluster through the real-time monitoring module, and combining the node ice thickness measurement data with the weight adjustment function calculation result. The real-time monitoring module is composed of a current sensor, a temperature sensor, an ice thickness measurement device, and a remote communication unit, and is used to collect equipment operating status and environmental parameters. The interaction similarity evaluation between the regional regulation state vector group and the grid boundary node structure is performed. The regulation behavior data includes the weighted regulation instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response results associated with the node ice thickness status. The boundary interaction response vector set is extracted, and the directional difference analysis is performed on the boundary interaction response vector set to output the boundary strategy difference matrix.

[0028] S1 also includes determining whether there is a response path in the boundary strategy difference matrix in which the cumulative difference in the adjustment direction of the nodes in the path is greater than a preset strategy deviation threshold. If so, extract the key nodes in the path and output them as a strategy inconsistent boundary node set; otherwise, output them as a consistency confirmation identifier set. Perform path connectivity expansion on the strategy-inconsistent boundary node set to generate a boundary strategy path set to be corrected;

[0029] in, For the current cycle The set of policy-adjusted response path values ​​for each boundary node in the ; For the Nodes at time The state trajectory vector includes the node ice thickness state, weight instruction, current response and energy consumption fluctuation index; For the The structural coupling relationship matrix of a node and its adjacent boundary nodes in the topological structure is used to reflect the node's regulation and influence on the propagation structure; For the Nodes at time The strategy direction change tension vector is used to measure the deviation trend of the node's current adjustment path; To adjust the start time of the history window; is the boundary node set, which includes all nodes participating in regulation on the boundaries of the power grid area; Current time Neidi The cumulative adjustment trend response value of each node; is the set of all paths to be analyzed in the area; When the cumulative response value of the adjustment trend exceeds the preset threshold The set of exception strategy paths; is the judgment threshold of the path cumulative offset response, which is used to identify abnormal policy behavior; Further, in In the formula, the adjustment speed change of each node in the path is used , and the structural coupling matrix and directional offset tension Convolution integral to obtain the overall path response value ; Then through the path response and the sum and the judgment threshold Compare and identify the critical paths of structural direction deviation.

[0030] S2 also includes obtaining a regulation sensitivity parameter group of each node in the path through the set of boundary strategy paths to be corrected, performing boundary direction tension modeling on the regulation sensitivity parameter group, and outputting a boundary control tendency tensor group, wherein the boundary direction tension modeling refers to calculating the mutual traction relationship of each node on the boundary strategy path in a multi-regional power grid according to the response intensity and directionality of each node in the regulation sensitivity parameter group to the weight change, thereby forming a vector field structure that characterizes the strategy coordination or confrontation trend between nodes, and is used to identify the inherent mechanical relationship of the boundary strategy consistency offset; Determine whether there is a tension substructure with mutually exclusive directions in the boundary control tendency tensor group. If so, extract the tension substructure and output it as a strategy confrontation chain group. Otherwise, output it as a consistent control chain group.

[0031] S2 also includes performing a direction decoupling operation on the strategy adversarial chain group and generating a weight callback sequence, which is then merged with the consistent control chain group to output a fused boundary feedback path chain; The fused boundary feedback path chain is embedded as a structural term in the original dynamic weight adjustment function. The original dynamic weight adjustment function is a function model that calculates the ice-melting priority weight of each node in the current period based on multi-dimensional input parameters such as ice thickness, grid topology, and ice-melting equipment status. The weight is dynamically adjusted over time to minimize the total power outage loss. The structure is updated and a set of boundary weight adjustment functions is output.

[0032]

[0033] in, For the The fusion adjustment feedback value of each node is used to reflect the tension response synthesis of the node in the path structure; For the The set of adjacent nodes of a node; For nodes The adjustment structure response gain coefficient is used to measure the response amplitude of the node to the feedback adjustment; For nodes and The tension space gradient tensor between them is used to measure the tension offset of the regulatory path in space; For the The response inertia parameters of each node are used to adjust the hysteresis control during direction feedback; For nodes At the moment The regulatory sensitivity function of For nodes and The boundary structure conductivity factor between them is used to reflect the information conduction capability of the path topology structure; is the integral path element, which is used to represent the integral interval along the boundary path; is a fusion strategy feedback weight function set, which is used to represent the final multi-node strategy feedback synthesis weight; is a multi-path tension response merging function, which is used to combine multiple The adjustment feedback values ​​are structurally unified and integrated to form an overall adjustment weight function at the path level.

[0034] Further, in In the formula, the tension gradient difference between nodes is integrated through the local path line integral Response block , construct feedback adjustment intensity function ; Through the path structure function All Aggregate to form a fusion feedback weight function .

[0035] S3 also includes applying the boundary weight adjustment function set to the policy inconsistent boundary node set, executing the weight policy callback operation, and outputting the node status data set after the callback; The difference calculation is performed on the node status data set after the callback and the status data of the same node at the corresponding time in the previous cycle, and the minimization index is output as the weighted callback effect index set.

[0036] S3 also includes determining whether there is a node in the weighted callback effect indicator set whose power outage loss increment is greater than the preset loss threshold, and the state change rate of the node also exceeds the stability fluctuation threshold. If the joint conditions are met, the output is a strategy imbalance node set; otherwise, the output is a regulation stability confirmation mark; Execute control freeze and path rollback operations on the strategy imbalance node set, and output the structure rollback path node group;

[0037]

[0038] in, For the Nodes in the time interval The policy perturbation response value within ; For nodes In time The state execution trajectory vector of For the The second-order derivative of the node state trajectory with respect to time. The node state trajectory is used to characterize the oscillation intensity and nonlinear acceleration trend of the state change during the regulation process; is the node's immediate benefit response factor, which is used to measure the change in the node's regulatory economic effect at that point in time; is the systemic imbalance threshold for responding to policy disturbances; is the set of nodes identified as having policy perturbation imbalance.

[0039] Further, in In the formula, the second-order derivative of the adjustment trajectory and the adjustment economic benefit response function are The product integral between them is used to construct the disturbance loss intensity index , used to determine whether a fallback policy path is needed.

[0040] S4 also includes writing the adjustment stability confirmation mark and the structure fallback path node group into the behavior history body, which includes the node response data of each cycle, the adjustment behavior trajectory and the weight configuration record, and outputs the strategy fusion evolution history body; The strategy is integrated with the evolution history body execution behavior clustering analysis and path response periodicity extraction to generate a sustainable control path sequence set.

[0041] S4 also includes determining whether there is a path group in the sustainable control path sequence set that satisfies the similarity of the weight evolution direction, callback amplitude and node response mode in multiple consecutive cycles, and the similarity is greater than a preset similarity threshold. If so, a path memory guidance structure is constructed; otherwise, an initialization reconstruction guidance structure is constructed. If it is determined that the construction is a path memory guidance structure, it will be used as the initial fitting term of the historical fusion weight and input into the dynamic weight function of the next cycle, where the historical fusion weight refers to the weight trajectory extracted from multiple strategy evolution cycles that shows high stability and minimum loss response in the process of maintaining boundary consistency. It is the initial reference structure of the dynamic weight function formed by aggregate fitting; if it is an initialization and reconstruction guidance structure, the basic response structure is constructed with it and a new parameter initialization is performed, and the final unified output is the starting weight configuration group of the new cycle;

[0042] in, For the The historical path compression vector of each node is used to represent the exponentially weighted fusion result of the historical trajectory; For nodes In time The state execution trajectory vector of The time decay coefficient of the historical trajectory is used to control the attenuation degree of the influence of earlier behaviors on the current strategy initialization; is the set of trajectories clustered as stable paths in historical behaviors; Behavior fitting function for path memory nodes; Reconstruct the state structure of the memoryless node; For the The target behavior expected trajectory of each node is used as the reference state input when initializing the strategy; For the The initial weight value of each node in the next cycle.

[0043] Further, in In the formula, through exponential decay history fusion Construct a path memory vector, and decide whether to call the path fitting function or reconstruction function based on whether it belongs to the stable trajectory set, to form the initial value input of the next round of weights.

[0044] It should be noted that in the formula structure involved in this solution, dimensionless terms can serve as proportionality or structural adjustment factors. When combined with quantities with units, they only play a numerical scaling role and do not introduce new physical dimensions. Therefore, they will not change or confuse the overall unit system of expression. This combination of "dimensionless terms and units" can be understood as a composite structural expression commonly used in mathematical and physical modeling, conforming to the principle of dimensional consistency and having a clear physical interpretation basis. Secondly, in the formula structure of this scheme, if multiple variables with different physical units are involved, including but not limited to time, mass or energy variables, their joint appearance is to express the collaborative modeling relationship of multiple physical mechanisms. Each variable can be formed into a unified structure through function mapping, ratio combination or normalization adjustment. The units and meanings are clear, and the overall expression conforms to the principle of dimensional consistency and the common formula of engineering modeling. In this solution, any design constants, weights, adjustment factors, threshold parameters, and proportional coefficients are adjustable control parameters for different application environments. Their values ​​depend on the target device configuration, data input characteristics, and performance optimization goals. During the implementation phase, they are set within a reasonable range through model verification, performance constraints, or engineering calibration. Although these parameters do not have unique preset values, they have clear adjustment logic and calculation paths, and are part of the deterministic setting process in engineering implementation. The purpose of such setting is to ensure that the solution is both universally adaptable, reproducible, and operable, without affecting its technical clarity and feasibility. The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing power grid ice melting strategy based on range icing, characterized in that: include: S1. Identify differences in policy adjustment behaviors between regions based on the actual status of ice melting equipment, extract key boundary paths with significant adjustment direction deviations, and construct an initial identification structure for cross-region consistency control; S2. Construct a fusion feedback path chain to deal with the boundary strategy direction confrontation problem and embed it into the dynamic weight function to achieve structural adaptive update of the boundary weight adjustment mechanism; S3. Evaluate the adjustment effect after the boundary policy callback, identify the imbalanced nodes through joint threshold judgment, and trigger the structural fallback mechanism to maintain the stability and convergence of the boundary scheduling; S4. Extract weight paths with periodic stable characteristics in the regulation evolution history, generate differentiated initial structures through path type determination, and realize adaptive guidance of cross-cycle weight regulation.

2. The method for optimizing power grid ice melting strategy based on range icing according to claim 1, characterized in that: S1 also includes obtaining a set of policy adjustment trajectories for each region through the adjustment behavior data returned by the range ice melting equipment cluster. The adjustment behavior data includes the weight adjustment instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response results associated with the ice thickness status of the node. The policy adjustment trajectory set is structured mapped to output a regional adjustment state vector group. The interaction similarity evaluation between the regional regulation state vector group and the grid boundary node structure is performed. The regulation behavior data includes the weighted regulation instruction value of the range ice melting equipment cluster in each time period, the actual ice melting execution status, the equipment energy consumption feedback, and the response results associated with the node ice thickness status. The boundary interaction response vector set is extracted, and the directional difference analysis is performed on the boundary interaction response vector set to output the boundary strategy difference matrix.

3. The method for optimizing power grid ice melting strategy based on range icing according to claim 2, characterized in that: S1 also includes determining whether there is a response path in the boundary strategy difference matrix in which the cumulative difference in the adjustment direction of the nodes in the path is greater than a preset strategy deviation threshold. If so, extract the key nodes in the path and output them as a strategy inconsistent boundary node set; otherwise, output them as a consistency confirmation identifier set. Perform path connectivity expansion on the strategy-inconsistent boundary node set to generate a boundary strategy path set to be corrected; ； in, For the current cycle The set of policy-adjusted response path values ​​for each boundary node in the ; For the Nodes at time The state trajectory vector of For the The structural coupling relationship matrix between a node and its adjacent boundary nodes in the topological structure; For the Nodes at time The strategic direction changes the tension vector; To adjust the start time of the history window; is the set of boundary nodes; Current time Neidi The cumulative adjustment trend response value of each node; is the set of all paths to be analyzed in the area; When the cumulative response value of the adjustment trend exceeds the preset threshold The set of exception strategy paths; The decision threshold for the path cumulative offset response.

4. The method for optimizing power grid de-icing strategy based on range icing according to claim 3, characterized in that: S2 also includes obtaining an adjustment sensitivity parameter group of each node in the path through the boundary strategy path set to be corrected, performing boundary direction tension modeling on the adjustment sensitivity parameter group, and outputting a boundary control tendency tensor group; Determine whether there is a tension substructure with mutually exclusive directions in the boundary control tendency tensor group. If so, extract the tension substructure and output it as a strategy confrontation chain group. Otherwise, output it as a consistent control chain group.

5. The method for optimizing power grid ice melting strategy based on range icing according to claim 4, characterized in that: S2 also includes performing a direction decoupling operation on the strategy adversarial chain group and generating a weight callback sequence, which is then merged with the consistent control chain group to output a fused boundary feedback path chain; The fused boundary feedback path chain is embedded into the original dynamic weight adjustment function as a structural term, the structure is updated, and a set of boundary weight adjustment functions is output. ； ； in, For the The fusion adjustment feedback value of each node; For the The set of adjacent nodes of a node; For nodes The adjustment structure response gain coefficient; For nodes and The tension between the spatial gradient tensor; For the Response inertia parameters of each node; For nodes At the moment The regulatory sensitivity function of For nodes and The boundary structure conduction factor between them; is the integral path element; Feedback weight function set for fusion strategy; is the multi-path tension response merging function.

6. The method for optimizing power grid ice melting strategy based on range icing according to claim 5, characterized in that: S3 also includes applying the boundary weight adjustment function set to the policy inconsistent boundary node set, executing the weight policy callback operation, and outputting the node status data set after the callback; The difference calculation is performed on the node status data set after the callback and the status data of the same node at the corresponding time in the previous cycle, and the minimization index is output as the weighted callback effect index set.

7. The method for optimizing power grid de-icing strategy based on range icing according to claim 6, characterized in that: S3 also includes determining whether there is a node in the weighted callback effect indicator set whose power outage loss increment is greater than the preset loss threshold, and the state change rate of the node also exceeds the stability fluctuation threshold. If the joint conditions are met, the output is a strategy imbalance node set; otherwise, the output is a regulation stability confirmation mark; Execute control freeze and path rollback operations on the strategy imbalance node set, and output the structure rollback path node group; ； ； in, For the Nodes in the time interval The policy perturbation response value within ; For nodes In time The state execution trajectory vector of For the The second-order derivative of the node state trajectory with respect to time; is the node’s immediate benefit response factor; is the systemic imbalance threshold for responding to policy disturbances; is the set of nodes identified as having policy perturbation imbalance.

8. The method for optimizing power grid de-icing strategy based on range icing according to claim 7, characterized in that: S4 also includes writing the adjustment stability confirmation mark and the structure fallback path node group into the behavior history body, which includes the node response data of each cycle, the adjustment behavior trajectory and the weight configuration record, and outputs the strategy fusion evolution history body; The strategy is integrated with the evolution history execution behavior clustering analysis and path response periodicity extraction to generate a sustainable control path sequence set.

9. The method for optimizing power grid de-icing strategy based on range icing according to claim 8, characterized in that: S4 also includes determining whether there is a path group in the sustainable control path sequence set that satisfies the similarity of the weight evolution direction, callback amplitude and node response mode in multiple consecutive cycles, and the similarity is greater than a preset similarity threshold. If so, a path memory guidance structure is constructed; otherwise, an initialization reconstruction guidance structure is constructed. If it is determined that the construction is a path memory guidance structure, it will be used as the initial fitting term of the historical fusion weight and input into the dynamic weight function of the next cycle; if it is an initialization reconstruction guidance structure, it will be used to construct the basic response structure and perform new parameter initialization, and finally the unified output will be the starting weight configuration group of the new cycle; ； in, For the The historical path compression vector of each node; For nodes In time The state execution trajectory vector of is the time decay coefficient of the historical trajectory; is the set of trajectories clustered as stable paths in historical behaviors; Behavior fitting function for path memory nodes; Reconstruct the state structure of the memoryless node; For the The expected trajectory of the target behavior of each node; For the The initial weight value of each node in the next cycle.

Images