A live working process optimization management method and system under complex terrain

Abstract

This invention discloses a method and system for optimizing and managing live-line work processes in complex terrain, belonging to the field of power system safety operation technology. The method includes complex terrain operation space modeling, real-time risk field construction, operation process generation, and dynamic optimization of the operation process. This invention employs a real-time risk field construction method based on multi-source risk information fusion, enabling the risk levels at different locations within the operation area to be intuitively represented in a continuous spatial distribution form and dynamically updated with environmental changes. It adopts a risk field-driven dynamic optimization method for the operation process, significantly reducing the degree of risk exposure during operation. By structurally modeling the operation space, uniformly assessing multi-source risks, and generating operation processes and dynamic adjustment mechanisms during implementation, it realizes a shift from experience-driven operation path planning and task arrangement to collaborative decision-making based on environment and risk.

Description

Technical Field

[0001] This invention belongs to the field of power system safety operation technology, specifically referring to a method and system for optimizing the management of live-line operation processes in complex terrain. Background Technology

[0002] As the scale of power systems continues to expand, transmission lines are gradually extending into areas with complex terrain and variable environments, such as mountains, valleys, and forests. When conducting live-line work in such environments, workers not only need to face the electric field risks brought by high voltage, but also need to cope with the combined effects of complex terrain conditions, weather changes, and environmental uncertainties, which significantly increases the safety and organizational difficulty of the operation.

[0003] In existing live-line work in complex terrain, environmental factors are highly variable, risk sources are scattered and difficult to assess uniformly. At the same time, existing live-line work procedures rely on pre-set plans and are difficult to adjust according to changes in on-site risks, which can easily lead to personnel entering high-risk areas or failing to take timely avoidance measures during the operation, thereby increasing the safety hazards of the operation. In addition, the selection of live-line work paths and the arrangement of procedures under existing complex terrain conditions mainly rely on manual experience and lack systematic collaborative optimization.

[0004] Therefore, there is an urgent need for a method and system for optimizing and managing live-line work processes, which can realize spatial modeling of complex terrain operations, unified assessment of multi-source risks, optimized generation of work processes, and dynamic adjustment of work processes, in order to improve work safety and execution efficiency, reduce risk exposure levels, and enhance the ability to respond to sudden risks. Summary of the Invention

[0005] To address the above issues and overcome the shortcomings of existing technologies, this invention provides a method and system for optimizing and managing live-line work processes in complex terrain. It creatively employs a real-time risk field construction method based on multi-source risk information fusion, achieving unified modeling and spatial representation of various risk factors such as wind speed changes, electric field strength, terrain stability, and personnel working status. This allows for a direct representation of the risk level at different locations within the work area in a continuous spatial distribution, dynamically updated with environmental changes. This provides a unified and reliable risk assessment basis for work path planning and task arrangement, improving the scientific rigor and consistency of work decisions. Furthermore, it creatively adopts a risk field-driven dynamic optimization method for the work process, enabling dynamic adjustments to personnel work paths, action arrangements, and work rhythm based on real-time risk changes during work execution. The adjustments enable the operation process to proactively avoid areas of rapidly escalating risk and promptly trigger control measures such as deceleration, suspension, or evacuation when risks exceed limits, thereby significantly reducing the degree of risk exposure during the operation process and improving the timeliness and effectiveness of response to sudden risks. By performing structured modeling of the work space and unified assessment of multi-source risks, and generating a dynamic adjustment mechanism for the work process and implementation based on this, the operation path planning and task arrangement have been transformed from experience-driven to collaborative decision-making based on environment and risk. This allows operators to continuously perceive changes in surrounding risks during the operation and adjust the work path and operation rhythm accordingly, thereby effectively reducing the time spent in high-risk areas, improving the rationality of the work path and the overall work efficiency, while enhancing the ability to respond to sudden risks and the safety and controllability of the operation process.

[0006] The technical solution adopted by this invention is as follows: This invention provides an optimized management method for live-line working processes in complex terrain, the method comprising the following steps:

[0007] Step S1: Spatial modeling for complex terrain operations;

[0008] Step S2: Real-time risk field construction;

[0009] Step S3: Job workflow generation;

[0010] Step S4: Dynamic optimization of the operation process.

[0011] Furthermore, in step S1, the complex terrain operation space modeling is used to establish a unified spatial representation basis required for subsequent risk field construction and operation process optimization. Specifically, it involves acquiring spatial data of the operation area, preprocessing and representing the spatial data with terrain, identifying and spatially calibrating the power facilities and environmental elements within the operation area, then discretizing the continuous space into three-dimensional spatial grid nodes, assigning corresponding spatial attributes and semantic labels to each grid node, and establishing adjacency relationships between grid nodes to form a structured three-dimensional operation space model containing spatial coordinates, terrain information, facility distribution information, and accessibility identifiers.

[0012] Further, in step S2, the real-time risk field construction is used to realize the dynamic perception and accurate assessment of multi-source risks in complex terrain live-line working environments. Specifically, based on a structured three-dimensional work space model, a real-time risk field construction method based on multi-source risk information fusion is adopted to obtain a risk field construction result containing the risk level of each grid node, risk field, risk evolution trend, and corresponding risk confidence interval, including the following steps:

[0013] Step S21: Multi-source state modeling, used to realize unified modeling of multi-source data in complex terrain operation environment, specifically by collecting environmental information data and personnel state information data at each grid node, performing unified feature mapping, constructing state vector, and obtaining unified state expression results for each grid node;

[0014] The state vector includes wind speed, wind direction, electric field strength, soil saturation, surface displacement rate, personnel heart rate, and continuous operation time.

[0015] Step S22: Single-source risk assessment, used to characterize the degree of independent impact of different risk factors on operational safety. Specifically, based on state vectors, corresponding risk mapping functions are constructed for different types of risk sources. The wind speed risk value, electric field risk value, landslide risk value and fatigue risk value of each grid node are calculated independently to achieve quantitative assessment of single-source risk and obtain the single-source risk value corresponding to each grid node.

[0016] The wind speed risk value is used to characterize the impact of wind speed and terrain disturbance on the stability of live-line work; the electric field risk value is used to characterize the impact of electric field strength in the work area on the safety of personnel and equipment; the landslide risk value is used to characterize the impact of ground stability on work safety; and the fatigue risk value is used to characterize the impact of continuous working conditions on safety.

[0017] Step S23: Data confidence modeling is used to characterize the assessment reliability of different types of risk sources at each grid node. Specifically, for each type of risk source, based on the data input it depends on in the risk calculation process, the uncertainty index of the corresponding data source is obtained, and exponential decay modeling is performed to calculate the data confidence of each grid node under each type of risk source, and the risk source-level confidence distribution results are obtained.

[0018] Step S24: Comprehensive risk assessment, used to achieve adaptive fusion assessment of multi-source risks. Specifically, based on obtaining the single-source risk value and corresponding data confidence of each grid node, a dynamic weighting mechanism is introduced to construct a multi-source risk fusion model, realize the nonlinear superposition and fusion of multi-source risks, obtain the comprehensive risk value of each grid node, and map the comprehensive risk value to the spatial grid structure to form a risk field that characterizes the spatial distribution of risks in the work area.

[0019] Step S25: Spatiotemporal propagation modeling of risk field, used to characterize the diffusion process of risk in space and time. Specifically, it involves constructing node adjacency relationships based on spatial grid, calculating the risk propagation weight between grid nodes by combining wind direction consistency, slope influence and spatial distance attenuation factors, establishing a risk propagation model based on neighborhood risk differences, updating the risk field in time series, and obtaining the dynamic evolution result of the risk field.

[0020] Step S26: Risk confidence assessment, used to quantify the credibility of the risk assessment results. Specifically, based on the dynamic weights, data confidence and uncertainty variance of each type of risk source, the uncertainty of each risk source is weighted and propagated to obtain the risk fluctuation characterization value of each grid node, and a risk confidence interval is constructed based on the risk fluctuation characterization value to obtain the risk credibility expression result including the upper and lower bounds.

[0021] Step S27: Risk classification output, specifically, based on the dynamic evolution results of the risk field, classifying and determining the comprehensive risk value of each grid node, generating the corresponding risk level, and calculating the risk evolution trend.

[0022] Further, in step S3, the work process generation specifically involves using a structured three-dimensional work space model, taking the risk field construction results, the set of work tasks, and the initial positions of personnel as inputs, and generating an initial work path in a three-dimensional space grid with risk constraints as boundary conditions through a path search algorithm; simultaneously, constructing a cost matrix between tasks based on path costs, and optimizing the task order using an approximate solution method for the traveling salesman problem; then, combining path length and task duration to perform time scheduling modeling, resulting in a structured work process scheme that includes path planning results, task execution order, and time arrangement.

[0023] Furthermore, in step S4, the dynamic optimization of the operation process is used to optimize and adjust the operation actions and paths based on real-time risk changes during live-line operations in complex terrain. Specifically, based on the structured three-dimensional operation space model, the risk field construction results and the structured operation process scheme are used as inputs. The risk field-driven dynamic optimization method of the operation process is adopted to obtain a real-time optimized operation execution scheme that includes the current optimal control strategy, dynamic safety constraint area, operation execution path and hierarchical response control strategy, so as to realize the collaborative optimization management of the live-line operation process.

[0024] The risk field-driven dynamic optimization method for work processes includes the following steps:

[0025] Step S41: Rolling time-domain predictive control modeling. Specifically, by obtaining the risk field construction results obtained in step S2 and the structured work process scheme generated in step S3, a system state variable including personnel spatial location, work posture state, task progress and tool state is constructed. A set of discrete control actions including movement, climbing, operation, pause and evacuation is defined, and a state transition model is established. Then, within the preset prediction time domain, an optimization function with risk cost, movement cost and task delay cost as objectives is constructed. The rolling time-domain predictive control is used to solve the work process to obtain the current optimal control strategy.

[0026] Step S42: Adaptive contraction of the safety envelope is used to achieve dynamic adjustment of the safety constraints of the work space. Specifically, a safety envelope region is constructed with the current work position as the center, and a direction-related safety distance model is established based on the comprehensive risk value, risk change rate and spatial risk gradient of the current work position. The safety radius of different directions is adaptively adjusted to form an asymmetric safety envelope constraint, so that the safety boundary of the high-risk direction contracts and the low-risk direction remains passable, thus obtaining a direction-aware dynamic safety constraint region.

[0027] Step S43: Path optimization and adjustment, used to perform local optimization and adjustment based on the initial operation path. Specifically, it uses the initial operation path generated in step S3 as a reference path, performs restricted sampling in its neighborhood space, and adaptively adjusts the sampling probability based on the risk field, so that low-risk areas have higher sampling priority. During the path adjustment process, a path risk integral is introduced to continuously evaluate the cumulative risk along the path, and the initial operation path is locally dynamically corrected in combination with the dynamic safety constraint region to construct a set of candidate paths that meet the safety envelope constraints and have the optimal risk integral. At the same time, global landmarks are pre-generated as structural support, and the candidate paths are optimized in combination with the local path replanning method to obtain the operation execution path.

[0028] Step S44: Graded response control, specifically, by calculating the comprehensive risk value of each grid node on the operation execution path, and extracting the maximum comprehensive risk value as the risk trigger indicator, the current operation status is graded based on the risk trigger indicator, and the operation status is divided into warning level, mandatory intervention level and evacuation level; the corresponding graded response control strategy is executed in different operation statuses to realize the dynamic switching of control strategy.

[0029] This invention provides a method for optimizing and managing live-line work processes in complex terrain, including a work space modeling module, a risk field construction module, a preliminary work process generation module, and a work process optimization module;

[0030] The operation space modeling module is used to model the operation space in complex terrain, obtain a structured three-dimensional operation space model, and send the structured three-dimensional operation space model to the risk field construction module, the preliminary operation process generation module, and the operation process optimization module.

[0031] The risk field construction module is used to construct the risk field in real time, obtain the risk field construction result, and send the risk field construction result to the preliminary work process generation module and the work process optimization module.

[0032] The preliminary work process generation module is used to generate a work process, obtain a structured work process scheme, and send the structured work process scheme to the work process optimization module.

[0033] The job process optimization module is used to dynamically optimize the job process and obtain a real-time optimized job execution plan.

[0034] The beneficial effects achieved by the present invention using the above solution are as follows:

[0035] (1) In response to the technical problems of variable environmental factors, dispersed risk sources and difficulty in unified assessment during live-line work in complex terrain, this solution creatively adopts a real-time risk field construction method based on multi-source risk information fusion. This method realizes unified modeling and spatial expression of various risk factors such as wind speed change, electric field strength, terrain stability and personnel working status. This allows the risk level of different locations in the work area to be intuitively represented in the form of continuous spatial distribution and to be dynamically updated with environmental changes. This provides a unified and reliable risk assessment basis for work path planning and task arrangement, and improves the scientificity and consistency of work decisions.

[0036] (2) In view of the technical problem that the existing live-line work process relies on a preset plan and is difficult to adjust according to changes in on-site risks, which may lead to personnel entering high-risk areas or failing to take timely avoidance measures during the operation, thereby increasing the safety hazards of the operation, this solution creatively adopts a risk field-driven dynamic optimization method for the operation process. It realizes the dynamic adjustment of personnel's work path, action arrangement and work rhythm according to real-time risk changes during the operation process, so that the operation process can actively avoid areas where risks rise rapidly, and trigger control measures such as deceleration, suspension or evacuation in a timely manner when the risk exceeds the limit, thereby significantly reducing the degree of risk exposure during the operation process and improving the timeliness and effectiveness of response to sudden risks.

[0037] (3) In view of the technical problem that the selection of live-line working paths and the arrangement of processes under existing complex terrain conditions mainly rely on human experience and lack systematic collaborative optimization, this solution realizes the transformation of work path planning and task arrangement from experience-driven to collaborative decision-making based on environment and risk by performing structured modeling of the work space, unified assessment of multi-source risks, and generating work processes and dynamic adjustment mechanisms in the implementation process. This enables workers to continuously perceive changes in surrounding risks during the work process and adjust the work path and operation rhythm accordingly, thereby effectively reducing the time spent in high-risk areas, improving the rationality of work paths and overall work efficiency, and enhancing the ability to respond to sudden risks and the safety and controllability of the work process. Attached Figure Description

[0038] Figure 1 A flowchart illustrating an optimized management method for live-line work in complex terrain, provided by the present invention.

[0039] Figure 2 A schematic diagram of a live-line work process optimization management system for complex terrain provided by the present invention;

[0040] Figure 3 This is a flowchart illustrating step S2;

[0041] Figure 4 This is a flowchart illustrating step S4.

[0042] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0045] Example 1, see Figure 1 This invention provides an optimized management method for live-line work processes in complex terrain, the method comprising the following steps:

[0046] Step S1: Spatial modeling for complex terrain operations;

[0047] Step S2: Real-time risk field construction;

[0048] Step S3: Job workflow generation;

[0049] Step S4: Dynamic optimization of the operation process.

[0050] By performing the above operations, this solution addresses the technical problem that the selection of live-line working paths and the arrangement of processes under existing complex terrain conditions mainly rely on manual experience and lack systematic collaborative optimization. This solution achieves a shift from experience-driven to environment- and risk-based collaborative decision-making in work path planning and task arrangement by performing structured modeling of the work space, conducting unified assessment of multi-source risks, and generating work processes and dynamic adjustment mechanisms during implementation. This enables workers to continuously perceive changes in surrounding risks and adjust work paths and operating rhythm accordingly, thereby effectively reducing the time spent in high-risk areas, improving the rationality of work paths and overall work efficiency, while enhancing the ability to respond to sudden risks and the safety and controllability of the work process.

[0051] Example 2, see Figure 1This embodiment is based on the above embodiment. In step S1, the complex terrain operation space modeling is used to establish a unified spatial expression basis required for subsequent risk field construction and operation process optimization. Specifically, it involves acquiring spatial data of the operation area, preprocessing the spatial data and expressing the terrain, identifying and spatially calibrating the power facilities and environmental elements in the operation area, then discretizing the continuous space into three-dimensional spatial grid nodes, assigning corresponding spatial attributes and semantic tags to each grid node, and establishing adjacency relationships between grid nodes to form a structured three-dimensional operation space model containing spatial coordinates, terrain information, facility distribution information and accessibility identifiers.

[0052] The acquisition of spatial data of the work area specifically involves using a drone equipped with a lidar sensor and a visible light camera to conduct aerial surveys of the work area, acquiring point cloud data and image data covering terrain undulations, power facility structures and surface cover. The point cloud data is used to characterize spatial geometric structure information, and the image data is used to assist in semantic recognition and target labeling.

[0053] The preprocessing and terrain representation of spatial data specifically involves filtering and denoising point cloud data, removing outliers, and classifying ground points, and generating a digital elevation model based on the ground points to obtain a continuous terrain surface that characterizes the terrain undulation features; preferably, the spatial resolution of the digital elevation model is 0.5m to 1m.

[0054] The identification and spatial labeling of power facilities and environmental elements within the work area specifically involves identifying towers, conductors, and insulator strings based on the geometric and strength features of point cloud data and obtaining their spatial coordinates. At the same time, vegetation areas, obstacle areas, steep slope areas, and potentially unstable areas are labeled to form spatial distribution information of work environment elements.

[0055] The discretization of continuous space into three-dimensional spatial grid nodes specifically involves dividing the work area according to a regular grid, constructing a set of spatial grid nodes, with each grid node corresponding to a spatial voxel unit, and assigning spatial coordinate information to each grid node; preferably, the resolution of the three-dimensional grid is 1m×1m×0.5m;

[0056] The process of assigning corresponding spatial attributes and semantic tags to each grid node specifically involves assigning terrain height attributes, land feature type attributes, and accessibility identifiers to each grid node based on the terrain model, power facility distribution, and environmental annotation results.

[0057] The establishment of adjacency relationships between grid nodes specifically involves constructing adjacency relationships between grid nodes based on their spatial location, preferably using a six-neighbor or twenty-six-neighbor method for connection.

[0058] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the real-time risk field construction is used to realize the dynamic perception and accurate assessment of multi-source risks in complex terrain live-line working environments. Specifically, based on the structured three-dimensional work space model, a real-time risk field construction method based on multi-source risk information fusion is adopted to obtain the risk field construction result containing the risk level of each grid node, risk field, risk evolution trend, and corresponding risk confidence interval. The steps include:

[0059] Step S21: Multi-source state modeling, used to realize unified modeling of multi-source data in complex terrain operation environment, specifically by collecting environmental information data and personnel state information data at each grid node, performing unified feature mapping, constructing state vector, and obtaining unified state expression results for each grid node;

[0060] The unified feature mapping specifically involves normalizing the dimensions and aligning the scales of the raw data from different data sources, standardizing the intervals of continuous variables, encoding the angles of directional variables, and numerically representing discrete variables, and then concatenating them according to a preset feature order to obtain a state vector.

[0061] The state vector includes wind speed, wind direction, electric field strength, soil saturation, surface displacement rate, personnel heart rate, and continuous operation time.

[0062] Step S22: Single-source risk assessment, used to characterize the degree of independent impact of different risk factors on operational safety. Specifically, based on state vectors, corresponding risk mapping functions are constructed for different types of risk sources. The wind speed risk value, electric field risk value, landslide risk value and fatigue risk value of each grid node are calculated independently to achieve quantitative assessment of single-source risk and obtain the single-source risk value corresponding to each grid node.

[0063] The wind speed risk value is used to characterize the impact of wind speed and terrain disturbance on the stability of live-line work, and the calculation formula is as follows:

[0064] ;

[0065] In the formula, This is the wind speed risk value of the k-th grid node, where k is the first index of the grid node, and f wind (·) is the risk mapping function, X k It is the state vector of the k-th grid node, exp(·) is the natural exponential function, v k It is the wind speed at the k-th grid node, v th It is the safe wind speed threshold for the k-th grid node. It is the topographic turbulence enhancement coefficient, used to characterize the amplification effect of complex terrain on airflow disturbance. Its value is determined based on the terrain undulation, roughness and obstacle density.

[0066] The electric field risk value is used to characterize the impact of the electric field strength in the work area on the safety of personnel and equipment. The calculation formula is as follows:

[0067] ;

[0068] In the formula, f is the electric field risk value of the k-th grid node. elec (·) is the electric field risk mapping function. This is a coefficient that adjusts the steepness of changes in electric field risk; its value is calibrated based on safety specifications or historical accident data. E k E is the electric field strength at the k-th grid node. safe It is the safe electric field threshold, which should preferably be set according to the safety specifications for live working or the equipment tolerance level;

[0069] The landslide risk value is used to characterize the impact of surface stability on operational safety, and the calculation formula is as follows:

[0070] ;

[0071] In the formula, f is the landslide risk value of the k-th grid node. slope (·) is the landslide risk mapping function, F s It is the slope stability coefficient, which is used to characterize the anti-sliding stability of a slope. The larger the value, the more stable the slope. It is specifically calculated based on soil saturation and surface displacement rate.

[0072] The fatigue risk value is used to characterize the impact of continuous working conditions on safety, and the calculation formula is as follows:

[0073] ;

[0074] In the formula, f is the fatigue risk value of the k-th grid node. fatigue (·) is the fatigue risk mapping function. , This is a weighting coefficient, the value of which is determined based on historical operation data. T k It refers to the continuous working time. It is a heart rate variability index used to characterize the fatigue state of personnel, specifically calculated based on the personnel's heart rate;

[0075] Step S23: Data confidence modeling is used to characterize the assessment reliability of different types of risk sources at each grid node. Specifically, for each type of risk source, based on the data input it depends on in the risk calculation process, the uncertainty index of the corresponding data source is obtained, and exponential decay modeling is performed to calculate the data confidence of each grid node under each type of risk source, and the risk source-level confidence distribution results are obtained.

[0076] The uncertainty index is composed of at least one of sensor measurement error, data missing rate and data fluctuation anomaly degree or a weighted combination thereof, and is used to characterize the uncertainty characteristics of the data source in terms of accuracy stability and integrity.

[0077] The calculation formula for the exponential decay model is as follows:

[0078] ;

[0079] In the formula, It represents the data confidence of the k-th grid node at time step t for the i-th type of risk source, where i is the first index of the risk source. It is the confidence decay coefficient of the i-th type of risk source. It is the data source uncertainty indicator corresponding to the i-th type of risk source;

[0080] Step S24: Comprehensive risk assessment, used to achieve adaptive fusion assessment of multi-source risks. Specifically, based on obtaining the single-source risk value and corresponding data confidence of each grid node, a dynamic weighting mechanism is introduced to construct a multi-source risk fusion model, realizing the nonlinear superposition and fusion of multi-source risks to obtain the comprehensive risk value of each grid node. The comprehensive risk value is then mapped to the spatial grid structure to form a risk field representing the spatial distribution of risks in the work area. Among these steps, risk contribution is suppressed through data confidence to reduce the impact of abnormal and incomplete data on the assessment results.

[0081] The dynamic weighting mechanism is specifically a confidence-driven weight allocation strategy;

[0082] Preferably, the formula for calculating the dynamic weight is:

[0083] ;

[0084] In the formula, w i (t) is the dynamic weight of the i-th type of risk source at time step t. Here, j is the global average confidence level of the i-th type of risk source at time step t, j is the second index of the risk source, and n is the number of risk sources. It is a minterm, used to prevent the denominator from being zero;

[0085] The formula for calculating the comprehensive risk value is as follows:

[0086] ;

[0087] In the formula, R k (t) is the comprehensive risk value of the k-th grid node at time step t. It is the single-source risk value of the i-th type of risk source at the k-th grid node;

[0088] Step S25: Spatiotemporal propagation modeling of risk field, used to characterize the diffusion process of risk in space and time. Specifically, it involves constructing node adjacency relationships based on spatial grid, calculating the risk propagation weight between grid nodes by combining wind direction consistency, slope influence and spatial distance attenuation factors, establishing a risk propagation model based on neighborhood risk differences, updating the risk field in time series, and obtaining the dynamic evolution result of the risk field.

[0089] The formula for calculating the risk propagation weight is as follows:

[0090] ;

[0091] In the formula, It is the risk propagation weight from the k-th grid node to the q-th grid node, where q is the second index of the grid node. , , It is an adjustment coefficient, and its value can be set by combining historical environmental data statistical analysis with engineering experience calibration. d kq It is the spatial distance between nodes. It is the direction angle from the k-th grid node to the q-th grid node. It is the prevailing wind direction, s kq It is the slope factor from the k-th grid node to the q-th grid node;

[0092] The calculation formula for the risk propagation model is as follows:

[0093] ;

[0094] In the formula, R k (t+1) is the comprehensive risk value of the k-th grid node at time step t, and N(k) is the set of neighboring nodes of the k-th grid node. It is a risk difference response function, preferably a ReLU or Sigmoid function, R q (t) is the comprehensive risk value of the q-th grid node at time step t;

[0095] Step S26: Risk confidence assessment, used to quantify the credibility of the risk assessment results. Specifically, based on the dynamic weights, data confidence and uncertainty variance of each type of risk source, the uncertainty of each risk source is weighted and propagated to obtain the risk fluctuation characterization value of each grid node, and a risk confidence interval is constructed based on the risk fluctuation characterization value to obtain the risk credibility expression result including the upper and lower bounds.

[0096] The formula for calculating the risk volatility characterization value is as follows:

[0097] ;

[0098] In the formula, w is the risk fluctuation representation value of the k-th grid node. i It is the dynamic weight of the i-th type of risk source. It is the data confidence level of the i-th type of risk source at the k-th grid node;

[0099] The formula for calculating the risk confidence interval is:

[0100] ;

[0101] In the formula, CI k R is the risk confidence interval for the k-th grid node. k It is the comprehensive risk value of the k-th grid node. It is the confidence adjustment coefficient, with a preferred range of [1.0, 2.0];

[0102] Step S27: Risk classification output, specifically, based on the dynamic evolution results of the risk field, the comprehensive risk value of each grid node is classified and determined, the corresponding risk level is generated, and the risk evolution trend is calculated;

[0103] The risk levels, from low to high, include safe, controllable, warning, dangerous, and prohibited levels.

[0104] By performing the above operations, this solution creatively adopts a real-time risk field construction method based on multi-source risk information fusion to address the technical problems of variable environmental factors, dispersed risk sources, and difficulty in unified assessment during live-line operations in complex terrain. This method achieves unified modeling and spatial representation of various risk factors such as wind speed changes, electric field strength, terrain stability, and personnel working status. It enables the risk level at different locations within the work area to be intuitively represented in a continuous spatial distribution form and can be dynamically updated with environmental changes. This provides a unified and reliable risk assessment basis for work path planning and task arrangement, improving the scientific nature and consistency of work decisions.

[0105] Example 4, see Figure 1This embodiment is based on the above embodiment. In step S3, the workflow generation specifically involves using a structured three-dimensional work space model, taking the risk field construction results, the set of work tasks, and the initial positions of personnel as inputs, and generating an initial work path in a three-dimensional space grid with risk constraints as boundary conditions through a path search algorithm. Simultaneously, a cost matrix between tasks is constructed based on path costs, and the task order is optimized using an approximate solution method for the traveling salesman problem. Then, time scheduling modeling is performed by combining path length and task duration to obtain a structured workflow scheme that includes path planning results, task execution order, and time arrangement.

[0106] The set of work tasks is pre-generated based on the live-line work plan or on-site maintenance tasks, and the initial position of the personnel is obtained through positioning equipment or set manually.

[0107] The path search algorithm may include, but is not limited to, the A algorithm, Dijkstra's algorithm, improved A algorithm, or a three-dimensional path search algorithm based on heuristic functions;

[0108] Preferably, the risk constraint as the boundary condition specifically involves constraining and filtering the risk values ​​of each grid node, allowing only grid nodes with risk values ​​below a preset threshold to participate in path search. The calculation formula is as follows:

[0109] ;

[0110] In the formula, G safe It is the set of walkable grid nodes, G k It is the kth grid node, R th It is a risk threshold, which is dynamically selected based on the risk level;

[0111] Preferably, the formula for calculating the path cost is:

[0112] ;

[0113] In the formula, g(q) is the path cost from the starting point to the q-th grid node, and path is the path from the starting point to the q-th grid node. It is the spatial distance between the k-th grid node and the (k+1)-th grid node in the path. It is a risk weighting coefficient, which is dynamically selected based on the risk level.

[0114] Example 5, see Figure 1 and Figure 4This embodiment is based on the above embodiment. In step S4, the dynamic optimization of the operation process is used to optimize and adjust the operation actions and paths based on real-time risk changes during live-line operations in complex terrain. Specifically, based on the structured three-dimensional operation space model, the risk field construction results and the structured operation process plan are used as inputs. The risk field-driven dynamic optimization method of the operation process is adopted to obtain a real-time optimized operation execution plan that includes the current optimal control strategy, dynamic safety constraint area, operation execution path and hierarchical response control strategy. This realizes the collaborative optimization management of the live-line operation process, thereby improving the safety of the operation, the rationality of the path and the execution efficiency in a dynamic risk environment.

[0115] The risk field-driven dynamic optimization method for work processes includes the following steps:

[0116] Step S41: Rolling time-domain predictive control modeling. Specifically, by obtaining the risk field construction results obtained in step S2 and the structured work process scheme generated in step S3, a system state variable including personnel spatial location, work posture state, task progress and tool state is constructed. A set of discrete control actions including movement, climbing, operation, pause and evacuation is defined, and a state transition model is established. Then, within the preset prediction time domain, an optimization function with risk cost, movement cost and task delay cost as objectives is constructed. The rolling time-domain predictive control is used to solve the work process to obtain the current optimal control strategy.

[0117] The current optimal control strategy includes a sequence of control actions generated within the prediction time domain;

[0118] Step S42: Adaptive contraction of the safety envelope is used to achieve dynamic adjustment of the safety constraints of the work space. Specifically, a safety envelope region is constructed with the current work position as the center, and a direction-related safety distance model is established based on the comprehensive risk value, risk change rate and spatial risk gradient of the current work position. The safety radius of different directions is adaptively adjusted to form an asymmetric safety envelope constraint, so that the safety boundary of the high-risk direction contracts and the low-risk direction remains passable, thus obtaining a direction-aware dynamic safety constraint region.

[0119] The calculation formula for the safety distance model is as follows:

[0120] ;

[0121] In the formula, It is the direction The safe distance below, The direction parameter d is the direction of movement from the current work position. base This is the basic safety distance. , , This is the safety distance weighting coefficient, which can be set by combining safety regulations, historical accident risk distribution statistics, and risk sensitivity analysis of the work scenario. R curr This is the current overall risk value, R. rate It is the rate of change of risk, used to characterize the speed at which the overall risk value evolves over time. It is along the direction The spatial risk gradient projection is used to characterize the upward trend of risk in that direction.

[0122] Step S43: Path optimization and adjustment, used to perform local optimization and adjustment based on the initial operation path. Specifically, it uses the initial operation path generated in step S3 as a reference path, performs restricted sampling in its neighborhood space, and adaptively adjusts the sampling probability based on the risk field, so that low-risk areas have higher sampling priority. During the path adjustment process, a path risk integral is introduced to continuously evaluate the cumulative risk along the path, and the initial operation path is locally dynamically corrected in combination with the dynamic safety constraint region to construct a set of candidate paths that meet the safety envelope constraints and have the optimal risk integral. At the same time, global landmarks are pre-generated as structural support, and the candidate paths are optimized in combination with the local path replanning method to obtain the operation execution path.

[0123] Step S44: Graded response control, specifically, by calculating the comprehensive risk value of each grid node on the operation execution path and extracting the maximum comprehensive risk value as the risk trigger index, the current operation status is graded based on the risk trigger index, and the operation status is divided into warning level, mandatory intervention level and evacuation level; corresponding graded response control strategies are executed under different operation statuses, including reducing the operation speed, prompting operators, pausing the current task, executing the minimum risk path evacuation, and adjusting the weights of each target in the rolling time domain predictive control, thereby realizing the dynamic switching of control strategies.

[0124] By performing the above operations, this solution addresses the technical problem that existing live-line working procedures rely on preset plans and are difficult to adjust according to changes in on-site risks, which can easily lead to personnel entering high-risk areas or failing to take timely avoidance measures during the operation, thereby increasing the risk of accidents. This solution creatively adopts a risk field-driven dynamic optimization method for the work process, which enables dynamic adjustment of personnel's work paths, action arrangements, and work rhythm based on real-time risk changes during the operation. This allows the work process to proactively avoid areas where risks rise rapidly and to trigger control measures such as deceleration, suspension, or evacuation in a timely manner when risks exceed limits, thereby significantly reducing the degree of risk exposure during the operation and improving the timeliness and effectiveness of response to sudden risks.

[0125] Example 6, see Figure 2Based on the above embodiments, this embodiment provides a method for optimizing and managing live-line work processes in complex terrain, including a work space modeling module, a risk field construction module, a preliminary work process generation module, and a work process optimization module.

[0126] The operation space modeling module is used to model the operation space in complex terrain, obtain a structured three-dimensional operation space model, and send the structured three-dimensional operation space model to the risk field construction module, the preliminary operation process generation module, and the operation process optimization module.

[0127] The risk field construction module is used to construct the risk field in real time, obtain the risk field construction result, and send the risk field construction result to the preliminary work process generation module and the work process optimization module.

[0128] The preliminary work process generation module is used to generate a work process, obtain a structured work process scheme, and send the structured work process scheme to the work process optimization module.

[0129] The job process optimization module is used to dynamically optimize the job process and obtain a real-time optimized job execution plan.

[0130] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0131] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0132] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for optimizing and managing live-line working processes in complex terrain, characterized in that: The method includes the following steps: Step S1: Complex terrain operation space modeling, resulting in a structured three-dimensional operation space model containing spatial coordinates, terrain information, facility distribution information, and accessibility markers; Step S2, real-time risk field construction, is based on the structured three-dimensional work space model and adopts a real-time risk field construction method based on multi-source risk information fusion. The result is a risk field construction result that includes the risk level of each grid node, risk field, risk evolution trend, and corresponding risk confidence interval. The steps are as follows: Step S21, multi-source state modeling; Step S22, single-source risk assessment; Step S23, data confidence modeling; Step S24, comprehensive risk assessment; Step S25, risk field spatiotemporal propagation modeling; Step S26, risk confidence assessment; and Step S27, risk classification output. Step S3 generates a work process, resulting in a structured work process plan that includes path planning results, task execution order, and time schedule. Step S4 is to dynamically optimize the operation process. Based on the structured three-dimensional operation space model, the risk field construction results and the structured operation process plan are used as inputs. The risk field-driven dynamic optimization method of the operation process is adopted to obtain a real-time optimized operation execution plan that includes the current optimal control strategy, dynamic safety constraint area, operation execution path and hierarchical response control strategy, so as to realize the collaborative optimization management of the live operation process. The risk field-driven dynamic optimization method for the operation process includes the following steps: step S41 rolling time-domain predictive control modeling, step S42 adaptive shrinking of the safety envelope, step S43 path optimization adjustment, and step S44 hierarchical response control.

2. The method for optimizing and managing live-line working processes in complex terrain according to claim 1, characterized in that: In step S21, the multi-source state modeling is used to realize the unified modeling of multi-source data in complex terrain operation environment. Specifically, it involves collecting environmental information data and personnel state information data at each grid node, performing unified feature mapping, constructing a state vector, and obtaining the unified state expression result of each grid node. The state vector includes wind speed, wind direction, electric field strength, soil saturation, surface displacement rate, personnel heart rate, and continuous operation time. In step S22, the single-source risk assessment is used to characterize the degree of independent impact of different risk factors on operational safety. Specifically, based on the state vector, corresponding risk mapping functions are constructed for different types of risk sources, and the wind speed risk value, electric field risk value, landslide risk value and fatigue risk value of each grid node are calculated independently to realize the quantitative assessment of single-source risk and obtain the single-source risk value corresponding to each grid node. The wind speed risk value is used to characterize the impact of wind speed and terrain disturbance on the stability of live-line work; The electric field risk value is used to characterize the impact of the electric field strength in the work area on the safety of personnel and equipment; The landslide risk value is used to characterize the impact of surface stability on operational safety; the fatigue risk value is used to characterize the impact of continuous working conditions on safety.

3. The method for optimizing and managing live-line working processes in complex terrain according to claim 2, characterized in that: In step S23, the data confidence modeling is used to characterize the assessment reliability of different types of risk sources at each grid node. Specifically, for each type of risk source, based on the data input it depends on in the risk calculation process, the uncertainty index of the corresponding data source is obtained, and exponential decay modeling is performed to calculate the data confidence of each grid node under each type of risk source, so as to obtain the risk source-level confidence distribution result. In step S24, the comprehensive risk assessment is used to achieve adaptive fusion assessment of multi-source risks. Specifically, based on obtaining the single-source risk value and corresponding data confidence of each grid node, a dynamic weighting mechanism is introduced to construct a multi-source risk fusion model, realize the nonlinear superposition and fusion of multi-source risks, obtain the comprehensive risk value of each grid node, and map the comprehensive risk value to the spatial grid structure to form a risk field that characterizes the spatial distribution of risks in the work area.

4. The method for optimizing and managing live-line working processes in complex terrain according to claim 3, characterized in that: In step S25, the risk field spatiotemporal propagation modeling is used to characterize the diffusion process of risk in space and time. Specifically, it involves constructing node adjacency relationships based on a spatial grid, calculating the risk propagation weights between grid nodes by combining wind direction consistency, slope influence, and spatial distance attenuation factors, establishing a risk propagation model based on neighborhood risk differences, performing time-series updates of the risk field, and obtaining the dynamic evolution results of the risk field. In step S26, the risk confidence assessment is used to quantify the credibility of the risk assessment results. Specifically, based on the dynamic weights, data confidence and uncertainty variance of each type of risk source, the uncertainty of each risk source is weighted and propagated to obtain the risk fluctuation characterization value of each grid node. Based on the risk fluctuation characterization value, a risk confidence interval is constructed to obtain the risk credibility expression result including the upper and lower bounds. In step S27, the risk classification output specifically involves classifying and determining the comprehensive risk value of each grid node based on the dynamic evolution results of the risk field, generating the corresponding risk level, and calculating the risk evolution trend.

5. The method for optimizing and managing live-line working processes in complex terrain according to claim 4, characterized in that: In step S3, the workflow generation specifically involves using a structured 3D work space model, taking the risk field construction results, the set of work tasks, and the initial positions of personnel as inputs, and generating an initial work path in a 3D spatial grid with risk constraints as boundary conditions through a path search algorithm. Simultaneously, a cost matrix between tasks is constructed based on path costs, and the task order is optimized using an approximate solution method for the traveling salesman problem. Then, time scheduling modeling is performed by combining path length and task duration to obtain a structured workflow scheme that includes path planning results, task execution order, and time arrangement.

6. The method for optimizing and managing live-line working processes in complex terrain according to claim 5, characterized in that: In step S41, the rolling time-domain predictive control modeling specifically involves acquiring the risk field construction results obtained in step S2 and the structured work process scheme generated in step S3, constructing system state variables including personnel spatial location, work posture state, task progress, and tool state, defining a set of discrete control actions including movement, climbing, operation, pause, and evacuation, establishing a state transition model, and then constructing an optimization function with risk cost, movement cost, and task delay cost as objectives within a preset prediction time domain, performing rolling time-domain predictive control to solve the work process, and obtaining the current optimal control strategy; In step S42, the adaptive shrinkage of the safety envelope is used to achieve dynamic adjustment of the safety constraints of the work space. Specifically, a safety envelope region is constructed with the current work position as the center, and a direction-related safety distance model is established based on the comprehensive risk value, risk change rate, and spatial risk gradient of the current work position. The safety radius in different directions is adaptively adjusted to form an asymmetric safety envelope constraint, so that the safety boundary in the high-risk direction shrinks while the low-risk direction remains passable, thus obtaining a direction-aware dynamic safety constraint region.

7. The method for optimizing and managing live-line working processes in complex terrain according to claim 6, characterized in that: In step S43, the path optimization adjustment is used to perform local optimization adjustment based on the initial operation path. Specifically, it involves using the initial operation path generated in step S3 as a reference path, performing restricted sampling in its neighborhood space, and adaptively adjusting the sampling probability based on the risk field to give low-risk areas a higher sampling priority. During the path adjustment process, a path risk integral is introduced to continuously assess the cumulative risk along the path. Combined with the dynamic safety constraint region, the initial operation path is locally and dynamically corrected to construct a set of candidate paths that meet the safety envelope constraints and have the optimal risk integral. At the same time, global landmarks are pre-generated as structural support, and combined with the local path replanning method, the candidate paths are optimized to obtain the operation execution path. In step S44, the hierarchical response control specifically involves calculating the comprehensive risk value of each grid node on the job execution path, extracting the maximum comprehensive risk value as a risk trigger indicator, and classifying the current job status based on the risk trigger indicator, dividing the job status into warning level, mandatory intervention level, and evacuation level; and executing the corresponding hierarchical response control strategy under different job statuses to achieve dynamic switching of control strategies.

8. The method for optimizing and managing live-line work processes in complex terrain according to claim 7, characterized in that: In step S1, the complex terrain operation space modeling is used to establish a unified spatial representation basis required for subsequent risk field construction and operation process optimization. Specifically, it involves acquiring spatial data of the operation area, preprocessing the spatial data and representing the terrain, identifying and spatially calibrating the power facilities and environmental elements in the operation area, then discretizing the continuous space into three-dimensional spatial grid nodes, assigning corresponding spatial attributes and semantic labels to each grid node, and establishing adjacency relationships between grid nodes to form a structured three-dimensional operation space model containing spatial coordinates, terrain information, facility distribution information, and accessibility identifiers.

9. A live-line work process optimization management system for complex terrain, used to implement the live-line work process optimization management method for complex terrain as described in any one of claims 1-8, characterized in that: It includes a workspace modeling module, a risk field construction module, a preliminary work process generation module, and a work process optimization module; The operation space modeling module is used to model the operation space in complex terrain, obtain a structured three-dimensional operation space model, and send the structured three-dimensional operation space model to the risk field construction module, the preliminary operation process generation module, and the operation process optimization module. The risk field construction module is used to construct the risk field in real time, obtain the risk field construction result, and send the risk field construction result to the preliminary work process generation module and the work process optimization module. The preliminary work process generation module is used to generate a work process, obtain a structured work process scheme, and send the structured work process scheme to the work process optimization module. The job process optimization module is used to dynamically optimize the job process and obtain a real-time optimized job execution plan.

Images