Intelligent recommendation system for inspection path of outer riser

Abstract

The invention relates to the technical field of patrol path optimization, and discloses an outer riser patrol path intelligent recommendation system, which comprises a patrol unit for recording a facade position, a center height and a patrol operation mode; the evaluation unit performs risk evaluation on each inspection unit according to historical defect information, riser types and environmental exposure conditions to obtain an inspection unit risk level and form an inspection unit time window constraint; the processing unit constructs an inspection business graph; and the recommendation unit determines the unit to be inspected, generates candidate inspection paths on the inspection service graph, combines the candidate inspection paths into a plurality of inspection task packets, and performs resource scheduling in combination with skill information and available time information of the inspection teams to obtain an inspection path recommendation result corresponding to each inspection team. According to the invention, the total inspection time and the switching frequency of high-cost inspection operation modes are reduced, and the inspection safety and the operation and maintenance management efficiency of the outer riser are improved.

Description

Technical Field

[0001] This invention relates to the field of inspection route optimization technology, and more specifically, to an intelligent recommendation system for external riser inspection routes. Background Technology

[0002] With the increasing number of existing buildings in cities and the long-term exposure of external pipes to wind, sun, temperature cycles, and pollution, the risks of corrosion, leakage, and detachment of facilities such as water supply, drainage, and gas external pipes are continuously rising. Property management and operation and maintenance units need to regularly organize inspections and maintenance of external pipes on building facades to prevent safety accidents such as leaks, gas leaks, and falls that could cause injuries to residents due to pipe failure. Traditional external pipe inspections rely heavily on manual experience to formulate inspection plans, with team leaders assigning tasks by building and responsibility area. Inspectors then check each building and facade along familiar routes, which is insufficient to meet the needs of refined operation and maintenance management in large-scale building complexes.

[0003] While existing inspection management and route planning technologies have incorporated comprehensive considerations of multiple sites, multiple inspectors, and risk factors, they primarily focus on general sites or equipment. They lack data models and management scheduling mechanisms specifically for scenarios involving building facade risers, which involve multiple floors, multiple work modes, and safety constraints related to high-altitude operations. For example, the invention patent publication number CN110826818A describes a method for multiple inspectors to plan inspection tasks and design routes for multiple sites. By preprocessing multiple sites, balancing task allocation, and identifying risks, it generates multiple inspection routes within the constraints of inspectors' daily working hours, reducing the overall operational risk of maintenance sites with lower inspection costs. However, this solution abstracts the inspection objects as planar sites, fails to classify the height and label the work modes of the risers distributed along the building facade axis, does not establish an integrated inspection business map encompassing access connecting edges, work connecting edges, and work mode switching connecting edges, and does not incorporate work mode switching costs into route recommendations.

[0004] Therefore, it is necessary to design an intelligent recommendation system for external riser inspection routes to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent recommendation system for inspection paths of external risers, which aims to solve the problems of high cost and unsatisfactory inspection results when switching between multiple inspection operation modes in the context of external risers in the existing technology.

[0006] This invention proposes an intelligent recommendation system for external riser inspection routes, comprising: Inspection unit: Each external riser is provided with multiple inspection units along the axial direction, and each inspection unit records the facade position, center height and inspection operation method. The inspection operation method includes at least one of walking operation, suspended platform operation, rope operation and drone operation. The assessment unit is configured to perform a risk assessment on each of the inspection units based on historical defect information, riser type and environmental exposure conditions, obtain the risk level of the inspection unit, and determine the inspection cycle and the latest allowed inspection date based on the risk level of the inspection unit, thus forming a time window constraint for the inspection unit. The processing unit is configured to construct an inspection business map based on the access locations within the building and the inspection units. In the inspection business map, access relationships within the building are represented by access connection edges, operation relationship relationships between inspection units and corresponding operation locations are represented by operation mode switching connection edges, and switching relationships between different inspection operation modes are represented by operation mode switching connection edges. Access time cost is set on access connection edges, operation time cost is set on operation connection edges, and operation mode switching cost is set on operation mode switching connection edges. The recommendation unit is configured to select inspection units whose time window constraints expire or whose risk levels reach preset conditions within a preset planning period as inspection units to be inspected. On the inspection business map, candidate inspection paths are generated by comprehensively considering travel time cost, operation time cost, operation mode switching cost, and inspection unit time window constraints. The candidate inspection paths are then combined into multiple inspection task packages. Resource scheduling is performed by combining the skill information and available time information of the inspection team to obtain the inspection path recommendation result corresponding to each inspection team.

[0007] Furthermore, each external riser is provided with multiple inspection units along its axial direction, and each inspection unit records the facade position, center height, and inspection operation method, including: Collect external riser layout data, and divide each external riser into multiple inspection sections along the axis according to the starting height and ending height of each external riser, based on the floor boundaries and preset height intervals, and set each inspection section as an inspection unit. For each inspection unit, the facade orientation, center height, horizontal distance and height difference between the inspection unit and the work position on the balcony, windowsill, roof or ground are determined according to the pre-stored inspection operation mode selection rules, and the inspection operation mode, facade position and center height of the inspection unit are recorded in the inspection unit.

[0008] Furthermore, when the assessment unit performs a risk assessment on each of the inspection units, it includes: The assessment unit statistically analyzes the number of historical defects, the severity of historical defects, and the time since the last maintenance for each inspection unit, and combines this with the riser type, connector type, and service life of the inspection unit to form basic risk information. The basic risk information is weighted and analyzed in conjunction with the orientation of the facade where the inspection unit is located, the degree of orientation towards the prevailing wind direction, the degree of street proximity, and the environmental exposure conditions to obtain the risk level of the inspection unit.

[0009] Furthermore, when determining the inspection cycle and the latest permitted inspection date based on the risk level of the inspection unit, the assessment unit includes: The evaluation unit selects the corresponding basic inspection cycle and basic grace period based on the risk level of the inspection unit, and shortens or extends the basic inspection cycle and basic grace period by combining the last actual inspection date and the most recent defect handling date of the inspection unit, thereby obtaining the inspection cycle and the latest allowed inspection date of the inspection unit, and forming a time window constraint for the inspection unit based on the inspection cycle and the latest allowed inspection date; wherein, when defects are found multiple times in the same inspection unit within a continuous planning cycle, the evaluation unit increases the inspection unit risk level of the inspection unit and shortens the inspection cycle of the inspection unit accordingly.

[0010] Furthermore, when constructing the inspection service map, the processing unit includes: Based on the building floor plan data and floor access data, the access locations within the building are determined at stairwell entrances, elevator lobbies, corridor turning points, roof entrances and exits, and ground passage nodes, and these access locations within the building are designated as access nodes. Based on the facade location and center height of the inspection unit, determine the work location at the balcony, window sill, roof edge and ground work point, and associate each work location with at least one access node. When there are direct staircases, elevators, or corridors connecting the buildings, the processing unit establishes a passage connection edge between the corresponding passage nodes. When personnel survey or the work platform can reach the corresponding work position, it establishes a work connection edge between the passage node and the work position. Only when the inspection work methods corresponding to adjacent work positions are different and the pre-set work mode switching safety conditions are met, a work mode switching connection edge is established between adjacent work positions.

[0011] Furthermore, when the processing unit sets the passage time cost on the passage connection edge, it includes: Based on the floor height difference, horizontal distance, medium type, and whether security access is required at the two ends of the access connection edge, combined with preset walking speed, stair up and down speed, elevator running speed, and average elevator waiting time, the estimated travel time of personnel on the access connection edge is calculated, and the estimated travel time is recorded as the travel time cost in the corresponding access connection edge.

[0012] Furthermore, when the processing unit sets the job time cost on the job connection edge and sets the job mode switching cost on the job mode switching connection edge, it includes: The processing unit determines the preparation time and observation time of the corresponding work connection edge based on the height of the inspection unit, the horizontal distance between the work position and the external riser, and the work specifications of the corresponding inspection work method, and records the sum of the preparation time and observation time as the work time cost in the work connection edge. When switching from one inspection operation mode to another, the operation mode switching time is determined based on the basket deployment time, rope deployment time, drone take-off and landing preparation time, and safety inspection time. The operation mode switching time and the preset weighted time used to penalize the number of operation mode switching times are recorded as the operation mode switching cost in the corresponding operation mode switching connection edge.

[0013] Furthermore, when the recommendation unit selects inspection units whose time window constraints have expired or whose risk levels have reached preset conditions as inspection units within a preset planning period, it includes: The recommendation unit calculates the remaining time until the latest allowed inspection date based on the inspection unit time window constraint of each inspection unit. It then performs a comprehensive analysis of the remaining time and the risk level of the inspection unit according to the pre-stored urgency assessment rules to form an urgency value. Based on the urgency value, the units are sorted from high to low, and the inspection units whose urgency value reaches the preset urgency threshold are selected as the units to be inspected.

[0014] Furthermore, when the recommendation unit generates candidate inspection paths, it includes: The recommendation unit, for each inspection team's starting point, takes the starting point as the starting node in the inspection business map, selects the unit to be inspected under the premise of satisfying the working time constraints recorded in the available time information of the corresponding inspection team, and extends the path between the access connection edge, the operation connection edge, and the operation mode switching connection edge through a heuristic path search strategy. The sum of the access time cost, the operation time cost, and the operation mode switching cost is used as the path evaluation basis. At the same time, a penalty is imposed on the path segment that violates the time window constraint of the inspection unit, and the candidate inspection path is generated.

[0015] Furthermore, when the recommendation unit combines the candidate inspection paths into multiple inspection task packages and combines the skill information and available time information of the inspection teams to perform resource scheduling to obtain the inspection path recommendation result corresponding to each inspection team, it includes: The recommendation unit uses the set of units to be inspected corresponding to each candidate inspection path, the type of inspection operation mode included in the candidate inspection path, the total travel time cost and the total operation time cost of the candidate inspection path as the inspection task package attributes. It matches the executable inspection operation mode recorded in the inspection team skill information with the inspection operation mode required by the inspection task package and eliminates combinations in which the inspection team skill information does not meet the inspection task package attributes. Under the premise that the sum of the cumulative travel time cost and the cumulative operation time cost of each inspection team does not exceed the available time recorded in the available time information of the corresponding inspection team, and that high-risk inspection units are covered first, the allocation scheme between each inspection task package and the inspection team is comprehensively evaluated based on the number of high-risk units to be inspected, the number of violations of the inspection unit time window constraint, and the cost of switching operation mode. The allocation scheme with the highest score is selected as the resource scheduling result, and the inspection path recommendation result corresponding to each inspection team is obtained.

[0016] Compared with the prior art, the beneficial effects of this invention are as follows: By subdividing each external riser along the axial direction into multiple inspection units and recording the inspection operation mode for each inspection unit, a detailed analysis of the inspection objects of high-altitude external risers is achieved. The evaluation unit, combining historical defect information, riser type, and environmental exposure conditions, gives the risk level of the inspection unit and forms a time window constraint for the inspection unit, enabling high-risk and near-expiration inspection units to be identified and prioritized. The processing unit further constructs an inspection business map containing access connection edges, operation connection edges, and operation mode switching connection edges based on the access positions within the building and the inspection units, and respectively... By setting access time cost, operation time cost, and operation mode switching cost on the aforementioned connecting edges, the time spent on accessing the building, the time spent on high-altitude operations, and the additional costs brought about by switching between different inspection operation modes are uniformly quantified into the same graph model. The recommendation unit generates candidate inspection paths, and combines the candidate inspection paths into multiple inspection task packages. Then, it combines the skill information and available time information of the inspection team for resource scheduling, ensuring that high-risk and near-expiration inspection units are prioritized for timely coverage, reducing the overall inspection mileage and the number of high-cost inspection operation mode switching times, improving the utilization rate of inspection team working time and the safety of external riser inspection. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a functional block diagram of the intelligent recommendation system for external riser inspection routes provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] In traditional external pipe inspection and management, the lack of refined data models that address the multi-story axial distribution characteristics of external pipes on building facades, the switching process between different work modes, and the safety constraints of high-altitude operations leads to inspection route planning relying on manual experience. This fails to integrate information on the physical height of external pipes, the cost of switching work modes, and a dynamic risk assessment mechanism. Consequently, inspection route planning is inefficient, high-risk areas are not fully covered, and high-altitude work safety hazards are prominent. Specifically, the physical characteristics of the external pipes' axial distribution along the building facade are not included in the data modeling, and the time costs and safety constraints during work mode switching are not quantified. This makes route planning unable to adapt to the operational needs of different floor heights, resulting in unreasonable allocation of inspection resources and the accumulation of safety risks.

[0020] For example, in a scenario involving the inspection of external risers in a high-rise residential community in a city, the external risers on the building facade are distributed along the axial direction at different floor heights. Inspectors need to switch between walking work areas on lower floors, suspended platform work areas on mid-to-high floors, and drone work areas on the roof. The existing inspection management system does not divide each external riser along the axial direction into multiple inspection units that record the facade position and center height, nor does it define work mode switching connection edges and set switching costs in the inspection business map. When the inspection path involves switching from the walking area on lower floors to the suspended platform area on higher floors, the system cannot quantify the suspended platform deployment time and safety inspection time, causing inspectors to frequently make unnecessary work mode switching during path planning, increasing the risk of high-altitude operations and reducing overall inspection efficiency. Furthermore, risk assessment relies only on a fixed period without dynamically adjusting based on historical defect information and environmental exposure conditions, resulting in high-risk areas not being prioritized for inclusion in the inspection plan and potential defects being overlooked.

[0021] If the above issues are not addressed, high-risk areas may be missed during external riser inspections due to unreasonable route planning, increasing the risk of corrosion, leakage, or detachment of the risers, which could lead to safety accidents such as leaks, gas leaks, or falls causing injuries to residents. Simultaneously, reduced inspection efficiency will render maintenance resources unusable, making it difficult to meet the demands of refined maintenance management in large-scale building complexes. Consequently, the safety and reliability of external riser inspections will be affected, and maintenance units will face higher pressure for safety supervision and increased accident prevention costs.

[0022] For this, please refer to Figure 1 As shown, this application proposes an intelligent recommendation system for external riser inspection routes, including: Inspection unit: Multiple inspection units are set along the axial direction for each external riser, and each inspection unit records the facade position, center height and inspection operation method. The inspection operation method includes at least one of walking operation, suspended platform operation, rope operation and drone operation. The assessment unit is configured to perform a risk assessment on each inspection unit based on historical defect information, riser type and environmental exposure conditions, obtain the risk level of the inspection unit, and determine the inspection cycle and the latest allowed inspection date based on the risk level of the inspection unit, thus forming the time window constraint of the inspection unit. The processing unit is configured to construct an inspection business map based on the access locations within the building and the inspection units. In the inspection business map, access relationships within the building are represented by access connection edges, operation relationship relationships between inspection units and corresponding operation locations are represented by operation mode switching connection edges, and switching relationships between different inspection operation modes are represented by operation mode switching connection edges. Access time cost is set on access connection edges, operation time cost is set on operation connection edges, and operation mode switching cost is set on operation mode switching connection edges. The recommendation unit is configured to select inspection units whose time window constraints expire or whose risk level reaches preset conditions within a preset planning period as inspection units to be inspected. On the inspection business map, candidate inspection paths are generated by comprehensively considering travel time cost, operation time cost, operation mode switching cost, and inspection unit time window constraints. The candidate inspection paths are combined into multiple inspection task packages. Resource scheduling is performed by combining the skill information and available time information of the inspection team to obtain the inspection path recommendation result for each inspection team.

[0023] This embodiment provides an intelligent recommendation system for external riser inspection routes. An inspection unit is defined as a physical segment along the axial direction of the external riser. Its facade location is determined as a specific position in the building facade coordinate system, its center height is measured, and the inspection operation method is selected as one or more of walking, suspended platform, rope, or drone operations. In implementation, the inspection operation method can be manually specified by on-site surveyors according to safety regulations, or determined by querying a preset operation method selection table. For example, the operation method can be directly matched based on the external riser height and operation location type, primarily to achieve targeted inspections of external riser segments at different heights. Furthermore, the risk assessment process is configured to calculate the risk level based on historical defect information, riser type, and environmental exposure conditions. Specifically, risk assessment can be implemented through statistical analysis of historical data or expert system scoring. For example, the number of historical defects and the degree of environmental exposure can be input into the risk assessment model to generate a risk level, thereby forming a time window constraint for the inspection unit. Specifically, the inspection operation map is constructed as a graph model containing access connection edges, operation connection edges, and operation mode switching connection edges. Access connection edges represent internal building access relationships, operation connection edges represent the operational relationships between inspection units and operation locations, and operation mode switching connection edges represent the switching relationships between different inspection operation modes. Time costs can be defined as access time cost, operation time cost, and operation mode switching cost, which can be calculated as functions based on distance, speed, and preparation time. For example, access time cost is calculated based on travel distance and walking speed, operation time cost considers equipment preparation and observation time, and operation mode switching cost includes equipment deployment and safety inspection time. As a preferred implementation, candidate inspection paths are generated by applying a graph search algorithm to the inspection operation map, comprehensively considering access time cost, operation time cost, operation mode switching cost, and inspection unit time window constraints to optimize the path. Therefore, this application constructs a multi-dimensional collaborative inspection business model by deeply integrating the physical characteristics of external risers, dynamic risk assessment, and operational constraints. This overcomes the limitations of existing technologies that only abstract inspection objects as planar stations. It solves the problem that the lack of a data model for multi-story distribution, multiple operation mode switching, and high-altitude operation safety constraints in the inspection of external risers on building facades leads to low efficiency, incomplete risk coverage, and prominent safety hazards in traditional manual experience-based path planning.

[0024] This intelligent recommendation system for external riser inspection routes achieves refined planning and dynamic scheduling of inspection routes for building facades through the collaborative operation of inspection units, evaluation units, processing units, and recommendation units. The inspection unit divides each external riser into multiple independent units along its axial direction. Each unit records its facade location, center height, and suitable inspection operation method, which includes at least one of walking, suspended platform, rope, and drone operations, thus transforming the physical characteristics of the external riser into quantifiable data nodes. The evaluation unit conducts a risk assessment for each inspection unit based on historical defect information, riser type, and environmental exposure conditions, generating a risk level for each unit. Based on this risk level, it determines the inspection cycle and the latest allowed inspection date, forming a time window constraint for the inspection unit to ensure priority coverage of high-risk areas. The processing unit constructs an inspection business map based on the access locations within the building and the inspection units. In this map, access connection edges represent the relationships between internal access paths such as stairs and elevators, operation connection edges represent the operational relationships between inspection units and operation locations such as balconies and windowsills, and operation mode switching connection edges represent the conversion relationships between different inspection operation modes. Simultaneously, access time costs are set on access connection edges, operation time costs on operation connection edges, and operation mode switching costs on operation mode switching connection edges, quantifying physical space constraints and time consumption into graph model parameters. The recommendation unit selects inspection units whose time window constraints expire or whose risk level meets the standards within a preset planning period as inspection targets. It generates candidate inspection paths on the inspection business map by comprehensively considering access time costs, operation time costs, operation mode switching costs, and time window constraints. These paths are then combined into inspection task packages, and resource scheduling is performed based on the skill information and available time information of the inspection teams. The recommended inspection paths are then output to suit each team.

[0025] Furthermore, as a preferred implementation, the inspection unit can be specifically implemented as an embedded sensor node installed on the surface of the riser, used to collect real-time data on the facade position and center height; the evaluation unit can be specifically implemented as a risk assessment engine deployed on a cloud server, which processes information in the historical defect database through a weighted algorithm; the processing unit can be specifically implemented as a modeling system based on the Neo4j graph database, used to store the topology and cost parameters of access connection edges, operation connection edges, and operation mode switching connection edges; the recommendation unit can be specifically implemented as a scheduling module integrating a heuristic search algorithm, which dynamically optimizes the path generation process based on the Dijkstra algorithm framework.

[0026] Therefore, by deeply integrating the axial distribution characteristics of external risers, the dynamic risk assessment mechanism, and the constraints of switching between multiple work modes, this system solves the problem of low path planning efficiency caused by the lack of refined data models in traditional manual experience-based methods for inspecting building facade external risers. This achieves a comprehensive improvement in risk coverage and a reduction in safety hazards associated with high-altitude operations. Specifically, the construction of the inspection business map fully depicts the relationship between internal building access paths and external work points, avoiding the shortcomings of existing technologies that only focus on planar sites while neglecting facade work transitions. The comprehensive application of inspection unit time window constraints and risk levels ensures that high-risk areas are prioritized for inclusion in the planning, overcoming the limitation that fixed-cycle inspections cannot respond to actual risk changes. The quantitative introduction of work mode switching costs ensures that path recommendations comply with safety regulations, transforming inspection path planning from extensive experience-based management to intelligent scheduling based on multi-dimensional constraints, ultimately improving the overall inspection execution efficiency and safety.

[0027] In some of the embodiments described above in this application, inspection units are proposed to record the facade location, center height, and inspection operation method. However, in its implementation, the external riser, as a continuous axial structure, lacks a standardized mechanism for how to scientifically divide the inspection units and accurately determine the operation method. This leads to reliance on manual experience for unit division and operation method selection in practical applications, which can easily result in problems such as misalignment between unit boundaries and building floors, and mismatch between operation method and location parameters. Consequently, it affects the accuracy of risk assessment and the feasibility of inspection path planning.

[0028] In response, this application further proposes that each external riser be equipped with multiple inspection units along its axial direction, and that each inspection unit, when recording the facade position, center height, and inspection operation method, includes: Collect external riser layout data, and divide each external riser into multiple inspection sections along the axis according to the starting height and ending height of each external riser, based on the floor boundaries and preset height intervals, and set each inspection section as an inspection unit. For each inspection unit, the orientation of the facade, the center height of the inspection unit, and the horizontal distance and height difference between the inspection unit and the work position on the balcony, windowsill, roof or ground are determined according to the pre-stored inspection operation method selection rules, and the inspection operation method, the facade position and center height of the inspection unit are recorded in the inspection unit.

[0029] Specifically, the division of external risers is based on the floor boundaries and preset height intervals. This can be achieved by combining the structural boundary points of the building floors with engineering experience values. The purpose is to ensure that the inspection unit is aligned with the natural boundary logic of the building and that the size is reasonable, so as to avoid the risk assessment granularity being too coarse due to the unit being too large or the path planning complexity being increased due to the unit being too small. The inspection operation mode selection rule can be understood as a decision logic based on location parameters. It can be implemented using a preset threshold judgment table or an expert system rule base. The purpose is to automatically match operation modes such as walking and suspended baskets based on objective parameters such as facade orientation and height difference, thereby reducing subjective errors in manual judgment.

[0030] Specifically, the proposed solution first collects data on the layout of external risers to obtain physical scope information. Then, based on floor boundaries and preset height intervals, it divides the area axially into inspection sections, standardizing these sections into inspection units as the basic unit of system operation. Subsequently, it analyzes the facade orientation, center height, and spatial relationship with the work location for each unit, automatically determining the appropriate work method based on pre-stored rules. Finally, it records the structured data in the inspection unit, thus establishing a standardized generation mechanism for external riser inspection units. This mechanism ensures that the division process strictly corresponds to the actual spatial distribution of the external risers, accurately aligns the unit boundaries with the building floors, and achieves automated matching of work methods based on objective parameters, providing a reliable data foundation for risk assessment and path planning.

[0031] As a specific embodiment, the solution of this application is implemented as follows: For the water supply riser of a residential building, its layout data from the ground to the roof is collected. According to the starting height and ending height, the riser is divided into multiple sections according to the floor slab position and the conventional height interval. For a certain section, its orientation, center height, distance and height difference from the balcony are analyzed. Based on the stored rules, the rope operation method is determined and the relevant information is recorded in the corresponding inspection unit.

[0032] By employing the above-mentioned solution, this application avoids the misalignment between the inspection unit boundary and the building floor, ensures the precise matching of the operation method and location parameters, and improves the accuracy of risk assessment data and the feasibility of inspection path planning.

[0033] In some of the embodiments described above in this application, an assessment unit is proposed to conduct risk assessment of the inspection unit. However, in its implementation, the risk assessment is based only on rough historical defect information, riser type and environmental exposure conditions, and fails to fully consider dynamic factors such as the frequency, severity and maintenance interval of historical defects, as well as the detailed differences in environmental exposure such as facade orientation and wind direction. This results in inaccurate risk level assessment, affecting the rationality and safety of the inspection plan.

[0034] In this regard, this application further proposes that when the assessment unit performs a risk assessment on each of the inspection units, it includes: The assessment unit statistically analyzes the number of historical defects, the severity of historical defects, and the time since the last maintenance for each inspection unit, and combines this with the riser type, connector type, and service life of the inspection unit to form basic risk information. The basic risk information is weighted and analyzed in conjunction with the orientation of the facade where the inspection unit is located, the degree of orientation towards the prevailing wind direction, the degree of street proximity, and the environmental exposure conditions to obtain the risk level of the inspection unit.

[0035] Among them, the historical defect count refers to the total number of defects that have occurred in the inspection unit throughout history, as recorded in the inspection system. This can be obtained by querying historical work order records or defect report documents in the database, with the aim of quantifying the frequency of failures in this area. The historical defect severity can be understood as a graded evaluation of the impact of historical defects, specifically using a three-level classification method (minor, moderate, severe) or a quantitative score based on maintenance costs, with the aim of distinguishing the contribution of different defects to safety risks. The time since the last maintenance is handled refers to the time interval from the completion of the most recent maintenance to the current moment, which can be calculated using the system timestamp, with the aim of reflecting the current maintenance status. Riser type refers to the material and use classification of the external riser, such as galvanized steel pipe, PVC plastic pipe, or stainless steel pipe, with the aim of identifying the material's durability and corrosion characteristics. Connector type refers to the type of connecting components such as brackets and clamps that fix the external riser, specifically metal clamps, plastic brackets, or welded connections, with the aim of reflecting the reliability of the structural connection. Service life refers to the continuous use time of the external riser since its installation, which can be calculated from the completion date, with the aim of assessing the degree of material aging. Facade orientation refers to the specific location and direction of the external riser on the building facade, such as the four basic directions of east, south, west, and north, or a further subdivided into eight directions. Its purpose is to differentiate between sunlight and wind erosion conditions. Orientation to the prevailing wind direction refers to the angle between the facade and the local prevailing wind direction, specifically expressed as an angle range of 0-90 degrees. Its purpose is to quantify the impact of wind loads. Street proximity refers to the distance of the external riser from the street, which can be understood as high, medium, and low levels. Its purpose is to assess the level of pollutant exposure. Environmental exposure conditions refer to external environmental factors affecting the corrosion of the external riser, such as humidity, salt spray concentration, and acid rain frequency. Its purpose is to comprehensively reflect the intensity of environmental erosion. Weighted analysis is a calculation method that assigns different weight coefficients to each factor based on its actual impact on risk. Specifically, it can use a linear weighted model or the analytic hierarchy process (AHP). Its purpose is to make the risk assessment results more consistent with reality.

[0036] Specifically, this application's solution first integrates historical defect dynamic data and inherent equipment attributes to form basic risk information, then incorporates environmental exposure details for weighted optimization, thereby improving the scientific rigor and accuracy of risk level assessment. The assessment unit statistically analyzes the number of historical defects, their severity, and the time elapsed since the last maintenance. This process captures the dynamic characteristics of risk evolution over time. The number of historical defects reflects the probability of failure, the severity reflects the potential hazard, and the time elapsed since the last maintenance indicates the current maintenance status. These three factors combine to construct a dynamic risk assessment framework. Basic risk information is formed by combining the riser type, connector type, and service life of the inspection unit, ensuring that basic risk encompasses both inherent equipment characteristics and subsequent usage conditions. The basic risk information is then weighted and analyzed in conjunction with facade orientation, orientation towards the prevailing wind direction, street proximity, and environmental exposure conditions. By incorporating environmental factors into the basic risk according to their actual impact weight, the risk level can differentiate the risk differences between different facades of the same building; for example, the windward side has a higher risk than the leeward side, and the street-facing side has a more prominent risk than the courtyard side, thus achieving spatial refinement of risk assessment.

[0037] As a specific implementation method, the solution of this application is implemented as follows: For the external riser inspection unit on the south facade of a high-rise residential building, the assessment unit first extracts multiple defect records that occurred in the past three years from the historical database, including several minor leaks and moderate corrosion events, with the most recent maintenance occurring several months ago; combined with the type of PVC drainage riser, the type of plastic clamp connector, and the relatively long service life of the unit, basic risk information is calculated; then, the basic risk information is weighted and analyzed with the south facade orientation, the degree of orientation to the prevailing wind direction, the degree of high street proximity, and the high salt fog environmental exposure conditions of the coastal city, with environmental factors allocated according to preset weight coefficients, and finally the risk level of the inspection unit is obtained as high risk level.

[0038] Through the above technical solution, this application achieves more precise risk level assessment, can distinguish subtle risk differences among different inspection units, avoids the one-sidedness of relying solely on static historical data, enables high-risk areas to receive inspection resources first, and improves the rationality and safety of inspection plans.

[0039] Specifically, in some of the embodiments described above in this application, risk assessment is proposed to determine the risk level. However, in its implementation, the conversion from risk level to inspection cycle lacks a dynamic adjustment mechanism, and the inspection plan cannot be optimized in a timely manner based on the actual inspection history and defect recurrence. This may result in insufficient inspection in high-risk areas due to rigid cycle settings, or waste of resources in low-risk areas due to excessive inspection.

[0040] In this regard, this application further proposes that when the assessment unit determines the inspection cycle and the latest allowed inspection date based on the risk level of the inspection unit, it includes: The evaluation unit selects the corresponding basic inspection cycle and basic grace period based on the risk level of the inspection unit, and shortens or extends the basic inspection cycle and basic grace period by combining the last actual inspection date and the most recent defect handling date of the inspection unit, thereby obtaining the inspection cycle and the latest allowed inspection date of the inspection unit, and forming a time window constraint for the inspection unit based on the inspection cycle and the latest allowed inspection date; wherein, when defects are found multiple times in the same inspection unit within a continuous planning cycle, the evaluation unit increases the inspection unit risk level of the inspection unit and shortens the inspection cycle of the inspection unit accordingly.

[0041] The basic inspection cycle refers to the standard inspection interval pre-set according to the risk level. It can be implemented using a fixed value or range derived from historical defect data statistics. The purpose is to transform the qualitative results of risk assessment into quantifiable initial cycle parameters, avoiding the problem of coarse cycle setting caused by simple mapping. The basic grace period refers to the buffer time allowed to be delayed based on the basic inspection cycle. It can be implemented using a fixed number of days or a proportional coefficient, aiming to address temporary scheduling difficulties and provide flexibility. The last actual inspection date refers to the specific date on which the inspection unit most recently completed an inspection. It can be implemented using a timestamp automatically recorded by the system, aiming to reflect the starting point of risk accumulation and the trend of facility status changes. The most recent defect... The defect handling date refers to the date on which the inspection unit most recently discovered and handled a defect. This can be achieved using time information from maintenance records, and its purpose is to indicate the level of risk activity and the frequency of defect recurrence. Shortening or extending the basic inspection cycle and basic grace period refers to dynamically adjusting cycle parameters based on historical dates. This can be achieved using a linear calculation model or rule engine, and its purpose is to ensure that the cycle settings closely match the actual changes in the facility's status. The mechanism for increasing the risk level when defects are found multiple times in the same inspection unit within a continuous planning cycle refers to identifying defect recurrence patterns to trigger a risk level escalation. This can be achieved using a threshold trigger algorithm or a weighted calculation model, and its purpose is to quickly respond to risk deterioration and break the defect accumulation chain.

[0042] Specifically, the proposed solution first selects the corresponding basic inspection cycle and basic grace period based on the risk level of the inspection unit through an assessment unit, transforming the risk assessment results into an initial cycle framework. Then, combining the previous actual inspection date and the most recent defect handling date, historical timestamps are used to capture risk evolution trends. For example, when the defect handling date is close to the current time, it reflects a recent surge in risk, and the cycle is proactively shortened to strengthen coverage. Conversely, when the last inspection date is far in the past, the grace period is extended to avoid plan interruptions. This dynamic adjustment process, based on a feedback mechanism of historical data, achieves flexible correction of cycle parameters, preventing insufficient coverage in high-risk areas due to fixed cycles while avoiding resource redundancy in low-risk areas. Finally, the adjustment results are solidified into inspection unit time window constraints, providing rigid time boundaries for path planning. In particular, when defects are found multiple times in the same inspection unit within a continuous planning cycle, the risk level jumps in real time and the inspection cycle is shortened simultaneously, ensuring that cycle adjustments keep pace with the rate of risk deterioration and that inspection activities are completed within a controllable risk window.

[0043] As a preferred embodiment, the solution of this application is implemented as follows: The evaluation unit selects a shorter basic inspection cycle and a smaller basic grace period for inspection units with high risk levels; considering the recent actual inspection date and the recent defect handling date, the basic parameters are shortened to obtain a more compact inspection cycle; when the same inspection unit is found to have defects multiple times within a continuous planning cycle, the risk level is increased and the inspection cycle is further shortened; the evaluation unit can dynamically retrieve the cycle parameter range based on the risk level table and automatically calculate the adjustment amount through the timestamp comparison module.

[0044] Through the above scheme, this application achieves adaptive adjustment of the inspection cycle, bridging the gap between risk assessment results and actual inspection needs. It prevents insufficient inspection coverage in high-risk areas due to rigid cycle settings, and avoids resource waste caused by excessive inspections in low-risk areas, thereby improving the accuracy of inspection resource allocation and the efficiency of risk control.

[0045] In some of the embodiments described above in this application, a processing unit is proposed to construct an inspection service map. However, in its implementation, the determination of access nodes and work locations lacks specific rules based on building data. The establishment of connecting edges does not clearly distinguish the conditions for access, work, and work mode switching. In particular, it ignores the safety constraints when switching work modes, which makes the service map unable to accurately reflect the actual building structure and the safety requirements for high-altitude operations, and may generate infeasible or dangerous inspection paths.

[0046] In this regard, this application further proposes that the processing unit, when constructing the inspection business diagram, includes: Based on the building floor plan data and floor access data, the access locations within the building are determined at stairwell entrances, elevator lobbies, corridor turning points, roof entrances and exits, and ground passage nodes, and these access locations within the building are used as access nodes. Based on the facade location and center height of the inspection unit, determine the work location at the balcony, window sill, roof edge and ground work point, and associate each work location with at least one access node. When there are direct staircases, elevators, or corridors connecting the processing unit within the building, a passage connection edge is established between the corresponding access nodes. When personnel survey or the work platform can reach the corresponding work position, a work connection edge is established between the access node and the work position. Only when the inspection work methods corresponding to adjacent work positions are different and the pre-set safety conditions for switching work methods are met, a work method switching connection edge is established between adjacent work positions.

[0047] In practical applications, building plan data and floor access data refer to digital information describing the internal spatial layout and access paths of a building. This can be achieved using Building Information Modeling (BIM) datasets, Computer-Aided Design (CAD) drawings, or on-site 3D scanning data. The purpose is to provide objective and accurate structural basis for determining access nodes. Access locations refer to inherent key access node sites within the building, such as stairwell entrances, elevator lobbies, corridor turning points, roof entrances / exits, and ground-level passageways. Their purpose is to identify essential path nodes for personnel moving within the building. Specifically, access nodes refer to access locations set as network topology nodes in the inspection business diagram. This can be achieved by coordinateizing access locations and assigning them unique logical identifiers, aiming to construct a network framework reflecting the actual access paths within the building. In practical applications, work locations refer to the specific operational sites used for performing external riser inspections. These can be building elements such as balconies, windowsills, roof edges, and ground-level work points, aiming to provide physical work sites that match the spatial characteristics of the inspection unit. The association between work locations and access nodes refers to establishing… The logical connection between the work location and access nodes can be achieved by recording the reachable path information from the work location to the nearest access node or by a direct topological connection. Its purpose is to ensure a smooth transition of the inspection path from the access network to the work point. Specifically, access connection edges refer to network connection lines representing access relationships within a building. They can be established between corresponding access nodes when there are physical connections such as direct staircases, elevators, or corridors within the building, aiming to accurately map the access paths of physical building connections. In practical applications, work connection edges refer to network connection lines representing the work relationship between the inspection unit and the work location. They can be established between access nodes and the work location when personnel surveying or the work platform can safely reach the corresponding work location, aiming to ensure the actual accessibility of the work point. Among these, work mode switching connection edges refer to network connection lines representing the switching relationship between different inspection work modes. They are established only between adjacent work locations when the inspection work modes corresponding to adjacent work locations are different and the pre-set work mode switching safety conditions are met, aiming to prevent work mode switching under unsafe conditions and ensure the safety of high-altitude operations.

[0048] Specifically, the processing unit first identifies key access points inherent in the building based on building plan data and floor access data, designating locations such as stairwell entrances and elevator lobbies as access nodes, thus constructing a basic topological framework reflecting the building's internal access network. Subsequently, based on the inspection unit's facade location and center height, corresponding work positions are dynamically generated at building elements such as balconies and window sills. A logical connection is established between each work position and at least one access node through an association mechanism, ensuring seamless integration between work points and the access network. Furthermore, the processing unit strictly adheres to the objective conditions of the building's physical connections, establishing access connection edges between access nodes that have direct access to staircases, elevators, or corridors, avoiding the creation of fictitious and infeasible access paths. Simultaneously, based on the actual accessibility constraints of personnel or work platforms, work connection edges are established between access nodes and work positions, ensuring the authenticity and effectiveness of work point connections. In particular, work mode switching connection edges are only established when there are differences in inspection operation methods between adjacent work positions and safety conditions are met, strictly controlling switching behavior through safety condition constraints to eliminate potential risks. This hierarchical and conditional mechanism for constructing connecting edges enables the inspection business map to accurately map the actual structural characteristics of the building and the safety regulations for high-altitude operations.

[0049] As a preferred embodiment, the solution of this application is implemented as follows: For a 12-story residential building, the processing unit extracts plan data and floor access data from the building BIM model, identifies the stair entrances, elevator lobbies, corridor turning points, roof entrances and ground passage nodes on each floor as access nodes; based on the facade position and center height of the external riser inspection unit, the operation position is determined at the balcony and window sill positions on each floor, and each operation position is associated with the nearest corridor turning point access node; when establishing connection edges, the processing unit establishes access connection edges between stair entrances on adjacent floors to represent stair access paths, and establishes operation connection edges between corridor turning points and balcony operation positions to represent accessible paths for personnel, and only when the operation position on the balcony of an adjacent floor changes from walking operation to suspended platform operation and meets the safety distance and wind speed conditions, does the operation mode switching connection edge establish between adjacent balcony operation positions.

[0050] Through the above solution, this application enables the inspection operation map to accurately reflect the actual building structure and high-altitude operation safety requirements, avoids the generation of infeasible or dangerous inspection paths due to inaccurate operation maps, and improves the safety and feasibility of inspection path planning.

[0051] In some of the embodiments described above in this application, the processing unit is proposed to set the passage time cost on the passage connection edge. However, in its implementation, the calculation of the passage time cost only relies on simple distance or empirical estimation, and does not fully incorporate key variables of passage within the building such as floor height difference, horizontal distance, passing medium type and security access control. This leads to the passage time estimation being out of touch with the actual operation scenario, which in turn distorts the passage time cost in the inspection path planning, affecting the feasibility of path recommendation and the efficiency of resource scheduling.

[0052] In this regard, this application further proposes that when the processing unit sets the passage time cost on the passage connection edge, it includes: Based on the floor height difference, horizontal distance, medium type, and whether security access is required at the two ends of the access connection edge, combined with preset walking speed, stair up and down speed, elevator running speed, and average elevator waiting time, the estimated travel time of personnel on the access connection edge is calculated, and the estimated travel time is recorded as the travel time cost in the corresponding access connection edge.

[0053] In practical applications, floor height difference refers to the vertical distance difference between access nodes, which can be achieved by using the standard floor height of adjacent floors or the cumulative height difference across floors, aiming to quantify the time consumption during vertical movement. Horizontal distance can be understood as the straight-line or path length of planar movement between access nodes, specifically obtained by measuring actual paths such as corridors and stairwells in the building floor plan, aiming to reflect the impact of planar displacement on travel time. Medium type specifically refers to the physical carrier on which the passage path relies, which can be classified by different passage medium types such as stairs, elevators, or corridors, aiming to associate different mediums with corresponding passage efficiency benchmarks. Whether security access control is required refers to whether there is an identity verification step in the passage path, specifically based on access control system configuration information or on-site survey records. The purpose of this is to incorporate the additional waiting time caused by access control verification; the preset walking speed refers to the standard speed at which people walk on flat ground, which can be set using industry-recommended values ​​or historical measured data, with the aim of standardizing the calculation of planar movement time; the stair speed refers to the average speed at which people move on a staircase, and different parameters can be set for uphill and downhill speeds, with the aim of accurately reflecting the speed differences in vertical movement; the elevator operating speed refers to the vertical operating speed of the elevator car, which can be configured based on the elevator equipment's technical parameters or building type, with the aim of calculating the elevator's vertical movement time; the average elevator waiting time refers to the average time from calling the elevator to the elevator arriving, which can be dynamically adjusted according to the building's usage time, with the aim of incorporating the impact of peak-hour queuing factors on travel time.

[0054] This application's solution integrates building physical characteristics and pedestrian traffic behavior parameters to transform static spatial parameters into dynamic time estimates. First, it obtains the floor height difference and horizontal distance between traffic nodes as basic spatial constraints. Second, it determines the traffic efficiency benchmark based on the medium type: when the medium is a staircase, it uses the staircase's vertical speed parameter; when it's an elevator, it uses the elevator's running speed parameter. Simultaneously, it checks the existence of security access control and adds corresponding waiting times. Finally, combining preset standardized parameters such as walking speed, it calculates vertical movement, horizontal movement, and waiting time in segments to form a complete traffic time cost. This multi-dimensional modeling mechanism covers the time-consuming details of the entire traffic process, dynamically coupling time estimates with actual operating conditions, avoiding biases from subjective experience-based estimates, and providing accurate time cost basis for inspection path optimization.

[0055] In one specific implementation, the processing unit of this application can be implemented using an industrial-grade ARM Cortex-M7 microcontroller. When calculating the passage connection edge connecting the stairwell entrance and the elevator lobby, the system obtains a 3.0-meter floor height difference, a 12-meter horizontal distance, a corridor as the medium type, and the presence of a security access control. Combining this with preset walking speeds of 1.0 meters per second, stairwell speeds of 0 (because the stairs are not used), elevator speeds of 0 (because the elevator is not used), and average elevator waiting time of 0, the system calculates a corridor passage time of 12 seconds, access control verification time of 45 seconds, and a total passage time of 57 seconds, which is recorded as the passage time cost in the passage connection edge. During this process, the microcontroller uses its built-in algorithm module to call the medium type parameters and access control configuration information in the building database to dynamically generate a passage time cost that conforms to the site conditions.

[0056] Through the above technical solutions, the calculation of travel time cost can accurately reflect the actual travel conditions inside the building, avoid path planning deviations caused by time estimation distortion, make the generated inspection path more in line with the actual on-site operation, and improve the feasibility and efficiency of resource scheduling.

[0057] In practical applications, some embodiments of this application propose that the processing unit set the operation time cost on the operation connection edge and the operation mode switching cost on the operation mode switching connection edge. However, in its implementation, since the operation time cost and operation mode switching cost are not accurately calculated, the inspection path recommendation cannot truly reflect the actual operation time consumption, which may lead to inspection task timeout or unreasonable resource scheduling.

[0058] In response, this application further proposes that when the processing unit sets the operation time cost on the operation connection edge and the operation mode switching cost on the operation mode switching connection edge, it includes: The processing unit determines the preparation time and observation time of the corresponding work connection edge based on the height of the inspection unit, the horizontal distance between the work position and the external riser, and the work specifications of the corresponding inspection work method. The sum of the preparation time and the observation time is recorded as the work time cost in the work connection edge. When switching from one inspection operation mode to another, the operation mode switching time is determined based on the basket deployment time, rope deployment time, drone take-off and landing preparation time, and safety inspection time. The operation mode switching time and the preset weighted time used to penalize the number of operation mode switching times are recorded as the operation mode switching cost in the operation mode switching connection edge.

[0059] The height of the inspection unit refers to its vertical coordinates, which can be obtained using Building Information Modeling (BIM) data or 3D laser scanning equipment. The purpose is to quantify the impact of height on operational safety preparation and difficulty. The horizontal distance between the work location and the external riser can be understood as the horizontal offset from the work point to the riser. This can be calculated through on-site surveying or computer-aided design models to assess movement efficiency and potential safety risks. The corresponding operational specifications for the inspection method refer to the operational requirements stipulated in industry standards or company procedures. These can be derived from national or industry standard documents to provide standardized time estimation criteria. Preparation time... Observation time refers to the time required for pre-operation safety checks and actual defect detection. It can be determined based on historical operation data or expert experience databases, with the aim of fully covering the time consumption of the entire operation process. Operation mode switching time refers to the equipment adjustment time required to switch from one inspection operation mode to another. It can include the specific operation time such as basket structure installation, rope anchor point fixing, or UAV system calibration, with the aim of accurately quantifying the time cost of the switching process. Preset weighted time refers to the penalty coefficient for frequent switching behavior. It can be set to a fixed time value or dynamically adjusted according to the switching frequency, with the aim of suppressing unnecessary changes in operation mode and reducing safety risks and time waste.

[0060] Specifically, the solution in this application uses a processing unit to perform refined modeling of the time costs of the operation connection edges and the operation mode switching connection edges, achieving accurate quantification of the inspection path time consumption. Based on the height of the inspection unit, the horizontal distance between the operation location and the external riser, and the corresponding operation specifications for the inspection operation mode, the processing unit dynamically calculates preparation and observation times to ensure that the operation time cost reflects the actual operating conditions. Simultaneously, for operation mode switching, the processing unit integrates the time for basket deployment, rope deployment, UAV take-off and landing preparation, and safety inspection to form an objective switching time benchmark, and introduces a preset weighted time as a penalty mechanism to suppress frequent switching behavior. This combination of a dual time cost model and the inspection business map enables path planning to realistically simulate the actual operation process, avoiding scheduling problems caused by time estimation errors.

[0061] As a specific implementation method, the solution of this application is implemented as follows: The processing unit is implemented using a microcontroller based on the ARM Cortex-M7 core. It connects to the building information management system via an Ethernet interface to acquire the center height and horizontal distance data of the inspection unit in real time, and calls the pre-stored operation specification database to determine the preparation time and observation time. In the operation mode switching scenario, when switching from rope operation to drone operation, the processing unit automatically calculates the rope dismantling time, drone assembly time, and safety inspection time, and combines them with the preset weighted time to generate the operation mode switching cost.

[0062] Through the above scheme, this application enables the inspection route recommendation to accurately reflect the actual operation time consumption, avoid the problems of inspection tasks exceeding the time limit and unreasonable resource scheduling, and improve the efficiency and safety of external riser inspection.

[0063] In the traditional planning process for external pipe inspection tasks, simple screening is carried out based solely on the expiration of time window constraints or the achievement of preset risk levels. The remaining time is not dynamically quantified and linked to the risk level, which may result in high-risk inspection units being delayed due to the long remaining time, or low-risk units being prioritized for resource allocation because the time window is approaching. This leads to confusion in the priority of inspection tasks, an increased risk of missing key defects, and a waste of human resources.

[0064] In response, this application further proposes that when the recommended unit selects inspection units whose time window constraints have expired or whose risk levels have reached preset conditions as inspection units within a preset planning period, the following applies: The recommendation unit calculates the remaining time until the latest allowed inspection date based on the inspection unit time window constraint of each inspection unit. It then combines the remaining time with the risk level of the inspection unit according to the pre-stored urgency assessment rules to form an urgency value. Based on the urgency values, the units are sorted from high to low, and the inspection units with urgency values ​​that reach the preset urgency threshold are selected as the units to be inspected.

[0065] In practical applications, remaining time refers to the time interval from the current date to the latest permitted inspection date, which can be expressed in days or hours. The purpose is to transform the abstract time window into a quantifiable urgency indicator. The urgency assessment rule can be understood as a mathematical model establishing the relationship between the time and risk dimensions. Specifically, it can be implemented using a linear weighted function or a non-linear piecewise function, aiming to eliminate the bias of a single indicator assessment through quantification rules. The urgency value is the priority quantification indicator output by the urgency assessment rule, which can be represented using a standardized scoring system, aiming to provide a unified priority comparison benchmark for different inspection units. In practical applications, the preset urgency threshold is the critical value for selecting units to be inspected. It can be set to a fixed value or dynamically adjusted according to system load, aiming to control the task scale and avoid resource overload.

[0066] Specifically, the solution in this application transforms the time window constraint of inspection units into a quantifiable remaining time indicator, and then combines this indicator with the risk level based on urgency assessment rules to form an urgency value, thereby achieving dynamic quantitative assessment of inspection task priority. First, the system calculates the remaining time for each inspection unit until the latest permitted inspection date, transforming the abstract time window into a concrete value. Second, the remaining time and risk level are input into a pre-stored urgency assessment rule for calculation. This rule establishes a quantitative relationship based on the principle that the shorter the remaining time and the higher the risk level, the greater the urgency value. Finally, the system sorts units from high to low urgency values ​​and selects units that reach a preset threshold, ensuring that high-risk units nearing their deadline are prioritized for coverage, while simultaneously controlling the task scale through the threshold. This dynamic quantitative mechanism allows the inspection plan to adaptively adjust priorities, avoiding resource allocation imbalances caused by single-indicator assessments.

[0067] As a specific implementation method, the recommendation unit performs the following operations within the preset planning cycle: First, obtain the latest allowed inspection date for all inspection units, and calculate the difference in the number of days between the current date and that date as the remaining time; then, call the urgency assessment rule stored in the database, which defines the urgency value as 0.7 × risk level coefficient + 0.3 × (1 - remaining time / basic inspection cycle), where the risk level coefficient is set as 1.0 for high risk, 0.6 for medium risk, and 0.3 for low risk; then, sort the calculated urgency values ​​from high to low, and select inspection units with urgency values ​​greater than 0.5 as the units to be inspected. For example, for an inspection unit with a high risk level and only 20% of the basic inspection cycle remaining, its urgency value is calculated as 0.7×1.0+0.3×(1-0.2)=0.94, which is much higher than the threshold of 0.5, and it will be selected first. On the other hand, for an inspection unit with a low risk level but only 10% of the cycle remaining, the urgency value is 0.7×0.3+0.3×(1-0.1)=0.48, which is lower than the threshold, and it will not be selected, thus ensuring that resources are prioritized for allocation to truly urgent high-risk units.

[0068] Through the above technical solutions, the system can dynamically and quantitatively assess the urgency of inspection tasks, giving priority to high-risk external riser inspection units that are close to the inspection deadline, avoiding the situation where critical defects are missed due to misjudgment of priority, while preventing low-risk units from occupying too many inspection resources, thus improving the scientific nature of inspection task allocation and resource utilization efficiency.

[0069] Traditional external pipe inspection route planning methods, when generating candidate routes, suffer from several drawbacks. They fail to integrate comprehensive evaluation mechanisms that consider inspection team work time constraints, multi-dimensional route costs, and dynamic penalty mechanisms related to time window constraints. This leads to route planning potentially exceeding the team's available time limits, ignoring the additional costs of switching work modes, or failing to respond promptly to the inspection timeliness requirements of high-risk units. Consequently, this results in low inspection task execution efficiency, resource waste, or safety hazards. To address this, this application further proposes a step-by-step approach for generating candidate inspection routes for recommended units, including: For each inspection team's starting point, the starting point is used as the starting node in the inspection business map. Under the premise of meeting the work duration constraints recorded in the available time information of the corresponding inspection team, the unit to be inspected is selected. The path is extended between the access connection edge, the operation connection edge, and the operation mode switching connection edge through a heuristic path search strategy. The sum of the access time cost, the operation time cost, and the operation mode switching cost is used as the path evaluation basis. At the same time, the path segment that violates the time window constraint of the inspection unit is penalized, and candidate inspection paths are generated.

[0070] In practical applications, heuristic path search strategies refer to an algorithmic mechanism that uses heuristic information to accelerate the path exploration process. This can be implemented using algorithms such as A*, simulated annealing, or tabu search. Its purpose is to efficiently address the path combination explosion problem caused by the complex structure of building facades, avoiding the waste of computational resources caused by exhaustive search. Path evaluation criteria can be understood as a comprehensive quantitative indicator of the total time consumption of a path. This can be achieved by using a linear weighted sum or normalized summation of travel time cost, operation time cost, and operation mode switching cost. Its purpose is to avoid the limitations of a single cost dimension and make path selection more aligned with actual inspection efficiency requirements. Specifically, the penalty mechanism refers to the rule of negatively adjusting path segments that violate the time window constraints of inspection units. This can be achieved by increasing virtual time costs, reducing path priority scores, or setting invalid path flags. Its purpose is to strengthen the constraint of timeliness violations and prioritize ensuring that the inspection windows of high-risk units are met.

[0071] Specifically, the recommendation unit first uses the starting point of the inspection team as the initial access node, and selects units to be inspected under the hard boundary condition of work time constraints. Then, through a heuristic path search strategy, it dynamically extends the path in a composite network composed of access connection edges, work connection edges, and work mode switching connection edges. The sum of access time cost, work time cost, and work mode switching cost is used as the core evaluation criterion. At the same time, a penalty mechanism is introduced for path segments that violate the time window constraint. Finally, candidate inspection paths that meet the requirements of timeliness and efficiency are generated. This process ensures the operability of the path by accurately matching the starting node with the actual work starting point, uses the work time constraint as a rigid boundary to prevent the task from exceeding the time limit, efficiently explores the combination path of multiple types of connection edges with heuristic search, and achieves a balance between timeliness and economy in path generation through the synergistic effect of comprehensive cost evaluation and time window penalty.

[0072] As a specific implementation method, the solution of this application is implemented as follows: For inspection teams starting from the stairwell entrance on the first floor of the building, the recommended unit takes the stairwell entrance as the starting access node. Based on the team's available working time, the units to be inspected with high urgency are selected. The A* algorithm is used to search for paths in the inspection business map. The heuristic function is dynamically adjusted based on the number of remaining units to be inspected and the spatial distance. During the path extension process, the sum of the time costs of the access connection edge, the work connection edge, and the work mode switching connection edge is calculated in real time as the evaluation basis. A preset penalty time adjustment is applied to the path segment that exceeds the latest allowed inspection date to generate candidate inspection paths that meet the constraints.

[0073] Through the above technical solution, this application integrates working time constraints, multi-dimensional cost evaluation and time window penalty mechanism to ensure that the inspection route planning strictly follows the available time limit of the team, comprehensively considers the total time consumption of personnel passage, operation execution and operation mode switching, and prioritizes the response to the inspection timeliness requirements of high-risk units, thereby improving the efficiency of inspection task execution and reducing resource conflicts and safety hazards.

[0074] In traditional external riser inspection management, when candidate inspection paths are combined into inspection task packages and resources are allocated to generate inspection path recommendation results, the matching of inspection team skill information with task requirements, the rigid constraints of available time, and the priority coverage requirements of high-risk inspection units are not fully considered. This may result in the recommendation results being unable to be executed due to skill mismatch, interruption of work due to time limits, or the existence of safety hazards due to the failure to deal with high-risk areas in a timely manner.

[0075] In response, this application further proposes that when the recommendation unit combines the candidate inspection paths into multiple inspection task packages and performs resource scheduling based on the skill information and available time information of the inspection teams to obtain the inspection path recommendation result corresponding to each inspection team, the following steps are taken: The set of units to be inspected corresponding to each candidate inspection path, the inspection operation mode type included in the candidate inspection path, the total travel time cost and the total operation time cost of the candidate inspection path are used as inspection task package attributes; the executable inspection operation mode recorded in the inspection team skill information is compared with the inspection operation mode required by the inspection task package. The matching process involves eliminating combinations where the inspection team's skill information does not meet the attributes of the inspection task package. Under the premise that the sum of the cumulative travel time cost and cumulative operation time cost for each inspection team does not exceed the available time recorded in the corresponding inspection team's available time information, and that high-risk inspection units are prioritized for coverage, the allocation scheme between each inspection task package and inspection team is comprehensively evaluated based on the number of high-risk units to be inspected, the number of violations of the inspection unit's time window constraint, and the cost of switching operation modes. The allocation scheme with the highest score is selected as the resource scheduling result, and the recommended inspection path result for each inspection team is obtained.

[0076] In practical applications, the inspection task package attribute refers to the technical feature that structurally represents the work content and constraints of candidate inspection paths. This can be achieved by combining the set of units to be inspected, the type of inspection operation, and time cost. The purpose is to provide a precise task description basis for resource scheduling and avoid allocation deviations due to ambiguous task information. Skill matching refers to the verification mechanism for the suitability of inspection team capabilities with task requirements. This can be achieved by comparing the set of inspection operation methods that the team can execute with the set of operation methods required by the task package. The purpose is to ensure that the assigned tasks strictly match the actual capabilities of the team and prevent work interruptions due to skill deficiencies. Specifically, time constraints refer to the mechanism for ensuring the time feasibility of inspection operations. This can be achieved by using a dynamic verification logic of the total accumulated time cost and the team's available time. The purpose is to ensure the executability of the recommended path in the time dimension. Furthermore, multi-dimensional evaluation refers to a decision-making mechanism that comprehensively considers key factors. This can be achieved by using a weighted scoring system covering the number of high-risk units, the number of time window defaults, and the cost of switching operation methods. The purpose is to balance risk control and operational efficiency.

[0077] This application's solution structures candidate inspection routes into task packages with clearly defined attributes. First, it matches and filters these task packages against the skills of the work teams, eliminating combinations with mismatched capabilities. Then, under the premise of meeting time constraints and prioritizing high-risk areas, it constructs a multi-dimensional evaluation system encompassing risk coverage, time compliance, and operational efficiency to score all feasible allocation schemes. Finally, it selects the scheme with the highest score as the resource allocation result. This mechanism ensures that the recommended inspection routes not only match the actual execution capabilities of the work teams but also prioritize high-risk areas within a limited timeframe, while simultaneously optimizing overall operational efficiency.

[0078] In one specific implementation, the processing unit can be a microcontroller based on an ARM Cortex-M7 core, which runs a resource scheduling algorithm. When the system receives candidate inspection path data, the processing unit uses the set of units to be inspected, the type of operation (e.g., including walking and drone operations), and the time cost as task package attributes for each path; simultaneously, it reads the skill information of team A (capable of performing walking and rope operations) and available time. The processing unit first matches and finds that the task package requires drone operation, which team A lacks the corresponding skill for, thus eliminating this combination; then, for team B (capable of performing all operation methods), it verifies that the cumulative time cost does not exceed the available time and calculates the score of this allocation scheme in terms of the number of high-risk units covered, the number of time window defaults, and the cost of switching operation methods; finally, it compares the scores of all feasible schemes and selects the optimal one as the recommended inspection path for team B.

[0079] Through the above technical solutions, this application solves the safety hazards in inspection route recommendation caused by skill mismatch leading to task inability to execute, time limit exceeding causing operation interruption, and high-risk areas not being dealt with in a timely manner, ensuring the feasibility of inspection task allocation and the priority of risk coverage.

[0080] In the above embodiments, by subdividing each external riser along the axial direction into multiple inspection units and recording the inspection operation methods for each unit, a detailed analysis of the high-altitude external riser inspection objects is achieved. The evaluation unit, combining historical defect information, riser type, and environmental exposure conditions, gives the risk level of the inspection unit and forms a time window constraint for the inspection unit, enabling high-risk and near-expiration inspection units to be identified and prioritized. The processing unit further constructs an inspection business map containing access connection edges, operation connection edges, and operation mode switching connection edges based on the access locations within the building and the inspection units, and respectively on the aforementioned connection edges. The system sets up access time costs, operation time costs, and operation mode switching costs, quantifying the time spent on in-building access, high-altitude operation time, and additional costs incurred by switching between different inspection operation modes into a single graphical model. The recommendation unit generates candidate inspection paths, and combines these candidate paths into multiple inspection task packages. Then, it uses the skill information and available time information of the inspection team to schedule resources, ensuring that high-risk and near-expiration inspection units are prioritized for timely coverage. This reduces the overall inspection mileage and the number of high-cost inspection operation mode switching times, improves the utilization rate of inspection team working hours, and enhances the safety of external riser inspections.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An intelligent recommendation system for external riser inspection routes, characterized in that, include: Inspection unit: Each external riser is provided with multiple inspection units along the axial direction, and each inspection unit records the facade position, center height and inspection operation method. The inspection operation method includes at least one of walking operation, suspended platform operation, rope operation and drone operation. The assessment unit is configured to perform a risk assessment on each of the inspection units based on historical defect information, riser type and environmental exposure conditions, obtain the risk level of the inspection unit, and determine the inspection cycle and the latest allowed inspection date based on the risk level of the inspection unit, thus forming a time window constraint for the inspection unit. The processing unit is configured to construct an inspection business map based on the access locations within the building and the inspection units. In the inspection business map, access relationships within the building are represented by access connection edges, operation relationship relationships between inspection units and corresponding operation locations are represented by operation mode switching connection edges, and switching relationships between different inspection operation modes are represented by operation mode switching connection edges. Access time cost is set on access connection edges, operation time cost is set on operation connection edges, and operation mode switching cost is set on operation mode switching connection edges. The recommendation unit is configured to select inspection units whose time window constraints expire or whose risk levels reach preset conditions within a preset planning period as inspection units to be inspected. On the inspection business map, candidate inspection paths are generated by comprehensively considering travel time cost, operation time cost, operation mode switching cost, and inspection unit time window constraints. The candidate inspection paths are then combined into multiple inspection task packages. Resource scheduling is performed by combining the skill information and available time information of the inspection team to obtain the inspection path recommendation result corresponding to each inspection team.

2. The intelligent recommendation system for external riser inspection routes according to claim 1, characterized in that, Each external riser is equipped with multiple inspection units along its axial direction, and each inspection unit records the facade position, center height, and inspection operation method, including: Collect external riser layout data, and divide each external riser into multiple inspection sections along the axis according to the starting height and ending height of each external riser, based on the floor boundaries and preset height intervals, and set each inspection section as an inspection unit. For each inspection unit, the facade orientation, center height, horizontal distance and height difference between the inspection unit and the work position on the balcony, windowsill, roof or ground are determined according to the pre-stored inspection operation mode selection rules, and the inspection operation mode, facade position and center height of the inspection unit are recorded in the inspection unit.

3. The intelligent recommendation system for external riser inspection routes according to claim 1, characterized in that, When the assessment unit performs a risk assessment on each of the inspection units, it includes: The assessment unit statistically analyzes the number of historical defects, the severity of historical defects, and the time since the last maintenance for each inspection unit, and combines this with the riser type, connector type, and service life of the inspection unit to form basic risk information. The basic risk information is weighted and analyzed in conjunction with the orientation of the facade where the inspection unit is located, the degree of orientation towards the prevailing wind direction, the degree of street proximity, and the environmental exposure conditions to obtain the risk level of the inspection unit.

4. The intelligent recommendation system for external riser inspection routes according to claim 3, characterized in that, When determining the inspection cycle and the latest permitted inspection date based on the risk level of the inspection unit, the assessment unit includes: The evaluation unit selects the corresponding basic inspection cycle and basic grace period based on the risk level of the inspection unit, and shortens or extends the basic inspection cycle and basic grace period by combining the last actual inspection date and the most recent defect handling date of the inspection unit, thereby obtaining the inspection cycle and the latest allowed inspection date of the inspection unit, and forming a time window constraint for the inspection unit based on the inspection cycle and the latest allowed inspection date; wherein, when defects are found multiple times in the same inspection unit within a continuous planning cycle, the evaluation unit increases the inspection unit risk level of the inspection unit and shortens the inspection cycle of the inspection unit accordingly.

5. The intelligent recommendation system for external riser inspection routes according to claim 1, characterized in that, The processing unit, when constructing the inspection service map, includes: Based on the building floor plan data and floor access data, the access locations within the building are determined at stairwell entrances, elevator lobbies, corridor turning points, roof entrances and exits, and ground passage nodes, and these access locations within the building are designated as access nodes. Based on the facade location and center height of the inspection unit, determine the work location at the balcony, window sill, roof edge and ground work point, and associate each work location with at least one access node. When there are direct staircases, elevators, or corridors connecting the buildings, the processing unit establishes a passage connection edge between the corresponding passage nodes. When personnel survey or the work platform can reach the corresponding work position, it establishes a work connection edge between the passage node and the work position. Only when the inspection work methods corresponding to adjacent work positions are different and the pre-set work mode switching safety conditions are met, a work mode switching connection edge is established between adjacent work positions.

6. The intelligent recommendation system for external riser inspection routes according to claim 5, characterized in that, When the processing unit sets the passage time cost on the passage connection edge, it includes: Based on the floor height difference, horizontal distance, medium type, and whether security access is required at the two ends of the access connection edge, combined with preset walking speed, stair up and down speed, elevator running speed, and average elevator waiting time, the estimated travel time of personnel on the access connection edge is calculated, and the estimated travel time is recorded as the travel time cost in the corresponding access connection edge.

7. The intelligent recommendation system for external riser inspection routes according to claim 6, characterized in that, When the processing unit sets the job time cost on the job connection edge and sets the job mode switching cost on the job mode switching connection edge, it includes: The processing unit determines the preparation time and observation time of the corresponding work connection edge based on the height of the inspection unit, the horizontal distance between the work position and the external riser, and the work specifications of the corresponding inspection work method, and records the sum of the preparation time and observation time as the work time cost in the work connection edge. When switching from one inspection operation mode to another, the operation mode switching time is determined based on the basket deployment time, rope deployment time, drone take-off and landing preparation time, and safety inspection time. The operation mode switching time and the preset weighted time used to penalize the number of operation mode switching times are recorded as the operation mode switching cost in the corresponding operation mode switching connection edge.

8. The intelligent recommendation system for external riser inspection routes according to claim 1, characterized in that, When the recommendation unit selects inspection units whose time window constraints have expired or whose risk levels have reached preset conditions as inspection units within a preset planning period, it includes: The recommendation unit calculates the remaining time until the latest allowed inspection date based on the inspection unit time window constraint of each inspection unit. It then performs a comprehensive analysis of the remaining time and the risk level of the inspection unit according to the pre-stored urgency assessment rules to form an urgency value. Based on the urgency value, the units are sorted from high to low, and the inspection units whose urgency value reaches the preset urgency threshold are selected as the units to be inspected.

9. The intelligent recommendation system for external riser inspection routes according to claim 8, characterized in that, When the recommendation unit generates candidate inspection paths, it includes: The recommendation unit, for each inspection team's starting point, takes the starting point as the starting node in the inspection business map, selects the unit to be inspected under the premise of satisfying the working time constraints recorded in the available time information of the corresponding inspection team, and extends the path between the access connection edge, the operation connection edge, and the operation mode switching connection edge through a heuristic path search strategy. The sum of the access time cost, the operation time cost, and the operation mode switching cost is used as the path evaluation basis. At the same time, a penalty is imposed on the path segment that violates the time window constraint of the inspection unit, and the candidate inspection path is generated.

10. The intelligent recommendation system for external riser inspection routes according to claim 9, characterized in that, When the recommendation unit combines the candidate inspection paths into multiple inspection task packages and performs resource scheduling based on the skill information and available time information of the inspection teams to obtain the inspection path recommendation result for each inspection team, it includes: The recommendation unit uses the set of units to be inspected corresponding to each candidate inspection path, the type of inspection operation mode included in the candidate inspection path, the total travel time cost and the total operation time cost of the candidate inspection path as the inspection task package attributes. It matches the executable inspection operation mode recorded in the inspection team skill information with the inspection operation mode required by the inspection task package and eliminates combinations in which the inspection team skill information does not meet the inspection task package attributes. Under the premise that the sum of the cumulative travel time cost and the cumulative operation time cost of each inspection team does not exceed the available time recorded in the available time information of the corresponding inspection team, and that high-risk inspection units are covered first, the allocation scheme between each inspection task package and the inspection team is comprehensively evaluated based on the number of high-risk units to be inspected, the number of violations of the inspection unit time window constraint, and the cost of switching operation mode. The allocation scheme with the highest score is selected as the resource scheduling result, and the inspection path recommendation result corresponding to each inspection team is obtained.

Images