Digital management method and system for main and distribution network engineering

Abstract

The embodiment of the invention discloses a digital management method and system for a main and distribution network project. The method comprises the following steps: performing project information structured modeling on a full life cycle of a target main and distribution network project to obtain a project stage node index model; chain type tracking analysis is carried out on the engineering stage nodes, an information interaction rule and a data transmission path are established, and a data flow topological graph is obtained; extracting front-end node state layer information and rear-end engineering monitoring layer information based on a front-end and rear-end separation strategy; and constructing a multi-dimensional influence factor association network by using the information to perform risk evolution path deduction and risk event simulation to obtain a simulation result and generate a cross-node cooperative response strategy. According to the embodiment of the invention, through the digital support of the node chain type tracking and the front and back end separation architecture, the data island of each stage of the main and distribution network project is eliminated, the cooperation efficiency of the project management is improved, and the controllability of the whole life cycle process is enhanced.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of digitization, in particular to a main distribution network engineering digitization management method and system. BACKGROUND

[0002] In the power system, the main distribution network engineering is the key infrastructure to ensure stable power supply. With the continuous growth of power demand and the increasing scale of power grid, the complexity and systematization of the main distribution network engineering are also continuously improving. At present, the management and risk assessment of the main distribution network engineering have been widely concerned. However, in the aspects of engineering information management and engineering risk assessment, the existing technology cannot comprehensively and systematically integrate and analyze the whole life cycle information of the main distribution network engineering, it is difficult to grasp the overall operation situation of the engineering, and it lacks overall and coordination in risk assessment and response strategy making, and cannot effectively cope with various risks in the whole life cycle of the engineering. SUMMARY

[0003] The embodiment of the present application provides a main distribution network engineering digitization management method, which is applied to a main distribution network engineering digitization management system. The method comprises: structuring modeling of whole life cycle information of a target main distribution network engineering to obtain an engineering stage node index model corresponding to the target main distribution network engineering, the engineering stage node index model comprising a plurality of engineering stage nodes corresponding to the target main distribution network engineering and a correlation constraint relationship between the engineering stage nodes; chain tracking analysis of the plurality of engineering stage nodes to obtain a data flow conversion topology graph of the target main distribution network engineering under the whole life cycle by establishing information interaction rules and data transmission paths between the engineering stage nodes, the data flow conversion topology graph being used to represent information input / output relationships and data processing logic of the engineering stage nodes; extracting front-end node state layer information and back-end engineering monitoring layer information of the data flow conversion topology graph based on a front-end and back-end separation strategy, the front-end node state layer information comprising real-time operation state parameters of the engineering stage nodes, and the back-end engineering monitoring layer information comprising engineering resource configuration data and environmental influence factor data; using the front-end node state layer information and the back-end engineering monitoring layer information to perform risk event simulation on the target main distribution network engineering by constructing a multi-dimensional influence factor correlation network to perform risk evolution path deduction, to obtain a risk event simulation result corresponding to the target main distribution network engineering, and generating a cross-node collaborative response strategy according to the risk event simulation result, the cross-node collaborative response strategy comprising a collaborative disposal process and a resource scheduling scheme of different engineering stage nodes.

[0004] The embodiment of the application provides a kind of main distribution network engineering digital management system, comprising: processor;Storage device, which is stored with computer program;Network interface, for providing network communication function;When the computer program is executed by the processor, so that the processor realizes any described main distribution network engineering digital management method.

[0005] The embodiment of the application provides a kind of readable storage medium, the program or instruction is stored on the readable storage medium, the program or instruction is executed by processor to realize the steps of the main distribution network engineering digital management method.

[0006] The embodiment of the application first constructs the engineering stage node index model containing engineering stage node and associated constraint relationship by structuring modeling of target main distribution network engineering full life cycle;On this basis, chain tracking analysis is carried out to obtain data flow conversion topological graph, the information input-output relationship and data processing logic of each engineering stage node are clarified, and the accurate grasp of engineering full life cycle data flow conversion is realized;Based on the front-end and back-end separation strategy, the front-end node state layer information and the back-end engineering monitoring layer information are extracted, the real-time running state, resource allocation and environmental influence and other aspects of data of the project can be comprehensively obtained;Using the above information, a multi-dimensional influence factor correlation network is constructed to carry out risk evolution path deduction and risk event simulation, which can accurately predict the risks that the project may face, and generate cross-node collaborative response strategy including collaborative disposal process and resource scheduling scheme, so as to improve the ability of target main distribution network engineering to cope with risks and ensure the stable and efficient operation of the project in the whole life cycle.

[0007] In this way, through the digital support of node chain tracking and front-end and back-end separation architecture, the data islands of each stage of main distribution network engineering are eliminated, the collaboration efficiency of engineering management is improved, and the controllability of the whole life cycle process is strengthened. BRIEF DESCRIPTION OF DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the application or the related art, the following will briefly introduce the drawings needed to be used in the embodiment or related art description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0009] Figure 1 The flow chart of the main distribution network engineering digital management method provided by the embodiment of the application; Figure 2 The basic structure of the main distribution network engineering digital management system provided by the embodiment of the application is shown in the schematic diagram; Figure 3A functional module block diagram of a main distribution network engineering digital management device provided by the embodiment of the application is provided. DETAILED DESCRIPTION

[0010] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work are within the protection scope of the application.

[0011] Please refer to Figure 1 , Figure 1 A flowchart of a main distribution network engineering digital management method provided by the embodiment of the application is provided. The method can be executed by a main distribution network engineering digital management system, or can be executed by the main distribution network engineering digital management system and a server together. The method can include steps 110-140.

[0012] The embodiment of the application discloses a main distribution network engineering digital management method. The method runs through the whole life cycle of a target main distribution network engineering. Through steps such as engineering information structured modeling, data flow topology construction, front and back end layer information extraction, risk event simulation, and cross-node collaborative response strategy generation, the digital and information management of the main distribution network engineering is realized.

[0013] Step 110: engineering information structured modeling is performed on the whole life cycle of a target main distribution network engineering, to obtain an engineering stage node index model corresponding to the target main distribution network engineering. The engineering stage node index model contains a plurality of engineering stage nodes corresponding to the target main distribution network engineering and the associated constraint relationship between the engineering stage nodes.

[0014] In the embodiment of the application, the target main distribution network engineering is a city regional power grid upgrading and reconstruction engineering. Its whole life cycle covers a planning and design stage, a material procurement stage, a construction stage, a commissioning and operation stage, and a maintenance and repair stage. When engineering information structured modeling is performed, the core business content and the key management elements of each stage are first sorted out. For example, the engineering stage nodes of the planning and design stage include a load prediction node, a network frame planning node, and a device selection node. The material procurement stage includes a supplier evaluation node, a procurement plan development node, and a material acceptance node. The construction stage includes a construction scheme development node, a site construction organization node, and a quality supervision node. The commissioning and operation stage includes a device commissioning node, a system integration node, and a trial operation evaluation node. The maintenance and repair stage includes a device state monitoring node, a fault diagnosis node, and a maintenance plan development node.

[0015] Each engineering stage node has a clear correlation constraint relationship, for example, the equipment parameter information output by the equipment selection node in the planning and design stage is the input condition of the supplier evaluation node and the purchase plan development node in the material procurement stage, and the construction scheme development node in the construction stage needs to be developed according to the network planning node result in the planning and design stage and the material inspection node information in the material procurement stage.

[0016] By storing these engineering stage nodes and their correlation constraint relationships in a structured data form, an engineering stage node index model is formed. Each engineering stage node in the model includes node number, node name, business attribute, input parameter list, output parameter list, pre-node list, and post-node list, etc. For example, the input parameter list of the equipment selection node includes load prediction data and network structure parameters, the output parameter list includes equipment model, technical parameters, and quantity demand, the pre-node list is the load prediction node and the network planning node, and the post-node list is the supplier evaluation node and the purchase plan development node.

[0017] Step 120: Chain tracking analysis is performed on the plurality of engineering stage nodes, information interaction rules and data transmission paths between the engineering stage nodes are established, and a data flow conversion topology graph of the target main and distribution network engineering in the whole life cycle is obtained. The data flow conversion topology graph is used to represent the information input and output relationship and data processing logic of each engineering stage node.

[0018] After obtaining the engineering stage node index model, in the embodiment of the present application, first, an engineering stage node correlation network is constructed based on the model, taking each engineering stage node as a network node, and the correlation constraint relationship between the nodes as a network edge, thereby forming a directed network structure. Through the engineering stage node correlation network, the pre-associated nodes and post-associated nodes of each engineering stage node can be identified, and the direct correlation relationship and indirect correlation relationship between the nodes can be determined. For example, the direct pre-associated node of the site construction organization node in the construction stage is the construction scheme development node, and the indirect pre-associated nodes include all nodes in the planning and design stage and all nodes in the material procurement stage, because the construction scheme development node depends on multiple output information of the planning and design and material procurement stages; the direct post-associated node is the quality supervision node, and the indirect post-associated nodes are the nodes in the debugging and running stage.

[0019] Step 121: Based on the engineering stage node index model, an engineering stage node correlation network is constructed, the pre-associated nodes and post-associated nodes of each engineering stage node are identified through the engineering stage node correlation network, and the direct correlation relationship and indirect correlation relationship between the nodes are determined.

[0020] In the embodiment of the present application, when constructing the node association network in the engineering stage, the adjacent matrix representation method in graph theory is adopted, the rows and columns of the matrix correspond to the node numbers of the engineering stage, and the matrix element value is 1, indicating that the node of the corresponding row is the preceding association node of the node of the corresponding column, and the value is 0, indicating that there is no preceding association relationship. By analyzing the reachability of the adjacent matrix, that is, calculating the transitive closure of the adjacent matrix, the indirect association relationship between the nodes can be determined, and the element value 1 in the transitive closure matrix indicates that the node of the corresponding row can indirectly affect the node of the corresponding column through a series of direct association relationships. For example, for a target main distribution network engineering containing n engineering stage nodes, an n*n adjacent matrix MA is constructed, where MA[i][j]=1 indicates that node i is a direct preceding association node of node j, and by calculating the transitive closure matrix T of MA, T[i][j]=1 indicates that node i is an indirect preceding association node of node j (when i and j have a direct association, it is also included). In this way, all preceding association nodes (direct and indirect) and following association nodes (direct and indirect) of each engineering stage node can be systematically identified.

[0021] Step 122: According to the business attributes and data characteristics of each engineering stage node, create information interaction rules between nodes, which include data format conversion requirements, information verification standards and interaction timing control parameters.

[0022] Due to different business attributes, the generated and required information data of each engineering stage node has different characteristics. For example, the device information data output by the equipment selection node in the planning and design stage is in XML format, containing structured fields such as device model, technical parameters and performance indicators; and the device information data required by the supplier evaluation node in the material procurement stage is in JSON format, and has specific requirements for the accuracy and completeness of the device technical parameters. Therefore, when creating information interaction rules between nodes, first, for data format conversion requirements, a mapping relationship table of XML and JSON formats is formulated to clearly specify the JSON field name and data type conversion method corresponding to each XML field, such as converting the "TechParam" field in XML to the "technical Parameters" field in JSON, and converting the data type from string type to numerical type. In terms of information verification standards, the supplier evaluation node sets verification rules for the received device technical parameters, such as requiring the device rated power parameter to be within a preset range interval, and requiring the device insulation level to meet a specific standard level requirement, and returning an error prompt information and specifying the specific verification failure item when the verification fails. The interaction timing control parameters specify the time sequence and time interval requirements of information interaction between nodes, for example, the procurement plan formulation node in the material procurement stage needs to complete the procurement plan formulation and deliver the results to the next node within a certain number of working days after receiving the information output by the equipment selection node, and if the time interval is exceeded, the system will automatically issue an overtime warning.

[0023] Step 123: According to the time sequence and logical relationship of project implementation, data flow transfer path tracking is performed on each project phase node in the project phase node association network, and the transmission direction, transmission content and transmission frequency of data between nodes are recorded.

[0024] In the embodiment of the application, the time sequence of project implementation follows the order of planning and design phase-material procurement phase-construction phase-commissioning phase-operation and maintenance phase, and the project phase nodes in each phase are executed in turn according to business logic. The data flow transfer path tracking starts from the load prediction node in the planning and design phase. The output data of this node is the load prediction report. The transmission direction is to the grid planning node. The transmission content includes the maximum load value of the predicted area, the load growth trend curve, the load distribution characteristics, etc. The transmission frequency is one-time transmission. After the grid planning node receives the load prediction data, it plans the power grid structure in combination with the regional geographic information, outputs the grid planning scheme, and transmits the direction to the equipment selection node. The transmission content includes substation site selection, line direction, capacity configuration, etc. The transmission frequency is one-time transmission. The equipment selection node selects equipment according to the grid planning scheme and load prediction data, outputs the equipment list and parameters, and transmits them to the supplier evaluation node and the procurement plan development node respectively. The transmission content is focused according to the needs of the receiving node. The transmission frequency is completed within h hours after the completion of equipment selection. In this way, the data flow transfer path of each project phase node is tracked and recorded in detail to form a complete data transmission record between nodes.

[0025] Step 124: Based on the data flow transfer path and the information interaction rules between nodes obtained by tracking, a directed graph structure containing node connection relationship, data flow direction identification and interaction rule attribute is constructed, and the directed graph structure is determined as the data flow transfer topology graph of the target main and distribution network project in the whole life cycle.

[0026] After the data flow path tracking and information interaction rule creation are completed, in the embodiment of the application, a directed graph structure is used to construct a data flow topology graph. The nodes of the directed graph represent engineering stage nodes, and the node connection relationship is represented by a directed edge. The direction of the directed edge is the data flow direction identifier, pointing from the information output node to the information input node. The directed edge is also attached with an interaction rule attribute, including the data format conversion requirement index corresponding to the edge, the information verification standard number, and the interaction timing control parameter value. For example, on the directed edge from the equipment selection node to the supplier evaluation node, the attribute information includes the data format conversion requirement index "FMT-001" (corresponding to the mapping relationship table of XML to JSON), the information verification standard number "VLD-005" (corresponding to the equipment parameter verification rule of the supplier evaluation node), and the interaction timing control parameter value "D=3" (indicating that the interaction is completed within 3 working days). In this way, the constructed directed graph structure clearly shows the information input and output relationship and data processing logic of each engineering stage node in the whole life cycle of the target main and distribution network engineering, that is, the data flow topology graph.

[0027] Step 130: Extracting the front-end node state layer information and the back-end engineering monitoring layer information of the data flow topology graph based on the front-end and back-end separation strategy. The front-end node state layer information includes real-time running state parameters of each engineering stage node, and the back-end engineering monitoring layer information includes engineering resource configuration data and environmental influence factor data.

[0028] In order to realize efficient management and analysis of the data flow topology graph, in the embodiment of the application, the front-end and back-end separation strategy is used to extract the layers. The front-end node state layer mainly focuses on the running state of the engineering stage node itself, and the back-end engineering monitoring layer focuses on the resource configuration and external environmental factors supporting the engineering operation.

[0029] Step 131: Layer separation processing is performed on the data flow topology graph, and the topology elements related to the running state of the engineering stage node are divided into the front-end node state layer, and the topology elements related to the engineering resource configuration and environmental monitoring are divided into the back-end engineering monitoring layer.

[0030] The topology elements in the data flow transformation topology diagram include node elements, edge elements, and attribute elements. The topology elements related to the running state of the engineering stage nodes include the running state identifier of the node element (such as the state markers of normal running, abnormal early warning, and suspended execution), the node processing progress parameter (such as the completed workload proportion and the remaining processing time estimation), and the node performance index data (such as the node data processing rate and the data storage capacity utilization rate), which are divided into the front-end node state layer. For example, the running state identifier of the site construction organization node in the construction stage is “normal running”, the processing progress parameter is “65%”, and the data processing rate is “v pieces / second”, which are all topology elements in the front-end node state layer. The topology elements related to the engineering resource allocation include the resource node element (such as the allocation nodes of manpower, materials, and funds), the resource allocation amount parameter, and the resource use efficiency index; the topology elements related to the environment monitoring include the environment node element (such as the environment parameter collection nodes of temperature, humidity, and wind speed), the environment parameter value, and the environment impact level assessment result, which are divided into the back-end engineering monitoring layer. For example, the resource allocation amount parameter of the material resource allocation node is “steel: w tons, cement: x bags”, and the environment parameter value of the construction site temperature collection node is “28°C”, which are all topology elements in the back-end engineering monitoring layer.

[0031] Step 132: Attribute analysis is performed on the topology elements in the front-end node state layer to extract the state identifier information, running progress parameter, and abnormal early warning threshold of each engineering stage node, and the front-end node state layer information containing the node state parameter and state change time sequence is generated.

[0032] Attribute analysis is performed on each engineering stage node element in the front-end node state layer. Taking the device debugging node in the debugging running stage as an example, the state identifier information includes enumerated values such as “debugging”, “debugging passed”, and “debugging failed”; the running progress parameter includes the percentage of the number of devices currently debugged to the total number of devices “G%” and the average device debugging time “M minutes per device”; and the abnormal early warning threshold includes the voltage fluctuation range of the device during debugging “[U1, U2]” and the current peak limit “I max ”. By analyzing these attributes, a node state parameter set is generated for each engineering stage node, such as the node state parameter set of the device debugging node {state identifier: debugging, progress percentage: G%, average time consumption: M minutes per device, voltage fluctuation range: [U1, U2], current peak limit: I maxMeanwhile, the changes of the state parameters of each node are recorded in time sequence to form a time sequence of state changes, for example, the progress percentage of the equipment debugging node at T1 is "30%", at T2 is "45%", and at T3 is "65%", and the parameter values at these time points constitute the time sequence of state changes of the node. The node state parameters and the time sequence of state changes of all the nodes in the engineering stages are summarized to generate the front-end node state layer information.

[0033] Step 133: performing data correlation processing on the topological elements in the back-end engineering monitoring layer to establish the corresponding relationship between the engineering resource configuration data and the engineering stage nodes and the spatial mapping relationship between the environmental influence factor data and the engineering implementation area.

[0034] The topological elements of the back-end engineering monitoring layer need to be associated with the engineering stage nodes and the engineering implementation area to play their monitoring role.

[0035] Step 1331: traversing all the topological elements in the back-end engineering monitoring layer to identify the resource node elements representing the engineering resource configuration data and the environmental node elements representing the environmental influence factor data.

[0036] The back-end engineering monitoring layer is traversed to identify according to the attribute labels of the topological elements. The attribute label of the resource node element contains a "Resource Type" field, and its value is "Human Resource" (human resource), "Material Resource" (material resource), "Financial Resource" (financial resource), etc.; the attribute label of the environmental node element contains an "Environment Type" field, and its value is "Temperature" (temperature), "Humidity" (humidity), "Wind Speed" (wind speed), etc. For example, the node element with "Resource Type: Material Resource" in the attribute label is a material resource node element, and the node element with "Environment Type: Temperature" in the attribute label is a temperature environmental node element.

[0037] Step 1332: extracting the resource type attribute and resource allocation period information of the resource node element, and extracting the resource demand type and resource use period parameters of each engineering stage node.

[0038] The resource type attribute of the resource node element is extracted from its "Resource Sub Type" field. For example, the "Resource Sub Type" value of the material resource node element can be "Steel", "Cement", "Transformer", etc. The resource allocation time range information is extracted from the "Allocation Time Range" field, in the format of "[Start Date, End Date]", such as "[2024-03-01, 2024-03-31]". The resource requirement type of each project stage node is extracted from its "Resource Requirement" attribute. For example, the "Resource Requirement" value of the site construction organization node in the construction stage is "Steel, Cement, Human Resource". The resource usage time range parameter is extracted from the "Resource Usage Time" field, also in the format of "[Start Date, End Date]", such as the resource usage time range parameter of the site construction organization node is "[2024-03-05, 2024-03-25]".

[0039] Step 1333: Match the resource type attribute with the resource requirement type, and calculate the time dimension overlap degree of the resource allocation time range information and the resource usage time range parameter. Based on the matching result and the overlap degree calculation result, establish the corresponding relationship between the project resource configuration data and the project stage node.

[0040] First, the resource type attribute of the resource node element is matched with the resource requirement type of the project stage node by string matching. For example, the resource type attribute "Steel" of the material resource node element is successfully matched with the resource requirement type "Steel" of the site construction organization node. Then, the time overlap degree of the resource allocation time range information and the resource usage time range parameter is calculated, using the method of dividing the overlapping days by the total days of the resource usage time range: overlap degree = overlapping days / (resource usage time range end date - resource usage time range start date + 1). For example, the resource allocation time range is [2024-03-01, 2024-03-31], the resource usage time range is [2024-03-05, 2024-03-25], the overlapping days are 21 days, and the total days of the resource usage time range are 21 days. The overlap degree is 21 / 21 = 1.0. When the matching is successful and the overlap degree is greater than a preset threshold (such as 0.8), the corresponding relationship between the resource node element and the project stage node is established, and the corresponding relationship table entry is recorded, such as (resource node ID: RS-008, project stage node ID: ST-12, matching degree: 1.0, overlap degree: 1.0).

[0041] Step 1334: Extract the environmental parameter collection location coordinates and environmental impact range description information of the environmental node element, and obtain the geographical boundary coordinates of each implementation area of the target main distribution network engineering.

[0042] Wherein, the environmental parameter collection location coordinates of the environmental node element are extracted from its "Location" attribute, in the format of longitude and latitude coordinates (longitude L1, latitude M1); the environmental impact range description information is extracted from the "Impact Range" attribute, such as "radius N meters circular area" or "east-west direction P meters, north-south direction Q meters rectangular area". The geographical boundary coordinates of each implementation area (such as construction area zoneA, construction area zoneB, equipment installation area zoneC, etc.) of the target main distribution network engineering are obtained through engineering design drawings and geographic information system, and each implementation area is represented by a sequence of polygon vertex coordinates, such as the geographical boundary coordinates of the construction area zoneA are (longitude L2, latitude M2), (longitude L3, latitude M3), (longitude L4, latitude M4), (longitude L5, latitude M5).

[0043] Step 1335: Calculate the spatial attribution relationship according to the environmental parameter collection location coordinates and the geographical boundary coordinates, determine the influence weight of the environmental impact factor on different implementation areas in combination with the environmental impact range description information, and establish the spatial mapping relationship between the environmental impact factor data and the engineering implementation area based on the spatial attribution relationship and the influence weight.

[0044] The spatial attribution relationship calculation uses the ray method to determine whether the environmental parameter collection location coordinates are located within the geographical boundary of a certain implementation area. If it is located inside, the environmental node element belongs to this implementation area; if it is located near the boundary or within the impact range of multiple implementation areas, the influence weight of the environmental impact factor on each implementation area needs to be further calculated. For example, the environmental impact range of the environmental node element is "radius N meters circular area", and there is an overlapping part between the circular area and the construction area zoneA and the construction area zoneB. By calculating the proportion of the overlapping area of the circular area and each construction area in the total area of the construction area, the influence weight is obtained. For example, the overlapping area proportion with the construction area zoneA is R%, and the influence weight is R%; the overlapping area proportion with the construction area zoneB is S%, and the influence weight is S%. Based on the spatial attribution relationship and the influence weight, the spatial mapping relationship between the environmental impact factor data and the engineering implementation area is established, such as (environmental node ID: ENV-012, implementation area ID: AREA-003, attribution relationship: inside, influence weight: 100%), (environmental node ID: ENV-012, implementation area ID: AREA-004, attribution relationship: overlap, influence weight: T%).

[0045] Step 134: Based on the correspondence and the spatial mapping relationship, the time-space feature extraction is performed on the back-end engineering monitoring data to generate the back-end engineering monitoring layer information containing the resource use efficiency index and the environmental impact assessment parameter.

[0046] The resource use efficiency index is calculated in combination with the correspondence between the engineering resource allocation data and the engineering stage node. For example, for the steel resource attributed to the site construction organization node, the resource use efficiency index can be calculated by the amount of work completed per unit of steel consumption, i.e., the amount of work / steel consumption, to obtain an index value such as "U cubic meters / ton". According to the spatial mapping relationship between the environmental impact factor data and the engineering implementation area, the time-space feature extraction is performed on the environmental parameter data, such as the average temperature, the maximum temperature, the temperature change trend of the construction area zone A in the past Q hours, and the impact level assessment results of the temperature parameter on the concrete pouring quality of the construction area zone A (such as "low impact", "medium impact", "high impact"). All resource use efficiency indexes and environmental impact assessment parameters are summarized to generate the back-end engineering monitoring layer information.

[0047] Step 135: Determine whether there is a missing key node state parameter or a discontinuous state change time sequence in the extracted front-end node state layer information. If there is a missing key node state parameter, make a supplementary prediction based on the historical state change law of the engineering stage node and the state characteristics of similar engineering stage nodes. If there is a discontinuous state change time sequence, perform a smooth transition processing according to the trend before and after the time sequence.

[0048] In the process of extracting the front-end node state layer information, the key node state parameter may be missing or the state change time sequence may be discontinuous due to data acquisition failure or transmission delay. The key node state parameter refers to a parameter that plays a decisive role in the evaluation of the running state of the engineering stage node, such as the "device debugging qualified rate" parameter of the device debugging node in the debugging and running stage. When it is detected that the parameter is missing, first obtain the change law of the "device debugging qualified rate" over time in the historical debugging process of the node, such as showing a gradual upward trend and conforming to a certain curve characteristic. At the same time, collect the "device debugging qualified rate" state characteristics of the device debugging node in similar projects, such as the average qualified rate level and the time period to reach the stable qualified rate. Based on the historical law and similar characteristics, a trend extrapolation method is used to make a supplementary prediction to obtain the missing parameter value. For a discontinuous state change time sequence, such as a data gap (T5-T4> preset time interval) between time points T4 and T5 for the progress parameter of the site construction organization node, the progress change trend before T4 (such as linear growth) and the progress change trend after T5 are analyzed, and a linear interpolation method is used to supplement the progress parameter value in the gap period to achieve smooth transition processing of the time sequence.

[0049] Step 136: Determine whether the timestamps of the extracted engineering resource allocation data and environmental impact factor data in the backend engineering monitoring layer information are synchronized, and whether the data measurement units are uniform; if the timestamps are not synchronized, perform time axis alignment adjustment based on key time nodes of engineering implementation; if the measurement units are not uniform, convert to a preset standard measurement unit system.

[0050] The engineering resource allocation data and environmental impact factor data may come from different collection systems, resulting in differences in timestamps. For example, the timestamp of material resource allocation data is "2024-04-01 08:30:00", while the timestamp of corresponding environmental temperature data is "2024-04-01 08:35:20", which are not synchronized. At this time, based on the key time nodes of engineering implementation (such as the hourly time), the data timestamps are aligned and adjusted, and the nearest neighbor interpolation method is used to adjust the timestamp of the environmental temperature data to the nearest key time node "2024-04-01 08:30:00" or "2024-04-01 09:00:00". In terms of measurement units, if there are two units of cement usage in material resources, "bags" and "tons", and the preset standard measurement unit system is "tons", then "bags" need to be converted to "tons" by the standard weight of each bag of cement (such as W kg / bag), that is, cement tons = bag number x W / 1000.

[0051] Step 137: Store the front-end node state layer information after completion of the judgment and the backend engineering monitoring layer information after completion of the judgment in association.

[0052] The front-end node state layer information and the backend engineering monitoring layer information after parameter supplement prediction, time series smoothing processing, timestamp synchronization and measurement unit unification are associated through engineering stage node ID and implementation area ID to establish a correlation relationship and stored in a relational database. For example, the state information of the site construction organization node (ID: ST-12) in the front-end node state layer and the resource allocation data (ID: RS-008) and the environmental impact factor data (ID: ENV-012) of the corresponding implementation area (ID: AREA-003) belonging to this node in the backend engineering monitoring layer are stored in association through the association fields ST-12, RS-008, AREA-003, so that the complete front-end and backend layer information of a certain engineering stage node or implementation area can be quickly obtained for subsequent query and analysis.

[0053] Step 140: using the front-end node state layer information and the back-end engineering monitoring layer information, risk evolution path deduction is performed by constructing a multi-dimensional influence factor correlation network to simulate risk events of the target main distribution network project, to obtain a risk event simulation result corresponding to the target main distribution network project, and a cross-node collaborative response strategy is generated according to the risk event simulation result.

[0054] In order to effectively identify and respond to potential risks in the whole life cycle of the target main distribution network project, a multi-dimensional influence factor correlation network is constructed in the embodiment of the application to simulate the evolution process of risk events and generate a corresponding collaborative response strategy.

[0055] Step 141: the node running state parameters in the front-end node state layer information are associated and fused with the resource configuration data and environmental influence factor data in the back-end engineering monitoring layer information to construct a multi-dimensional influence factor correlation network including node state dimension, resource dimension and environmental dimension.

[0056] The node running state parameters include processing progress, performance indicators, abnormal early warning states and the like of nodes in each engineering stage; the resource configuration data includes allocation amount, use efficiency, reserve amount and the like of various resources; and the environmental influence factor data includes environmental parameters such as temperature, humidity and wind speed and influence evaluation results. Through data association and fusion technology, these multi-dimensional data from different layers are integrated into a unified network structure. For example, the “processing progress 65%” (node state dimension) of the field construction organization node, the “steel resource use efficiency U cubic meters / ton” (resource dimension), and the “average temperature 28℃ in the construction area zone A, environmental influence level medium influence” (environmental dimension) are associated to form an associated node in the network. The edges in the network represent the mutual influence relationship between different dimensional factors, such as high temperature environment (environmental dimension) may cause the node processing progress to slow down (node state dimension), and low steel resource use efficiency (resource dimension) may affect the node processing progress (node state dimension). In this way, a multi-dimensional influence factor correlation network is constructed.

[0057] Step 142: setting risk trigger conditions and influence transmission rules in the multi-dimensional influence factor correlation network, and generating a risk event evolution path set under different initial condition combinations through random sampling.

[0058] Step 1421: for each engineering stage node in the multi-dimensional influence factor correlation network, setting a risk trigger threshold corresponding to the engineering stage node according to historical engineering data and prior knowledge, the risk trigger threshold including a node running state parameter threshold, a resource configuration threshold and an environmental influence factor threshold.

[0059] The historical engineering data includes parameter records when risk events occur at each node in similar main and distribution network engineering in the past, and the priori knowledge includes industry standards, expert experience, etc. For example, for the device debugging node, the node running state parameter threshold is set as "device debugging qualified rate is lower than W%", when the actual qualified rate is lower than the value, triggering the debugging risk; the resource configuration threshold is set as "the number of debugging personnel is less than K", when the actual number of personnel is insufficient, triggering the resource insufficient risk; the environmental influence factor threshold is set as "the debugging environment humidity is higher than Y%", when the humidity is out of standard, triggering the environmental risk.

[0060] Step 1422: when the actual parameter value of the engineering stage node exceeds the corresponding risk triggering threshold, it is determined that the risk event occurs at the engineering stage node, and the occurrence time and initial influence degree of the risk event are recorded.

[0061] The actual parameter value of each engineering stage node is monitored in real time, and compared with the preset risk triggering threshold. For example, the actual debugging qualified rate of the device debugging node is Z% (Z

[0062] Step 1423: defining the influence transmission coefficient between the engineering stage nodes in the multi-dimensional influence factor correlation network, the influence transmission coefficient representing the probability and degree of influence of one engineering stage node on its associated nodes when the risk event occurs.

[0063] The influence transmission coefficient is obtained based on statistical analysis of the propagation data of historical risk events. For example, when the device debugging node occurs debugging risk event, the probability of influence on the system debugging node is α, and the influence degree quantization value is β (range 0-1, the larger the value, the higher the influence degree), then the influence transmission coefficient is (α, β). For directly associated nodes, the influence transmission coefficient is usually high; for indirectly associated nodes, the influence transmission coefficient is low or zero.

[0064] Step 1424: constructing an influence transmission matrix based on the influence transmission coefficient, the element value in the matrix representing the influence transmission probability and influence degree quantization value from one engineering stage node to another engineering stage node.

[0065] For example, there are M engineering stage nodes, and an MxM influence transmission matrix is constructed, the matrix row representing the risk event source node, the list representing the possible affected target node, and the matrix element being (influence transmission probability, influence degree quantization value). For example, the element in the i-th row and j-th column of the matrix is (α ij ,βij , which represents the influence transmission probability of node j when node i has a risk event ij , which represents the influence degree quantization value of node j when node i has a risk event ij .

[0066] Step 1425: Randomly perturb the risk trigger threshold in the multi-dimensional influence factor correlation network multiple times, and generate a new set of initial condition combinations for each perturbation.

[0067] Random perturbation refers to randomly adjusting the value of the risk trigger threshold within a predetermined range, such as randomly adjusting the pass rate threshold W% of the device debugging node within ±Δ%, generating new thresholds W1%, W2%, etc. After each perturbation, a set of initial condition combinations containing new risk trigger thresholds for all nodes is obtained.

[0068] Step 1426: Under each set of initial condition combinations, determine the initial risk event node according to the risk trigger condition, and then gradually deduce the diffusion process of the risk event in the network from the initial risk event node according to the influence transmission rules and the influence transmission matrix, record the node sequence, occurrence time and influence degree of the risk event, and form a risk event evolution path.

[0069] Under a set of initial condition combinations, for example, the new pass rate threshold of the device debugging node is W1%, while the current actual pass rate is still Z%, if Z < W1%, the device debugging node is still the initial risk event node; if Z ≥ W1%, the node does not trigger a risk, and other nodes need to be re-determined to determine whether they trigger a risk. After determining the initial risk event node, the diffusion of the risk event from the initial node to the associated node is deduced according to the influence transmission rules (such as transmitting in the order of influence transmission probability from high to low, and continuing to transmit only when the influence degree quantization value reaches a certain threshold, etc.) and the influence transmission matrix. For example, the device debugging node has a risk, and influences the system integration node with a probability of α ij , if the influence is successful, the system integration node is recorded as the next node through which the risk event passes, the occurrence time is the risk occurrence time of the device debugging node plus the transmission delay time, and the influence degree is the initial influence degree of the device debugging node multiplied by β ij . In this way, a complete risk event evolution path is formed until the risk event no longer spreads or reaches the preset maximum number of deduction steps.

[0070] Step 1427: Repeat the random sampling and deduction process multiple times until a preset number of risk event evolution paths are generated, forming a risk event evolution path set.

[0071] The preset number is determined according to the engineering complexity and risk analysis requirements, such as generating K risk event evolution paths. Through multiple random sampling of different initial condition combinations and deduction, a risk event evolution path set containing K different paths is obtained.

[0072] Step 143: The impact degree of each risk event evolution path is evaluated, the risk loss value and risk diffusion speed parameter of each engineering stage node in the path are calculated, and the severity level of the risk event is determined according to the risk loss value and risk diffusion speed parameter.

[0073] Step 1431: The node impact chain identification process is performed on each risk event evolution path, all engineering stage nodes directly affected by the risk event and indirectly affected by the risk event in the risk event evolution path are extracted, and a risk impact hierarchical structure containing a direct impact node layer and an indirect impact node layer is constructed.

[0074] The direct impact node layer is the node directly affected by the initial risk event node in the risk event evolution path, the indirect impact node layer is the node affected by the direct impact node, and a multi-level structure is formed in this way. For example, the initial risk event node is the equipment debugging node, the direct impact node layer is the system integration node, the indirect impact node layer is the trial operation evaluation node, and a three-level risk impact hierarchical structure of “equipment debugging node-system integration node-trial operation evaluation node” is formed.

[0075] Step 1432: The core business interruption mode identification is performed on the direct impact node layer in the risk impact hierarchical structure, the business function loss degree of each direct impact node and the blocking type of the associated business process are identified, and the risk transmission path and transmission intensity characteristics of the direct impact node are determined.

[0076] The core business interruption mode identification is performed by analyzing the business function implementation of the direct impact node under the risk event. For example, the loss degree of the business function “system compatibility test” of the system integration node affected by the equipment debugging node risk is “50%” (i.e. only half of the test work can be completed), and the blocking type of the associated business process “debugging-trial operation process” is “partial blocking” (the process is not completely stopped, but the progress is seriously delayed). According to the business function loss degree and the blocking type, the risk transmission path is determined as “equipment debugging node-system integration node-trial operation evaluation node”, and the transmission intensity characteristics are “medium intensity, gradually weakened”.

[0077] Step 1433: Based on the risk transmission path and transmission intensity characteristics, the spread range deduction is performed on the indirect impact node layer, and the potential secondary risk event type and risk coupling effect mode in the indirect impact node layer are identified.

[0078] The secondary risk event type refers to a new risk type caused by the transmission of the initial risk event in the indirectly affected node. For example, the commissioning evaluation node cannot carry out the evaluation work on time due to the progress delay of the system integration node, which may cause the “commissioning delay risk” as a secondary risk event. The risk coupling effect mode refers to the interaction mode of multiple risk events, such as the “stacking effect” of the commissioning delay risk and the equipment warranty expiration risk, which may cause the overall risk loss to increase.

[0079] Step 1434: According to the loss degree of the business function of the directly affected node, the secondary risk event type of the indirectly affected node, and the risk coupling effect mode, a risk impact degree qualitative assessment matrix is constructed, the row dimension of the assessment matrix represents the set of directly affected nodes, the column dimension represents the set of indirectly affected nodes, and the matrix element represents the qualitative description of the risk impact between the corresponding nodes.

[0080] The qualitative description of the assessment matrix element is based on a preset description system, such as “no impact”, “slight impact”, “moderate impact”, and “severe impact”. For example, the row dimension is the system integration node, the column dimension is the commissioning evaluation node, and the matrix element is “severe impact (causing commissioning delay risk, with stacking effect)”.

[0081] Step 1435: Hierarchical weighting processing is performed on the risk impact degree qualitative assessment matrix, and a risk event impact degree comprehensive description vector is generated according to the core business weight of the directly affected node and the spread range weight of the indirectly affected node.

[0082] The core business weight is determined according to the importance of the directly affected node in the engineering whole life cycle, such as the core business weight of the system integration node being V1; the spread range weight is determined according to the size of the range affected by the indirectly affected node, such as the spread range weight of the commissioning evaluation node being V2. In the hierarchical weighting processing, the qualitative description of each element in the assessment matrix is converted into a quantitative score (such as “severe impact” being recorded as V3 points), and then multiplied by the corresponding core business weight and spread range weight to obtain the weighted score. All weighted scores are arranged in node order to generate a risk event impact degree comprehensive description vector, such as [V4, V5, V6,...], where each element corresponds to the comprehensive impact score of a node.

[0083] Step 1436: Time series analysis is performed on the node impact order in the risk event evolution path, the propagation time sequence characteristics of the risk event from the initial node to each level node are identified, and the bottleneck node and the acceleration node in the propagation process are extracted.

[0084] The propagation timing features include propagation delay time, propagation interval time, etc. After the risk of the device debugging node occurs, it propagates to the system integration node after T1 time, and then propagates to the trial operation evaluation node after T2 time. The size relationship between T1 and T2 reflects the change of the propagation speed. The bottleneck node refers to a node that significantly slows down the propagation speed in the propagation process. For example, the system integration node has limited processing capacity, resulting in a long propagation delay time T1. The acceleration node refers to a node that can accelerate the propagation speed. For example, a node with high information sharing degree can shorten the propagation interval time.

[0085] Step 1437: Based on the propagation timing features, bottleneck nodes and acceleration nodes, a risk diffusion mode atlas is constructed, which includes a qualitative description of the key path of risk diffusion, node blocking coefficient and node acceleration coefficient.

[0086] The key path is the path that has the greatest impact and the fastest propagation speed in the risk event propagation process. The node blocking coefficient qualitatively describes the degree of hindering of the bottleneck node to the risk diffusion, such as “high blocking”, “medium blocking” and “low blocking”. The node acceleration coefficient qualitatively describes the degree of promotion of the acceleration node to the risk diffusion, such as “high acceleration”, “medium acceleration” and “low acceleration”. For example, the key path in the risk diffusion mode atlas is “device debugging node-system integration node-trial operation evaluation node”, the blocking coefficient of the system integration node is “medium blocking”, and there is no acceleration node.

[0087] Step 1438: Based on the risk event influence degree comprehensive description vector and the risk diffusion mode atlas, the risk event is multi-dimensionally qualitatively classified to determine the severity level of the risk event. The severity level is determined based on the number of influence levels, the type of core business blocking and the complexity of the diffusion mode.

[0088] The more the number of influence levels, the more serious the type of core business blocking (such as “complete blocking” is more serious than “partial blocking”), and the more complex the diffusion mode (such as multiple bottleneck nodes and acceleration nodes interact with each other), the higher the severity level of the risk event. The severity level can be divided into five levels: “slight”, “general”, “relatively large”, “major” and “particularly major”. For example, a risk event has 3 influence levels, the core business blocking type is “partial blocking”, and the diffusion mode complexity is “medium”, so the comprehensive judgment of the severity level is “relatively large”.

[0089] Step 144: The risk event evolution paths with the same severity level and similar evolution characteristics are clustered and analyzed, the typical evolution mode and key influence node of each type of risk event are extracted, and the risk event simulation results including the type of risk event, occurrence probability, influence range and severity level are generated.

[0090] The clustering algorithm (such as K-means clustering) is used to cluster the risk event evolution path set, and the paths with the same severity level and similar evolution characteristics (such as node sequence, influence degree change trend, diffusion mode, etc.) are clustered into a class. For each class of risk event, extract its typical evolution mode (such as "device debugging risk-system debugging risk-delayed commissioning risk") and key influence node (such as device debugging node, system debugging node, commissioning evaluation node). By counting the proportion of each category path in the total path set, the risk event occurrence probability is obtained; the influence range is determined according to the number of nodes involved in the path and the implementation area range. The final risk event simulation result contains information such as "risk event type: device debugging risk triggered chain risk, occurrence probability: P(w)%, influence range: 3 nodes in the debugging running stage, severity level: large".

[0091] Step 145: According to the risk event type and severity level in the risk event simulation result, combined with the business attributes of the key influence nodes, create a cross-node collaboration response strategy containing resource scheduling priority, node collaboration sequence and disposal process specification.

[0092] Step 1451: Feature deconstruction is performed on the risk event type in the risk event simulation result, and the trigger condition feature, influence range feature and evolution trend feature of the risk event are extracted to construct a risk event feature vector library; based on the business attributes of the key influence nodes, the core ability type and resource reserve state of each key influence node in the risk disposal process are identified, and a node ability-resource mapping relationship model is established; the risk event feature vector library and the node ability-resource mapping relationship model are associated and matched to determine the candidate response node set that produces disposal effect on the risk event type and the response ability weight of each candidate node; according to the severity level of the risk event and the response ability weight of the candidate response node, a response node dynamic selection mechanism is constructed, and when the severity level of the risk event changes, the candidate response node set and the response ability weight distribution proportion are automatically adjusted.

[0093] The trigger condition features of the risk event type include, for example, "device commissioning qualification rate is lower than a threshold value", "high temperature environment impact", and the like; the influence range features include, for example, "involving the commissioning running stage", "affecting 3 nodes", and the like; and the evolution trend features include, for example, "risk diffusion speed is moderate", "may cause secondary risks", and the like. After quantifying these features, a risk event feature vector library is constructed. The business attributes of the key influence nodes include the core capability types (such as device maintenance capability, resource allocation capability, technical support capability, and the like) and the resource reserve states (such as the number of spare devices, the number of emergency personnel, the amount of special funds, and the like) possessed by the nodes, and accordingly, a node capability-resource mapping relationship model is established, such as that the core capability type of the device maintenance node is "device maintenance capability" and the resource reserve state is "X spare transformers and Y maintenance personnel". Through association matching, candidate response nodes capable of disposing a specific risk event type are found, such as that the device commissioning risk can be responded by the device maintenance node, the technical support node, and the like, and the response capability weights of each candidate node are determined according to the matching degree of the capability of the node and the risk event and the resource reserve sufficiency. The response node dynamic selection mechanism stipulates that when the risk event severity level is upgraded from "moderate" to "major", the senior technical expert node and the emergency command node are added to the candidate response node set, and the response capability weight of the device maintenance node is increased.

[0094] Step 1452: based on the candidate response node set and the response capability weight, a cross-node collaboration network topology structure is constructed, and the topology structure includes information interaction channels and resource sharing paths between nodes.

[0095] The nodes in the candidate response node set are connected through the information interaction channels and the resource sharing paths, forming a cross-node collaboration network topology structure. For example, a high-speed data communication channel (information interaction channel) is established between the device maintenance node and the technical support node, for sharing device fault diagnosis data and maintenance schemes; a material allocation path (resource sharing path) is established between the device maintenance node and the material reserve node, for quickly obtaining spare parts required for maintenance. The edge weight in the topology structure can be determined by the response capability weight and the connection strength (such as information transmission rate, resource allocation efficiency) between nodes.

[0096] Step 1453: in the cross-node collaboration network topology structure, the node cooperation sequence is determined according to the evolution trend features of the risk event and the geographical distribution attributes of the nodes, and the cooperation sequence follows the joint sorting rule of the risk influence area from near to far and the response capability weight from high to low.

[0097] The evolution trend characteristics of the risk event, such as "risk spreading to the northwest construction area", are located in the northwest direction of the candidate response node in the geographical distribution, and the node with high response ability weight (such as the equipment maintenance node weight qZ%, the technical support node weight qA%, and qZ>qA) is preferentially cooperated. Under the joint sorting rule, first, the risk influence area is sorted from near to far, and the nodes in the same area are sorted from high to low according to the response ability weight, and the node cooperation order is obtained, such as "equipment maintenance node (northwest area, qZ%) - technical support node (northwest area, qA%) - emergency command node (central area, qB%)".

[0098] Step 1454: For the determined node cooperation order, a treatment process framework containing information transmission protocol, resource transfer specification and decision chain level is created, and the decision threshold and information feedback mechanism of the key node are set in the treatment process framework.

[0099] The information transmission protocol specifies the format, frequency and security requirements of information interaction between nodes, such as that the equipment maintenance node sends a fault handling progress report to the technical support node every set minutes, and the report format is encrypted JSON format. The resource transfer specification clearly specifies the procedures, responsibility division and acceptance standards of resource transfer between nodes, such as when the material reserve node allocates spare parts to the equipment maintenance node, it needs to fill out the allocation form and be signed and confirmed by the responsible persons of both parties, and the spare parts acceptance needs to check the model, quantity and integrity. The decision chain level determines the decision authority of different nodes, such as the general maintenance scheme is decided by the equipment maintenance node, and the major maintenance scheme needs to be reported to the emergency command node for approval. The decision threshold of the key node is "when the maintenance cost estimation exceeds R ten thousand yuan, it needs to be reported to the emergency command node"; the information feedback mechanism requires each node to feed back the treatment result to the previous node and the emergency command node in the cooperation order after completing the treatment task.

[0100] Step 1455: According to the influence range characteristics of the risk event and the node cooperation order, the resource scheduling priority rules are embedded in the treatment process framework, and the resource scheduling priority rules are dynamically adjusted based on the influence degree of the risk event on each engineering stage node and the business recovery urgency of the node.

[0101] The higher the influence degree of the risk event on the engineering stage node and the more urgent the business recovery, the higher the priority of the node to obtain resource scheduling. For example, the equipment debugging node is stalled due to the risk influence, and the business recovery urgency is the highest, so the resource scheduling priority is set to "level 1" (the highest level); the system integration node is less affected, and the priority is set to "level 2". When the influence range of the risk event expands or shrinks, or the business recovery urgency of the node changes, the resource scheduling priority is dynamically adjusted, such as the business recovery urgency of the system integration node increases due to the secondary risk, and the priority is adjusted to "level 1".

[0102] Step 1456: The treatment process framework is instantiated with resource scheduling priority rules to generate a cross-node collaboration response strategy including a node collaboration timeline, resource scheduling path planning, and decision authority allocation table.

[0103] The node collaboration timeline shows the time arrangement of the collaborative work of each candidate response node in the form of a time axis, such as "the equipment maintenance node starts troubleshooting at time g1 for L1 hours; the technical support node joins the collaboration at time g1+L2 hours to provide technical guidance." The resource scheduling path planning clearly specifies the optimal transmission path and scheduling order of various resources (personnel, materials, and funds) from the reserve node to the demand node, such as "allocate spare parts from the warehouse C1 of the material reserve center, transport to the equipment debugging site via R1 route, and prioritize VE model vehicles." The decision authority allocation table lists the decision-making authority of each node at different treatment stages in detail, such as "equipment maintenance node: general fault treatment scheme approval, maintenance personnel allocation; emergency command node: major fault treatment scheme approval, more than R million maintenance fund approval." These contents are integrated to form a complete cross-node collaboration response strategy.

[0104] In an optional embodiment, the method further comprises: step 210: analyzing the historical execution data of the risk event simulation results and the cross-node collaboration response strategy, extracting the key influence factor combination of the risk event, the execution path characteristics of the response strategy, and the qualitative description information of the strategy execution effect, and constructing a risk event-response strategy association dataset.

[0105] The historical execution data includes past risk event cases, corresponding cross-node collaboration response strategies, and their execution process and result records. By analyzing these data, the key influence factor combination of the risk event is extracted, such as "high temperature environment + low equipment debugging qualification rate + insufficient human resources"; the execution path characteristics of the response strategy include the actual collaborative node sequence, resource scheduling path, decision-making process, and deviation from the preset strategy; the qualitative description information of the strategy execution effect is, for example, "the risk event was controlled within S hours and no serious loss was caused," "part of the node collaboration was delayed, resulting in the expansion of the risk diffusion range," etc. These information is associated according to the risk event type and the response strategy to construct a risk event-response strategy association dataset.

[0106] Step 220: Feature engineering processing is performed on the risk event-response strategy association dataset to identify risk event features of the risk event and response strategy features of the response strategy; the internal correlation between the risk event features and the response strategy features is mined to generate a risk-response association rule set containing multiple groups of premise conditions and conclusion relationships.

[0107] The feature engineering process includes standardization and discretization of risk event features and response strategy features, so as to perform correlation mining. The risk event features are, for example, risk event type, severity level, key influence factor combination, etc. The response strategy features are, for example, resource scheduling priority, node coordination sequence, disposal process complexity, etc. Through an association rule mining algorithm (such as the Apriori algorithm), the internal correlation between the risk event features and the response strategy features is mined, and a risk-response association rule set is generated, such as “if the risk event type is device debugging risk and the severity level is greater, then the resource scheduling priority is level 1, the node coordination sequence is the device maintenance node-technical support node, and the response strategy execution effect is good” (premise: risk event type = device debugging risk, severity level = greater; conclusion: response strategy feature = resource scheduling priority level 1, node coordination sequence = device maintenance node-technical support node, execution effect = good).

[0108] Step 230: According to the risk-response association rule set, a risk event disposal knowledge graph is constructed, which contains risk event type nodes, response strategy nodes, influence factor nodes, and association relationship edges between the nodes.

[0109] The knowledge graph represents risk event types, response strategies, influence factors, etc. as nodes, and represents the association relationships between entities as edges. For example, the risk event type node “device debugging risk” is connected to the influence factor nodes “high temperature environment” and “device quality problem” through the association relationship edge “caused by…”, and is connected to the response strategy node “strategy TS1” (containing the features of resource scheduling priority level 1) through the association relationship edge “adopted for disposal”. The association relationship edge can be labeled with a rule confidence description, such as the rule confidence description of “device debugging risk caused by high temperature environment” being “LL%”.

[0110] Step 240: Feature extraction is performed on a newly occurring risk event or a simulation generated risk event, similarity matching is performed between the extracted features and the risk event type nodes in the risk event disposal knowledge graph, and the risk event type node with the highest matching degree is determined.

[0111] A newly occurring risk event is, for example, “rainstorm in construction area zone A causes 20% delay in the progress of the on-site construction organization node”. The features extracted therefrom include “triggering condition: rainstorm environment, affected node: on-site construction organization node, influence degree: 20% delay in progress”. Similarity calculation is performed between these features and the features of each risk event type node in the knowledge graph, such as the highest similarity (B2%) with the features of the “construction progress risk caused by environmental factors” type node, so that the risk event type node with the highest matching degree is determined as the node.

[0112] Step 250: Based on the risk event type node, traverse the risk event handling knowledge graph to obtain the response strategy node associated with the risk event type node and the corresponding association relationship edge information, and generate a candidate response strategy set.

[0113] Traverse the "adopt … handling" association relationship edge of the "construction progress risk caused by environmental factors" type node in the knowledge graph, and obtain all associated response strategy nodes such as "strategy TS2", "strategy TS3", etc. These nodes and the strategy characteristics they contain jointly constitute a candidate response strategy set.

[0114] Step 260: Evaluate and sort the response strategies in the candidate response strategy set, select the optimal candidate response strategy as the recommended handling strategy for the current risk event according to the evaluation and sorting result, and feed back the recommended handling strategy and its decision basis to the decision support system.

[0115] The evaluation and sorting basis includes the historical execution effect of the response strategy, the matching degree with the current risk event characteristics, the rule confidence description, etc. For example, the historical execution effect of strategy TS2 is "average progress delay reduction J1%", the matching degree with the current risk event characteristics is J2%, the rule confidence description is "J3%", and the comprehensive score is higher than that of strategy TS3. Therefore, strategy TS2 is selected as the optimal recommended handling strategy. The decision basis includes the matched risk event type node, the associated response strategy node, and the key indicators in the evaluation and sorting process. These information is fed back to the decision support system to assist the management personnel in decision-making.

[0116] Step 270: In the execution process of the recommended handling strategy, real-time collection of execution data and effect feedback information, updating the rule confidence description of the association relationship edge in the risk event handling knowledge graph according to the execution data and the effect feedback information, and adding new risk event characteristics, response strategy characteristics and effect characteristics to the risk event handling knowledge graph.

[0117] In the execution process of the recommended handling strategy TS2, the collected execution data includes "actual progress delay reduction 15%", and the effect feedback information is "basically achieves the expected effect". According to these information, if the actual effect is consistent with the expected effect of the association relationship edge "environmental factors cause construction progress risk to adopt strategy TS2 handling", the rule confidence description of the association relationship edge is improved, such as from "J3%" to "J4%"; if not, the rule confidence description is reduced. At the same time, the characteristics of the new risk event, the new characteristics of strategy TS2 in this execution (such as actual resource usage), and the effect characteristics (progress delay reduction 15%) are added to the knowledge graph, enriching the content of the knowledge graph.

[0118] In an optional embodiment, the method further comprises: step 310: after the cross-node collaborative response strategy is generated, collecting real-time node state data and real-time resource environment data in the execution process of the cross-node collaborative response strategy, the real-time node state data including actual operation state parameters of each engineering stage node in the strategy execution, and the real-time resource environment data including actual resource configuration data and environmental influence factor data in the strategy execution process.

[0119] In the execution process of the cross-node collaborative response strategy, data is collected in real time by sensors, data collection terminals and other devices. The real-time node state data is, for example, the actual debugging qualification rate B1% of the device debugging node after the strategy execution, and the processing progress B2%; the real-time resource environment data is, for example, the actual number of spare parts TT allocated to the device maintenance node, and the real-time temperature 25℃ of the construction area zone A.

[0120] Step 320: comparing and analyzing the real-time node state data and the real-time resource environment data with the expected node state parameters, the expected resource configuration data and the expected environmental influence factor data in the risk event simulation result, identifying data deviation characteristics, the data deviation characteristics including deviation type, deviation degree and deviation duration.

[0121] The expected node state parameter is, for example, the expected qualification rate B3% of the device debugging node after the strategy execution in the risk event simulation result. The actual qualification rate B1% is compared with the expected qualification rate B3%, the deviation type is “negative deviation” (actual lower than expected), the deviation degree is “|B1-B3|%”, and the deviation duration is “B4 minutes” (from the discovery of the deviation to the current duration). The deviation characteristics of the resource configuration data and the environmental influence factor data are analyzed in the same way.

[0122] Step 330: determining the associated changes of influencing factors causing the deviation based on the data deviation characteristics, the associated changes of influencing factors including actual execution deviation of inter-node information interaction rules, blocking condition of data flow path and actual effectiveness degree of resource scheduling priority.

[0123] The data deviation characteristics are “negative deviation of device debugging qualification rate, deviation degree B5%, and duration B4 minutes”. The possible associated changes of influencing factors are analyzed. The actual execution deviation of inter-node information interaction rules is, for example, that the debugging scheme information feedback from the technical support node to the device debugging node is sent with delay, causing the device debugging node to fail to adjust the debugging strategy in time; the blocking condition of the data flow path is, for example, that the data transmission channel between the device debugging node and the system integration node is blocked due to network failure, causing the data transmission rate to decrease by B6%; and the actual effectiveness degree of the resource scheduling priority is, for example, that the special debugging tool that should be allocated to the device debugging node is allocated to other nodes, causing the debugging efficiency to decrease.

[0124] Step 340: According to the data bias characteristics and the influence factor correlation changes, adjust the node coordination sequence and resource scheduling priority rules in the cross-node collaboration response strategy, and generate adjusted cross-node collaboration response strategy parameters.

[0125] For information interaction rule execution bias, adjust the node coordination sequence, advance the feedback link of the technical support node, and ensure timely information transmission. For insufficient actual effectiveness of resource scheduling priority, increase the resource scheduling priority level of the device debugging node, such as from “level 2” to “level 1”, and add a resource locking mechanism to prevent special resources from being adjusted. These adjustment contents constitute the adjusted cross-node collaboration response strategy parameters.

[0126] Step 350: Input the adjusted cross-node collaboration response strategy parameters into the risk evolution scenario corresponding to the risk event simulation result, simulate the execution effect of the adjusted cross-node collaboration response strategy under the current real-time data condition, and if the simulation execution effect meets the preset bias convergence condition, determine the adjusted cross-node collaboration response strategy as the dynamically optimized cross-node collaboration response strategy; if it does not meet the bias convergence condition, repeat the data comparison and analysis, influence factor correlation change analysis, and strategy parameter adjustment steps until the simulation execution effect meets the bias convergence condition.

[0127] The bias convergence condition is, for example, “the device debugging qualified rate bias degree is reduced to within B7%, and the continuous time length is not more than B8 minutes”. Input the adjusted strategy parameters into the risk evolution scenario simulation, and if the simulation result shows that the device debugging qualified rate bias degree is reduced to within B7%, and the continuous time length is only qA minutes, the bias convergence condition is met, and the adjusted strategy is determined as the dynamically optimized cross-node collaboration response strategy; if the simulation result still does not meet the condition, continue to analyze the bias reasons, adjust the node coordination sequence (such as adding the coordination link of the emergency command node) and the resource scheduling priority rule (such as configuring exclusive resource reserves for the device debugging node), and simulate again until the condition is met.

[0128] The embodiment of the application firstly performs engineering information structured modeling on the whole life cycle of the target main distribution network project, constructs an engineering stage node index model containing engineering stage nodes and associated constraint relationships; on this basis, chain tracking analysis is performed to obtain a data flow conversion topological graph, the information input-output relationship and data processing logic of each engineering stage node are clarified, and accurate grasp of the data flow conversion of the whole life cycle of the project is realized; based on the front-end and back-end separation strategy, the front-end node state graph layer information and the back-end engineering monitoring graph layer information are extracted, and various data such as real-time operation state, resource allocation and environmental impact of the project can be comprehensively obtained; using the above information, a multi-dimensional influence factor correlation network is constructed to perform risk evolution path deduction and risk event simulation, which can accurately predict the risks that the project may face, and generate a cross-node collaborative response strategy containing a collaborative disposal process and a resource scheduling scheme, thereby improving the ability of the target main distribution network project to cope with risks and ensuring the stable and efficient operation of the project in the whole life cycle.

[0129] In this way, through the digital support of node chain tracking and front-end and back-end separation architecture, the data islands of each stage of the main distribution network project are eliminated, the collaboration efficiency of project management is improved, and the controllability of the whole life cycle process is strengthened.

[0130] Referring to Figure 2 , which is a schematic diagram of the basic structure of a main distribution network project digital management system 200 provided by the embodiment of the application, the main distribution network project digital management system 200 comprises: a processor 201; a storage device 202 having a computer program 2020 stored thereon; a network interface 203 for providing network communication function; when the computer program 2020 is executed by the processor 201, the processor 201 realizes the main distribution network project digital management method.

[0131] Referring to Figure 3The function module block diagram of the main distribution network engineering digital management device is provided in the embodiments of the present application, and the main distribution network engineering digital management device comprises: an engineering information modeling module, which is configured to perform engineering information structured modeling on the whole life cycle of a target main distribution network engineering, to obtain an engineering stage node index model corresponding to the target main distribution network engineering, wherein the engineering stage node index model comprises a plurality of engineering stage nodes corresponding to the target main distribution network engineering and a correlation constraint relationship between the engineering stage nodes; a chain tracking analysis module, which is configured to perform chain tracking analysis on the plurality of engineering stage nodes, to obtain a data flow conversion topology graph of the target main distribution network engineering under the whole life cycle by establishing information interaction rules and data transmission paths between the engineering stage nodes, and to use the data flow conversion topology graph to represent information input / output relationships and data processing logic of the engineering stage nodes; a layer information extraction module, which is configured to extract front-end node state layer information and back-end engineering monitoring layer information of the data flow conversion topology graph based on a front-end / back-end separation strategy, wherein the front-end node state layer information comprises real-time running state parameters of the engineering stage nodes, and the back-end engineering monitoring layer information comprises engineering resource configuration data and environmental influence factor data; and a risk simulation response module, which is configured to use the front-end node state layer information and the back-end engineering monitoring layer information, to perform risk event simulation on the target main distribution network engineering by constructing a multi-dimensional influence factor correlation network to perform risk evolution path deduction, to obtain a risk event simulation result corresponding to the target main distribution network engineering, and to generate a cross-node collaborative response strategy according to the risk event simulation result, wherein the cross-node collaborative response strategy comprises a collaborative disposal process and a resource scheduling scheme of different engineering stage nodes.

[0132] On the basis described above, a readable storage medium is provided, and the readable storage medium stores programs or instructions. When the programs or instructions are executed by a processor, the steps of the method described above are implemented.

[0133] In addition, it should be noted that the embodiments of the present application also provide a computer program product, which can include a computer program that can be stored in a computer readable storage medium. The processor of the main distribution network engineering digital management system reads the computer program from the computer readable storage medium, and the processor can execute the computer program to make the main distribution network engineering digital management system execute the foregoing Figure 1 The description of the method in the corresponding embodiments is described above, and therefore, the description will not be repeated here. In addition, the beneficial effects of using the same method will not be repeated. For technical details not disclosed in the computer program product embodiments of the present application, please refer to the description of the method embodiments of the present application.

[0134] It should be noted that the various embodiments described in the specification are intended to be exemplary only and that the scope of the application is not intended to be limited to the embodiments described in the specification.

Claims

1. A digital management method for main and distribution network projects, characterized in that, include: A structured model of the entire lifecycle of the target main distribution network project is performed to obtain the project stage node index model corresponding to the target main distribution network project. The project stage node index model includes multiple project stage nodes corresponding to the target main distribution network project and the correlation constraints between each project stage node. Chain tracking analysis is performed on the multiple project stage nodes. By establishing information interaction rules and data transmission paths between project stage nodes, a data flow topology diagram of the target main distribution network project under the entire lifecycle is obtained. The data flow topology diagram is used to characterize the information input and output relationships and data processing logic of each project stage node. Based on a front-end and back-end separation strategy, the front-end node status layer information and back-end engineering monitoring layer information of the data flow topology are extracted. The front-end node status layer information includes real-time operating status parameters of nodes at each engineering stage, and the back-end engineering monitoring layer information includes engineering resource configuration data and environmental influencing factor data. Using the front-end node status layer information and the back-end engineering monitoring layer information, a multi-dimensional influencing factor association network is constructed to extrapolate the risk evolution path and simulate risk events for the target main distribution network project. The simulation results of risk events corresponding to the target main distribution network project are obtained. Based on the simulation results of risk events, a cross-node collaborative response strategy is generated. The cross-node collaborative response strategy includes collaborative handling procedures and resource scheduling schemes for nodes at different engineering stages.

2. The digital management method for main and distribution network projects as described in claim 1, characterized in that, The chain-like tracking analysis of the multiple project stage nodes, by establishing information interaction rules and data transmission paths between project stage nodes, yields the data flow topology of the target main distribution network project throughout its entire lifecycle. This includes: constructing a project stage node association network based on the project stage node indicator model; identifying the preceding and following associated nodes of each project stage node through the project stage node association network to determine the direct and indirect associations between nodes; creating information interaction rules between nodes based on the business attributes and data characteristics of each project stage node, the information interaction rules including data format conversion requirements, information verification standards, and interaction timing control parameters; tracking the data flow path of each project stage node in the project stage node association network according to the project implementation time sequence and logical relationship, and recording the transmission direction, content, and frequency of data between nodes; and constructing a directed graph structure containing node connection relationships, data flow direction identifiers, and interaction rule attributes based on the tracked data flow paths and information interaction rules between nodes, and determining the directed graph structure as the data flow topology of the target main distribution network project throughout its entire lifecycle.

3. The digital management method for main and distribution network projects as described in claim 1, characterized in that, The step of extracting front-end node status layer information and back-end engineering monitoring layer information from the data flow topology based on the front-end-back-end separation strategy includes: performing layer separation processing on the data flow topology, assigning topology elements related to the running status of engineering stage nodes to the front-end node status layer, and assigning topology elements related to engineering resource allocation and environmental monitoring to the back-end engineering monitoring layer; performing attribute parsing on the topology elements in the front-end node status layer, extracting status identifier information, running progress parameters, and abnormal early warning thresholds for each engineering stage node, and generating front-end node status layer information containing node status parameters and status change time series; performing data association processing on the topology elements in the back-end engineering monitoring layer, establishing the correspondence between engineering resource allocation data and engineering stage nodes, and the spatial mapping relationship between environmental impact factor data and engineering implementation area; and extracting spatiotemporal features from the back-end engineering monitoring data based on the correspondence and spatial mapping relationship to generate a package. The system includes backend engineering monitoring layer information containing resource utilization efficiency indicators and environmental impact assessment parameters; it also includes: determining whether there are missing key node status parameters or discontinuous status change time series in the extracted frontend node status layer information; if there are missing key node status parameters, supplementary predictions are made based on the historical status change patterns of nodes in this engineering stage and the status characteristics of nodes in similar engineering stages; if there are discontinuous status change time series, smooth transition processing is performed based on the trends before and after the time series; determining whether the timestamps of engineering resource allocation data and environmental impact factor data in the extracted backend engineering monitoring layer information are synchronized, and whether the data measurement units are consistent; if the timestamps are not synchronized, time axis alignment adjustment is performed based on key time nodes of engineering implementation; if the measurement units are not consistent, they are converted to a preset standard measurement unit system; and storing the frontend node status layer information and the backend engineering monitoring layer information that have completed the judgment in association.

4. The digital management method for main and distribution network projects as described in claim 3, characterized in that, The process of performing data association processing on the topology elements in the backend engineering monitoring layer to establish the correspondence between engineering resource allocation data and engineering stage nodes, as well as the spatial mapping relationship between environmental impact factor data and engineering implementation area, includes: traversing all topology elements in the backend engineering monitoring layer, identifying resource node elements representing engineering resource allocation data and environmental node elements representing environmental impact factor data; extracting the resource type attributes and resource allocation time period information of the resource node elements, and extracting the resource demand type and resource usage time period parameters of each engineering stage node; matching the resource type attributes with the resource demand types, calculating the overlap of the resource allocation time period information and the resource usage time period parameters in the time dimension, and establishing the correspondence between engineering resource allocation data and engineering stage nodes based on the matching results and overlap calculation results; extracting the environmental parameter collection location coordinates and environmental impact range description information of the environmental node elements, and obtaining the geographical boundary coordinates of each implementation area of ​​the target main distribution network project; calculating the spatial attribution relationship based on the environmental parameter collection location coordinates and geographical boundary coordinates, determining the impact weight of environmental impact factors on different implementation areas in combination with the environmental impact range description information, and establishing the spatial mapping relationship between environmental impact factor data and engineering implementation area based on the spatial attribution relationship and impact weight.

5. The digital management method for main and distribution network projects as described in claim 1, characterized in that, The process involves utilizing the front-end node status layer information and the back-end project monitoring layer information to construct a multi-dimensional influencing factor association network for risk evolution path deduction, thereby simulating risk events for the target main distribution network project. This yields the simulation results of the risk events corresponding to the target main distribution network project. Based on these simulation results, a cross-node collaborative response strategy is generated. This includes: associating and fusing node operating status parameters from the front-end node status layer information with resource configuration data and environmental influencing factor data from the back-end project monitoring layer information to construct a multi-dimensional influencing factor association network encompassing node status, resource, and environmental dimensions; setting risk triggering conditions and influence transmission rules within the multi-dimensional influencing factor association network; and generating different initial risk triggering conditions through random sampling. The system sets a set of risk event evolution paths under initial conditions; it assesses the impact of each risk event evolution path, calculates the risk loss value and risk diffusion rate parameter of each engineering stage node in the path, and determines the severity level of the risk event based on the risk loss value and risk diffusion rate parameter; it performs cluster analysis on risk event evolution paths with the same severity level and similar evolution characteristics, extracts typical evolution patterns and key impact nodes for each category of risk events, and generates risk event simulation results including risk event type, probability of occurrence, scope of impact, and severity level; based on the risk event type and severity level in the risk event simulation results, combined with the business attributes of key impact nodes, it creates a cross-node collaborative response strategy that includes resource scheduling priority, node collaboration order, and handling process specifications.

6. The digital management method for main and distribution network projects as described in claim 5, characterized in that, The process of setting risk triggering conditions and influence transmission rules in the multi-dimensional influencing factor association network, and generating a set of risk event evolution paths under different initial condition combinations through random sampling, includes: for each engineering stage node in the multi-dimensional influencing factor association network, setting a risk triggering threshold corresponding to that engineering stage node based on historical engineering data and prior knowledge, wherein the risk triggering threshold includes node operating status parameter threshold, resource allocation threshold, and environmental influence factor threshold; when the actual parameter value of the engineering stage node exceeds the corresponding risk triggering threshold, determining that a risk event has occurred at that engineering stage node, and recording the occurrence time and initial impact degree of the risk event; defining an influence transmission coefficient between engineering stage nodes in the multi-dimensional influencing factor association network, wherein the influence transmission coefficient characterizes the influence of a risk event occurring at an engineering stage node on its associated nodes. The probability and degree of influence generated by a point; an influence transmission matrix is ​​constructed based on the influence transmission coefficient, where the element values ​​represent the quantitative values ​​of the probability and degree of influence transmission from one engineering stage node to another; the risk trigger threshold in the multi-dimensional influence factor association network is randomly perturbed multiple times, with each perturbation generating a new set of initial condition combinations; under each set of initial condition combinations, the initial risk event node is determined according to the risk trigger condition, and then, based on the influence transmission rules and the influence transmission matrix, the diffusion process of the risk event in the network is gradually deduced from the initial risk event node, recording the node sequence, occurrence time, and degree of influence of the risk event, forming a risk event evolution path; the random sampling and deduction process is repeated multiple times until a preset number of risk event evolution paths are generated, constituting a risk event evolution path set.

7. The digital management method for main and distribution network projects as described in claim 5, characterized in that, The assessment of the impact of each risk event evolution path, calculating the risk loss value and risk diffusion speed parameters of each engineering stage node in the path, and determining the severity level of the risk event based on the risk loss value and risk diffusion speed parameters, includes: identifying the node impact chain for each risk event evolution path, extracting all engineering stage nodes directly and indirectly affected by the risk event in the risk event evolution path, and constructing a risk impact hierarchy structure including a direct impact node layer and an indirect impact node layer; identifying the core business interruption mode of the direct impact node layer in the risk impact hierarchy structure, identifying the degree of business function loss of each directly impact node and the type of blockage to related business processes, and determining the risk transmission path and transmission intensity characteristics of the directly impact nodes; based on the risk transmission path and transmission intensity characteristics, extrapolating the scope of impact for the indirect impact node layer, identifying the potential secondary risk event types and risk coupling effect modes in the indirect impact nodes; and constructing a risk impact hierarchy structure based on the degree of business function loss of the directly impact nodes, the secondary risk event types of the indirect impact nodes, and the risk coupling effect modes. A qualitative assessment matrix for the degree of risk impact is constructed. The row dimension of the matrix represents the set of directly affected nodes, the column dimension represents the set of indirectly affected nodes, and the matrix elements represent qualitative descriptions of the risk impact between corresponding nodes. This matrix is ​​then subjected to hierarchical weighting. Based on the core business weights of directly affected nodes and the ripple range weights of indirectly affected nodes, a comprehensive description vector of the degree of risk event impact is generated. A time-series analysis is performed on the node impact sequence in the risk event evolution path to identify the propagation time sequence characteristics of the risk event from the initial node to nodes at each level, extracting bottleneck nodes and accelerating nodes in the propagation process. Based on the propagation time sequence characteristics, bottleneck nodes, and accelerating nodes, a risk diffusion pattern map is constructed. This risk diffusion pattern map includes qualitative descriptions of the critical path of risk diffusion, node obstruction coefficients, and node acceleration coefficients. Combining the comprehensive description vector of the degree of risk event impact and the risk diffusion pattern map, the risk event is qualitatively graded in multiple dimensions to determine its severity level. The severity level is determined comprehensively based on the number of impact levels, the type of core business obstruction, and the complexity of the diffusion pattern.

8. The digital management method for main and distribution network projects as described in claim 5, characterized in that, The process involves creating a cross-node collaborative response strategy based on the risk event types and severity levels from the risk event simulation results, combined with the business attributes of key impact nodes. This strategy includes: deconstructing the risk event types from the simulation results to extract triggering condition features, impact range features, and evolution trend features, and constructing a risk event feature vector library; identifying the core capability types and resource reserve status of each key impact node during the risk handling process based on the business attributes of the key impact nodes, and establishing a node capability-resource mapping relationship model; matching the risk event feature vector library with the node capability-resource mapping relationship model to determine the set of candidate response nodes that have a handling effect on the risk event type and the response capability weight of each candidate node; constructing a dynamic selection mechanism for response nodes based on the severity level of the risk event and the response capability weight of the candidate response nodes, automatically adjusting the set of candidate response nodes and the allocation ratio of response capability weights when the severity level of the risk event changes; and based on the candidate response nodes' business attributes, establishing a node capability-resource mapping relationship model. Based on the node set and response capability weights, a cross-node collaborative network topology is constructed, which includes information interaction channels and resource sharing paths between nodes. Within this topology, the node collaboration order is determined according to the evolutionary trend characteristics of risk events and the geographical distribution attributes of nodes. This order follows a joint sorting rule: risk impact area from near to far, and response capability weight from high to low. For the determined node collaboration order, a handling process framework is created, including information transmission protocols, resource handover specifications, and decision chain levels. This framework sets decision thresholds and information feedback mechanisms for key nodes. Based on the impact range characteristics of risk events and the node collaboration order, resource scheduling priority rules are embedded in the handling process framework. These rules are dynamically adjusted based on the impact of risk events on nodes at each project stage and the urgency of node business recovery. The handling process framework and resource scheduling priority rules are instantiated to generate a cross-node collaborative response strategy that includes a node collaboration sequence diagram, resource scheduling path planning, and a decision authority allocation table.

9. A digital management system for main and distribution network projects, characterized in that, include: processor; A storage device storing a computer program; a network interface for providing network communication functions; when the computer program is executed by the processor, the processor implements the digital management method for main and distribution network projects as described in any one of claims 1-8.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a processor, implement the digital management method for main and distribution network projects as described in any one of claims 1-8.

Images