A method for dynamic identification of target deviation in different stages of reservoir EPC project

Abstract

The present application relates to the technical field of project management analysis, and particularly relates to a method for dynamic identification of target deviation in different stages of a reservoir EPC project, comprising the following steps: constructing a three-stage correlated multiple regression model, and configuring a dynamic influence coefficient for each EPC stage node; collecting real-time operation data of each stage, using the dynamic influence coefficient to perform stage weight distribution on the data, and generating a node state vector including a stage identifier; establishing a Markov transition chain according to the stage identifier of the node state vector, calculating the deviation transition probability between adjacent stages, and when the deviation transition probability exceeds an alarm threshold, triggering a reverse stage compensation instruction and updating the dynamic influence coefficient of S1, to form a closed-loop regulation and control. The present application has stronger environmental adaptability and real-time performance in the scene of large terrain undulation and complex conduction path in reservoir projects, and improves the stability and robustness of the deviation chain diagnosis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of project management analysis, and particularly relates to a method for dynamically identifying target deviation in different stages of a reservoir EPC project. BACKGROUND

[0002] As a major infrastructure project, a reservoir project generally adopts an EPC general contracting mode of design, procurement and construction integration to improve overall progress coordination and resource allocation efficiency. In this mode, the target connection relationship between different stages is highly coupled, and the deviation of any stage may cause cumulative amplification effect on the subsequent stage, thereby affecting the quality, safety and schedule control of the entire project.

[0003] In the prior art, the engineering deviation management is mainly based on static progress plan and single-stage data monitoring, and lacks cross-stage information fusion mechanism and dynamic response capability. On the one hand, design change information cannot be fed back to procurement and construction plans in time; on the other hand, the delay or quality problems generated in the procurement and construction links lack traceable source identification mechanism, which leads to delayed problem response and difficult responsibility determination, affecting the efficiency and accuracy of target closed-loop control. SUMMARY

[0004] The present application provides a method for dynamically identifying target deviation in different stages of a reservoir EPC project, which combines multi-stage key parameters, identifies deviation transmission path and has dynamic feedback adjustment capability, and is suitable for realizing engineering whole-process, cross-link coordinated regulation and risk suppression in complex reservoir projects.

[0005] A method for dynamically identifying target deviation in different stages of a reservoir EPC project, comprising the following steps:

[0006] S1, stage target correlation modeling:

[0007] Based on the design stage BIM parameters, the procurement stage material list and the construction stage process plan, a multivariate regression model associated with three stages is constructed, and a dynamic influence coefficient is configured for each EPC stage node, wherein the dynamic influence coefficient is calculated and generated according to the cross-stage deviation transmission law in historical data;

[0008] S2, multi-stage data fusion and state quantization:

[0009] Real-time acquisition of operation data of each stage, including time difference of material arrival at procurement node, mechanical efficiency value at construction node and parameter change frequency at design node, stage weight distribution of data is carried out by using the dynamic influence coefficient, and a node state vector including stage identifier is generated;

[0010] S3: cross-stage deviation propagation identification and closed-loop control:

[0011] According to the phase identifier of the node state vector, a Markov transition chain is established, the deviation transition probability between adjacent phases is calculated, when the transition probability of design procurement or procurement construction exceeds the warning threshold, the reverse phase compensation instruction is triggered and the dynamic influence coefficient of S1 is updated, forming a closed loop control.

[0012] Optionally, the S1 specifically comprises:

[0013] S11, extracting the physical parameter set of the component in the design phase BIM model, including concrete strength grade, steel bar specification and impermeable coefficient, generating a design feature matrix with a unique component ID;

[0014] S12, matching the material technical parameters in the procurement material list with the design feature matrix, establishing a component ID-material code association mapping table, calculating the influence weight of material parameter deviation on design parameter based on historical supply data, denoted as procurement influence factor ;

[0015] S13, analyzing the process constraint conditions in the construction process plan, identifying the key construction path strongly associated with the design component, establishing a multivariate regression model of construction delay days, design parameter deviation and material arrival delay based on historical construction log, and outputting the construction influence factor ;

[0016] S14, introducing the environmental sensitivity coefficient, calculating the dynamic influence coefficient across stages.

[0017] Optionally, the construction influence factor is obtained by constructing a multivariate regression model, represented as:

[0018] , represents the actual delay days of the component on the key path; is the mean value of the parameter deviation caused by procurement, defined as: , is the regression coefficient obtained by model training, is the construction influence factor of the component , reflecting the comprehensive influence of construction plan deviation on design target, is the procurement influence factor of the component .

[0019] Optionally, the dynamic influence coefficient is represented as:

[0020] ;

[0021] wherein, is the dynamic influence coefficient of the component in the current cycle, is the upper cycle influence coefficient, is the procurement and construction influence factor calculated in the current cycle.

[0022] Optionally, the collection of the operation data of each stage in S2 specifically includes:

[0023] The design node extracts the parameter change operation in the record by calling the version management interface of the building information model, and within the set time window, counts the number of change events of all design parameters to measure the dynamic change frequency of the design link;

[0024] The procurement node accesses the material transportation scheduling platform, analyzes the geographic trajectory data of the transportation vehicle, extracts the time difference between the actual arrival time and the scheduled plan, and at the same time, combined with the acceptance information based on the blockchain, excludes records with failed verification or abnormalities, and only keeps valid data;

[0025] The construction node obtains the work quantity data in a unit of time through the engineering machinery controller, and combines the rated power information of the equipment to calculate the actual work efficiency in the current working state as the mechanical efficiency value of the construction node.

[0026] Optionally, the stage weight distribution specifically includes:

[0027] According to the extracted procurement influence factor and the environmental sensitivity coefficient, the design node is given a corresponding fusion weight to reflect the sensitivity of the design link to the downstream deviation;

[0028] According to the construction influence factor and the cross-stage dynamic image coefficient, the procurement node is configured with a fusion weight to quantify the amplification or buffering effect of the intermediate link in the target deviation transmission;

[0029] The fusion weight of the construction node is determined by the sum of the weights of the design and procurement nodes, so that the overall weight is evenly distributed among the three stages, and the data fusion is closed in quantization.

[0030] Optionally, the design node vector includes its stage identifier, weighted parameter change frequency indicator, and corresponding component number; the procurement node vector includes stage identifier, weighted arrival time difference indicator, and its material batch number; the construction node vector includes stage identifier, weighted work efficiency indicator, and associated process code, and S2 further includes vectorization coding, which respectively weights and corrects the data of the three nodes, and then structures the expression, and adds stage identifier and accessory information to form the node state vector.

[0031] Optionally, the establishment of the Markov transition chain in S3 includes defining a set of state transition directions according to the stage identifier in the node state vector, , and constructing a binary transition probability matrix, wherein:

[0032] D→P transition probability , the proportion of the number of times of the procurement node anomaly after the design node anomaly in the statistical historical data;

[0033] P→C transition probability , the proportion of the number of times of the construction node anomaly after the procurement node anomaly.

[0034] Optionally, the S3 further comprises setting of an alert value: based on the reservoir area environment complexity index Set the alert threshold , the more complex the terrain , the greater the value, when Or , the generation of the reverse compensation instruction is activated.

[0035] Optionally, the generation of the reverse compensation instruction comprises:

[0036] If the D→P probability is over-standard, a parameter optimization instruction is sent to the design node, and the weight of the procurement influence factor a is reduced synchronously;

[0037] If the P→C probability is over-standard, a construction resource re-allocation instruction is triggered, and the construction influence factor is updated based on the actual delay days ;

[0038] The compensated actual deviation data is input into the multivariate regression model of S1, the dynamic influence coefficient is recalculated, and a cross-stage regulation and control closed loop is formed.

[0039] The beneficial effects of the present application are:

[0040] The present application can obtain the state change of the key node in real time and vectorize the structured state code by constructing a design, procurement and construction three-stage associated multivariate regression model, combining multi-source data fusion and dynamic weight distribution strategy. Compared with the traditional single-stage progress tracking mode, the present application can identify the potential deviation node in the cross-stage conduction chain, realize the change from "lag response" to "feedforward warning", and improve the deviation identification accuracy and intervention timeliness in the EPC process.

[0041] The present application fuses the terrain complexity index K calculated by the geographic information system, adaptively adjusts the alert threshold of the deviation transition, cooperates with the sparse Markov chain model of only retaining the D→P→C path, reduces the false positive rate and the operation complexity, and has stronger environmental adaptability and real-time performance in the scene of large terrain undulation and complex conduction path in the reservoir project, and improves the stability and robustness of the deviation chain diagnosis.

[0042] This invention activates reverse compensation commands through deviation propagation analysis, and links the influencing factors and regression models in the S1 stage to perform incremental coefficient updates, forming an adaptive closed-loop path of "anomaly identification - compensation adjustment - model reconstruction". Compared with the static target assessment mechanism in the traditional EPC system, this invention has the ability to dynamically adjust in the scenario of continuous evolution of target deviation, which can effectively suppress system oscillation and maintain the consistency and controllability of the target throughout the process. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Fig. 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0045] Fig. 2 This is a schematic diagram of multi-stage data fusion and state quantization in an embodiment of the present invention. Detailed Implementation

[0046] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. For some well-known technologies, those skilled in the art may also use other alternative methods to implement the invention. Moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0047] like Figs. 1-2 As shown, a method for dynamic identification of target deviations at different stages of a reservoir EPC project includes the following steps:

[0048] S1, Stage Goal Association Modeling:

[0049] Based on BIM parameters in the design phase, material list in the procurement phase, and work plan in the construction phase, a multivariate regression model is constructed to link the three phases. A dynamic influence coefficient is configured for each EPC phase node. The dynamic influence coefficient is calculated and generated based on the cross-phase deviation transmission law in historical data.

[0050] S2, Multi-stage data fusion and state quantization:

[0051] Real-time collection of operational data at each stage, including the time difference of material arrival at procurement nodes, mechanical efficiency values ​​at construction nodes, and the frequency of parameter changes at design nodes; using dynamic influence coefficients to assign stage weights to the data and generate node state vectors including stage identifiers.

[0052] S3: Cross-stage deviation propagation identification and closed-loop control:

[0053] According to the phase identifier of the node state vector, a Markov transition chain is established, the deviation transition probability between adjacent phases is calculated, when the transition probability of design→ procurement or procurement→ construction exceeds the warning threshold, the reverse phase compensation instruction is triggered and the dynamic influence coefficient of S1 is updated, forming a closed loop control.

[0054] S1 specifically includes:

[0055] S11, extract the physical parameter set of the component in the design phase BIM model, including concrete strength grade, steel bar specification and impermeability coefficient, generate a design feature matrix with unique component ID, and the design feature matrix is constructed as follows:

[0056]

[0057] wherein, is the unique identifier of the i-th component, represents the concrete strength grade (C30, C35, etc.), is the steel bar specification, is the impermeability coefficient, is the total number of components.

[0058] S12, match the material technical parameters in the procurement material list with the design feature matrix, establish the component ID material code correlation mapping table: and match the design parameters and procurement parameters of each component, construct the deviation matrix: ; based on the historical supply data, calculate the influence weight of material parameter deviation on design parameter, denoted as procurement influence factor , further calculate the procurement influence factor : ; wherein, is the actual supply value of the corresponding component parameter in the procurement list, i.e. the supply value of the i-th technical parameter matched by the i-th component in the procurement material list, is the design parameter value, is the relative deviation, is the influence weight of the i-th parameter, satisfying , determined according to simulation or expert scoring method, is the procurement influence factor of the component . i represents a component (uniquely identified by the component ID in the BIM model);

[0059]

[0060] ​​​​j represents the jth design technical parameter related to the component, for example:

[0061] j = 1: concrete strength grade (for example, the design value is C30, and the corresponding procurement supply is the compressive strength MPa value, such as 32.5 MPa);

[0062] j = 2: diameter and yield strength of steel bar (for example, the design is HRB400-16, and the supply is marked as HRB400 with a measured yield strength of 420 MPa);

[0063] j = 3: impermeability (for example, the design impermeability grade is P6, the procurement is a certain brand of water stop material, and the corresponding permeability coefficient is obtained);

[0064] S13, analyze the process constraints in the construction process plan, identify the key construction path closely related to the design component, and establish a multiple linear regression model of construction delay days, design parameter deviation, and material arrival delay based on historical construction logs, and output construction influence factors First, extract the key path delay information and historical records related to the component, including the actual delay days on the key path of the component The mean value of parameter deviation caused by procurement is defined as: The construction influence factor is fitted using a multiple linear regression model : wherein, is the regression coefficient obtained by model training, satisfying the normalization condition , based on the historical project log data set, is the construction influence factor of the component , reflecting the comprehensive influence of construction plan deviation on design target.

[0065] S14, introduce the environmental sensitivity coefficient When the temperature in the warehouse area is , otherwise , defined as follows:

[0066] In the first period, the cross-stage dynamic influence coefficient of the component is dynamically updated:

[0067] ;

[0068] wherein, is the dynamic influence coefficient of the component in the current period, is the influence coefficient of the previous period, and the initial value is ​Determined by engineering type lookup table, such as "dam type = 0.6, diversion tunnel type = 0.4", The procurement and construction impact factors calculated in the current period.

[0069] S2 specifically includes:

[0070] S21, real-time collection of multi-source data:

[0071] S211, design node data collection: through the version control interface of the BIM platform, capture parameter change events, and within a unit time window count the frequency of changes in design parameters, defined as: ; wherein, is the frequency of parameter changes in the design node, is the total number of design parameter change events within the time window, is the duration of the statistical period;

[0072] S212, procurement node data collection: analyze the GPS trajectory data of the material transportation system to obtain the actual arrival time and the planned time , and filter abnormal data combined with the blockchain inspection record to calculate the net arrival delay: ;

[0073] S213, construction node data collection: read the work volume within a unit of working hours from the engineering machinery controller , and calculate the mechanical efficiency based on the rated power : ;

[0074] S22, dynamic weight allocation: according to the key factors obtained in S1, assign dynamic weights to the three types of nodes:

[0075] Design node weight: ;

[0076] Procurement node weight: ;

[0077] Construction node weight: ;

[0078] wherein, is the procurement impact factor, is the construction impact factor, is the environmental sensitivity coefficient, is the cross-stage dynamic impact coefficient, respectively represent the fusion weights of the design, procurement, and construction nodes, if , then automatic normalization processing ensures that the total weight is 1.

[0079] S23, vectorization coding: encode the data of the three stages respectively according to the weight and attach the identification information to form the node state vector of unified format.

[0080] Design node vector: ;

[0081] Purchase node vector: ;

[0082] Construction node vector: ;

[0083] Wherein, is the stage identifier, is the component ID, is the material batch number, is the process code, the second item in the vector is the fused weighted feature value, which is used for subsequent state sequence processing or deviation identification modeling.

[0084] Design node weight In is the purchase influence factor, indicating the feedback degree of purchase error to the design target, is the environmental sensitivity coefficient, reflecting the amplification effect of high temperature and other external factors on design change. The combination of the two factors represents the indirect impact pressure of "actual deviation of the procurement link + environmental risk" on the design target. Therefore, the weight of the design node is set as the product of the two factors, which reflects the upstream sensitivity of its deviation transmission.

[0085] Purchase node weight In, is the construction influence factor, indicating the influence of construction delay or low efficiency on procurement rhythm and material effectiveness, is the cross-stage dynamic influence coefficient, reflecting the accumulation and transmission trend of deviation in the historical period. The product of the two represents "the degree of dependence of construction on procurement x the coupling strength of historical stages, so the weight of the procurement node is set to this to reflect the core regulation of the upstream coordination function to the overall deviation.

[0086] Construction node weight In, the remaining all weights are automatically assigned to the construction node, because the construction stage is the final executor, and its deviation is more of a result and landing. When the design and procurement risk is high, the interference weight of the construction end data should be reduced accordingly to maintain the balance of the overall fusion. In this way, the design can ensure that the total weight of the three nodes is 1, and the construction stage assumes the role of "dynamic self-balancing".

[0087] The weight design logic is to dynamically allocate according to the risk transmission path and the actual error action direction, so that each stage assumes the data fusion weight corresponding to its "influence degree" in deviation identification.

[0088] S3 specifically comprises:

[0089] S31, Markov transition chain modeling: based on the phase identifier in the node state vector, a transition state set is constructed:

[0090] The transition probability matrix of phase abnormal conduction is constructed, defined as follows:

[0091] Design The transition probability of procurement:

[0092] Procurement The transition probability of construction:

[0093] Wherein, is the probability of design node abnormality leading to procurement node abnormality, is the probability of procurement node abnormality leading to construction node abnormality, represents the number of times of procurement node abnormality immediately following design node abnormality in the historical record, represents the total number of times of design node or procurement node abnormality in the history.

[0094] S32, dynamic alarm value setting: according to the terrain complexity index of the warehouse area Adaptive adjustment of alarm threshold: Wherein, is the early warning trigger threshold of the current period, is the terrain complexity index, which is calculated by geographic information system according to slope, fault density, ground relief, etc. The greater the value, the more complex the environment, and the trigger judgment condition is: ; any condition is met, the reverse compensation mechanism is activated.

[0095] S33, reverse compensation instruction generation:

[0096] S331, if the transition probability from design to procurement exceeds the threshold value, then:

[0097] Parameter optimization instructions are issued to the design node;

[0098] Synchronization update of procurement influence factor, reduce its weight coefficient in the following way: Wherein is the adjustment coefficient, set to 0.8.

[0099] S332, if the transition probability from procurement to construction exceeds the threshold value, then:

[0100] Trigger construction resource reallocation;

[0101] ​​​​And according to the following increment model corrects the construction influence factor: Wherein, The actual delay days of the current construction node, The adjusted construction influence factor.

[0102] S34, the above compensation adjusted feedback data is re-input into the multiple regression model established in S1, based on the latest design deviation, procurement difference and construction delay, re-calculate the cross-stage dynamic influence coefficient So as to realize the deviation conduction identification Compensation feedback The closed-loop control mechanism of model recalculation.

[0103] The state transition is limited to the two directions of design→purchase and purchase→construction, a sparse Markov transition model is constructed, invalid state traversal paths are reduced, the real-time and efficiency of abnormal identification are improved, the terrain complexity index is introduced as an adjustment parameter, the alarm threshold is dynamically reduced under the complex geological conditions of reservoir construction environment, the false alarm and missed alarm problems are alleviated, and the adaptability of the model under extreme environment is enhanced. In the construction deviation compensation mechanism, the smooth correction coefficient increasing with the delay time is introduced, the parameter mutation is avoided, the shock of the scheduling system is caused, and the stability of the target control chain of each stage is maintained.

[0104] The present application covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can also be fully understood without the description of these details to those skilled in the art. In addition, in order to avoid unnecessary confusion to the essence of the present application, well-known methods, processes, procedures, elements and circuits are not described in detail.

[0105] The above is only the preferred embodiment of the present application, it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can also be made, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for dynamic identification of target deviation in different stages of a reservoir EPC project, characterized in that, Comprise the following steps: S1, stage target correlation modeling: Based on the design stage BIM parameters, procurement stage material list, construction stage process plan, a three-stage correlated multiple regression model is constructed, and a dynamic influence coefficient is configured for each EPC stage node, which is calculated and generated according to the cross-stage deviation conduction law in historical data; Specifically, it includes: S11, extracting the physical parameter set of the component in the design stage BIM model, including concrete strength grade, reinforcement specification and impermeability coefficient, and generating a design feature matrix with a unique component ID; S12, match the material technical parameters in the procurement material list with the design feature matrix, establish a component ID→material code correlation mapping table, calculate the influence weight of material parameter deviation on design parameters based on historical supply data, denoted as procurement influence factor ; S13, analyze the process constraints in the construction process plan, identify the key construction path strongly associated with the design component, establish a multiple regression model of construction delay days, design parameter deviation, and material delay according to the historical construction log, and output the construction influence factor ; the construction influence factor Obtained by constructing a multiple regression model, represented as: , represents the actual delay days on the critical path of the component; is the mean value of the parameter deviation caused by procurement, defined as: , is the regression coefficient obtained by model training, is the construction impact factor of the component , reflecting the comprehensive impact of construction plan deviation on the design target, is the procurement impact factor of the component ; S14, introducing the environmental sensitivity coefficient calculating the dynamic influence coefficient across stages, specifically comprising: When the library temperature is greater than 0, else is defined as follows: In the first cycle, the dynamic influence coefficient of the component across the stages is dynamically updated: ； wherein, is the component is the dynamic influence coefficient of the current period, is the influence coefficient of the previous period, , is the procurement and construction influence factor calculated in the current period; S2, multi-stage data fusion and state quantization: Real-time collection of operation data in each stage, including procurement node material arrival time difference, construction node mechanical efficiency value, design node parameter change frequency, using the dynamic influence coefficient to distribute the stage weight, and generating a node state vector including stage identifier; The stage weight distribution specifically includes: According to the extracted procurement influence factor and environmental sensitivity coefficient, the design node is given the corresponding fusion weight, reflecting the sensitivity of the design link to the downstream deviation; According to the construction influence factor and the cross-stage dynamic influence coefficient, the fusion weight of the procurement node is configured to quantify the amplification or buffering effect of the intermediate link in the target deviation transmission; The remaining value of the construction node fusion weight is determined by the sum of the design and procurement node weights, so that the overall weight is evenly distributed among the three stages, and the data fusion is closed in quantization; S3: Cross-stage deviation propagation identification and closed-loop control: A Markov transition chain is established according to the phase identifier of the node state vector, deviation transition probabilities between adjacent phases are calculated, when the transition probability of design→procurement or procurement→construction exceeds an alarm threshold, a reverse phase compensation instruction is triggered and the dynamic influence coefficient of S1 is updated, forming a closed-loop regulation; the establishment of the Markov transition chain includes defining a state transition direction set according to the phase identifier in the node state vector, , a binary transition probability matrix is constructed, wherein: D→P transition probability The number of times of purchasing node abnormalities after the design node abnormalities in the statistical historical data is counted. P→C transition probability The proportion of the number of times of the construction node exception after the statistics procurement node exception wherein, is the probability that a design node exception leads to a procurement node exception, is the probability that a procurement node exception leads to a construction node exception, represents the number of times in the history that a design node exception was followed by a procurement node exception, , represents the total number of design node or procurement node exceptions in the history; The S3 further comprises setting of an alert value: based on the warehouse environment complexity index Setting an alert threshold Wherein, the more complex the terrain The greater the value, when Or The generation of the reverse compensation instruction is activated; The generation of the reverse compensation instruction includes: If the D→P probability is out of standard, send parameter optimization instruction to the design node, and reduce the weight of the procurement influence factor simultaneously ; If the P→C probability exceeds the standard, a construction resource reallocation instruction is triggered, and the construction influence factor is updated based on the actual delay days ​ Input the actual deviation data after compensation into the multiple regression model of S1, recalculate the dynamic influence coefficient, and form a cross-stage regulation and control closed loop.

2. The method according to claim 1, characterized in that, The collection of operation data in each stage in S2 specifically includes: The design node calls the version management interface of the building information model to extract the parameter change operation in the record, and within the set time window, the number of all design parameter change events is counted to measure the dynamic change frequency of the design link; The procurement node accesses the material transportation scheduling platform, analyzes the geographic trajectory data of the transportation vehicle, extracts the time difference between the actual arrival time and the scheduled plan, and at the same time, combined with the acceptance information based on blockchain, excludes records with failed verification or abnormalities, and only keeps valid data; The construction node obtains the work quantity data in unit time through the engineering machinery controller, and calculates the actual work efficiency in the current work state combined with the rated power information of the equipment as the construction node mechanical efficiency value.

3. The method according to claim 1, characterized in that, The design node vector includes its stage identifier, weighted parameter change frequency index and corresponding component number; The procurement node vector includes stage identifier, weighted arrival time difference index and its material batch number; The construction node vector includes stage identifier, weighted work efficiency index and associated process code, and S2 further includes vectorization coding, which respectively weights and corrects the data of the three nodes, and adds stage identifier and attached information to form the node state vector.

Images