Pumped storage power station infrastructure period AI dominated multi-dimensional management system

Abstract

The invention relates to the technical field of building management, in particular to a pumped storage power station infrastructure period AI dominant multi-dimensional management system, which comprises a first acquisition unit used for acquiring initial geological survey data which is geological feature record data of a target infrastructure area before an infrastructure process is carried out; the second acquisition unit is used for acquiring process implementation record data; and the third acquisition unit is used for acquiring process acceptance check data, and the process acceptance check data is acceptance record data of the reconstruction effect of the target capital construction area after the preset capital construction process is completed. According to the invention, the multi-dimensional data of the capital construction area is comprehensively acquired through the three acquisition units, and the accurate control of capital construction projects can be ensured through the multi-angle and real-time data acquisition mode, so that the construction risk caused by incomplete or inaccurate data is reduced.

Description

Technical Field

[0001] This invention relates to the field of building management technology, specifically to an AI-led multidimensional management system for the infrastructure construction phase of pumped storage power stations. Background Technology

[0002] Pumped storage power stations are large-scale infrastructure projects involving multiple complex stages such as geological exploration, civil engineering, mechanical equipment installation, and power system connection. Traditional management models often lack efficient real-time data collection and feedback mechanisms, which can easily lead to slow project progress, waste of resources, and construction quality problems.

[0003] Currently, traditional systems often rely on manual recording and offline data collection, which results in incomplete data collection and poor real-time performance. This may fail to reflect the actual situation on site in a timely manner, and this approach is prone to data omissions or errors, thereby affecting project decisions and progress. In addition, it usually relies on human judgment and experience, lacking automated evaluation and correction mechanisms. Manual analysis is not only inefficient, but may also introduce errors due to human factors. When dealing with deviations in the construction process, it often requires multiple manual checks and adjustments, which is inefficient and makes it difficult to detect potential problems in a timely manner.

[0004] Furthermore, traditional systems tend to have a rather coarse assessment of the suitability of infrastructure projects, typically relying on limited historical data and fixed standards. They lack dynamic adaptation assessments tailored to specific projects and real-time conditions. This can lead to situations during construction that do not meet geological conditions or construction specifications, resulting in rework or construction delays. Moreover, the control plans are generally static and lack flexibility, failing to make rapid adjustments based on changes in real-time data. This may result in the inability to update the plans or adjust implementation details in a timely manner when new and complex problems arise during construction, thereby affecting the quality and schedule of the project. Summary of the Invention

[0005] To achieve the above objectives, the present invention provides the following technical solution: an AI-led multi-dimensional management system for the infrastructure construction phase of a pumped-storage power station, comprising:

[0006] The first acquisition unit is used to acquire initial geological survey data, wherein the initial geological survey data is geological feature record data of the target infrastructure area before the commencement of infrastructure construction procedures;

[0007] The second acquisition unit is used to acquire process implementation record data, wherein the process implementation record data is the process record data after the preset infrastructure process is carried out in the target infrastructure area and the preset infrastructure process has completed the modification of the initial geological conditions;

[0008] The third acquisition unit is used to acquire process acceptance verification data, wherein the process acceptance verification data is the acceptance record data of the transformation effect of the target infrastructure area after the completion of the preset infrastructure process;

[0009] The control and assessment unit is used to compare the control and assessment fit of the target infrastructure area in the first control and assessment result with the control and assessment fit of the target infrastructure area in the second control and assessment result; the first control and assessment result is obtained by performing difference analysis on the process acceptance verification data and the process implementation record data; the second control and assessment result is obtained by performing difference analysis on the process implementation record data and the initial geological survey data;

[0010] The infrastructure management unit is used to determine the infrastructure management plan for the target infrastructure area based on the first management assessment result, when the management adaptability of the target infrastructure area in the first management assessment result is greater than the management adaptability of the target infrastructure area in the second management assessment result.

[0011] Preferably, the method for calculating the control adaptability of the target infrastructure area includes:

[0012] The target infrastructure area is assessed for control based on a pre-trained infrastructure control assessment model to obtain the control suitability of the target infrastructure area.

[0013] Preferably, determining the infrastructure management and control plan for the target infrastructure area based on the first management and control assessment results includes:

[0014] Compare the control indicator parameters of the target infrastructure area in the first control assessment result with the control indicator parameters of the target infrastructure area in the second control assessment result;

[0015] If there is a deviation between the control indicator parameters of the target infrastructure area in the first control assessment result and the control indicator parameters of the target infrastructure area in the second control assessment result, the first control assessment result shall be corrected based on the second control assessment result.

[0016] The revised first control assessment result is determined as the infrastructure control plan for the target infrastructure area.

[0017] Preferably, the first control assessment result is corrected based on the second control assessment result, including:

[0018] Based on the second control assessment result, the first deviation location of the target infrastructure area in the first control assessment result is identified; wherein, the first deviation location is the location where the process implementation form of the target infrastructure area is abnormal; the abnormal form includes at least one of the following: pouring thickness deviation, support structure misalignment, and excavation slope deviation;

[0019] Based on the standard parameters of the target infrastructure area at the first deviation position in the second control assessment result, the first deviation position of the target infrastructure area in the first control assessment result is corrected to obtain the corrected first control assessment result.

[0020] Preferably, correcting the first control assessment result based on the second control assessment result further includes:

[0021] Based on the second control assessment result, a second deviation location is identified in the target infrastructure area mentioned in the first control assessment result; wherein, the second deviation location is the location in the target infrastructure area where there is excessive material accumulation or process implementation oversight;

[0022] Based on the standard parameters of the target infrastructure area at the second deviation position in the second control assessment result, the second deviation position of the target infrastructure area in the first control assessment result is corrected to obtain the corrected first control assessment result.

[0023] Preferably, the initial geological survey data, the process implementation record data, and the process acceptance verification data are all collected using high-precision infrastructure monitoring equipment; the high-precision infrastructure monitoring equipment includes a three-dimensional geological scanner, a stress monitor, and a material density tester.

[0024] Preferably, the training method for the pre-trained infrastructure management and assessment model includes:

[0025] Create an initial infrastructure management and control assessment model;

[0026] Obtain an infrastructure management and control sample dataset, wherein the infrastructure management and control sample dataset includes multiple infrastructure management and control sample pairs with management and control adaptability labeling information, wherein any one of the multiple infrastructure management and control sample pairs includes initial geological sample data and process implementation sample data, wherein the process implementation sample data is generated by process planning based on the geological feature parameters in the initial geological sample data;

[0027] The initial infrastructure management assessment model is iteratively trained based on the infrastructure management sample dataset.

[0028] If the preset convergence condition is met, the initial infrastructure management and control assessment model will be determined as the pre-trained infrastructure management and control assessment model.

[0029] Preferably, the initial infrastructure management and control assessment model includes a management and control feature extraction layer, a management and control difference analysis layer, and a management and control fit evaluation layer;

[0030] The control feature extraction layer is used to receive the input initial geological sample data and the process implementation sample data, and extract the control feature matrix of the initial geological sample data and the control feature matrix of the process implementation sample data, respectively.

[0031] The control difference analysis layer is used to determine the control difference information between the initial geological sample data and the process implementation sample data based on the control feature matrix of the initial geological sample data and the control feature matrix of the process implementation sample data.

[0032] The control fit evaluation layer is used to determine the control fit result of the process implementation sample data based on the control difference information between the initial geological sample data and the process implementation sample data.

[0033] Preferably, the control difference analysis layer is further configured to extract the global control feature vector of the initial geological sample data from the control feature matrix of the initial geological sample data, extract the global control feature vector of the process implementation sample data from the control feature matrix of the process implementation sample data, and determine the control difference information between the initial geological sample data and the process implementation sample data based on the global control feature vector of the initial geological sample data and the global control feature vector of the process implementation sample data.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] (1) This invention uses three acquisition units to comprehensively acquire multi-dimensional data of the infrastructure area. This multi-angle, real-time data acquisition method can ensure accurate control of the infrastructure project, thereby reducing construction risks caused by incomplete or inaccurate data. The control and evaluation unit compares the initial geological data, process implementation records and process acceptance data to obtain the control and adaptation of the infrastructure area. The automation and intelligence of this process make the adaptation assessment of the entire project more scientific and accurate, and can correct deviations in time to ensure that the geological conditions and process implementation are matched during the construction process.

[0036] (2) This invention can automatically generate or correct infrastructure management plans. This automated plan generation mechanism reduces manual intervention, improves efficiency, and can optimize the entire infrastructure process through continuous feedback and adjustment. It accurately identifies deviations in the implementation of procedures (such as pouring thickness deviations, support structure misalignments, etc.) through the control difference analysis layer and corrects them based on the second control assessment results. This precise deviation correction can effectively avoid major construction problems and ensure infrastructure quality. By monitoring and adjusting various procedures in the infrastructure process in real time, the system can help project managers identify and respond to potential risks in a timely manner, reducing the possibility of project delays or cost overruns. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the overall system architecture in one embodiment of the present invention.

[0038] In the diagram: 1. First data acquisition unit; 2. Second data acquisition unit; 3. Third data acquisition unit; 4. Control and assessment unit; 5. Infrastructure control unit. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example 1, please refer to Figure 1 This invention provides a technical solution: an AI-led multi-dimensional management system for the infrastructure construction phase of a pumped storage power station, comprising:

[0041] The first acquisition unit 1 is used to acquire initial geological survey data, wherein the initial geological survey data is the geological feature record data of the target infrastructure area before the commencement of infrastructure construction procedures;

[0042] The second acquisition unit 2 is used to acquire process implementation record data, wherein the process implementation record data is the process record data after the preset infrastructure process is carried out in the target infrastructure area and the preset infrastructure process has completed the modification of the initial geological conditions;

[0043] The third acquisition unit 3 is used to acquire process acceptance verification data, which is the acceptance record data of the transformation effect of the target infrastructure area after the completion of the preset infrastructure process;

[0044] Control and assessment unit 4 is used to compare the control and assessment fit of the target infrastructure area in the first control and assessment result with the control and assessment fit of the target infrastructure area in the second control and assessment result; the first control and assessment result is obtained by performing difference analysis on the process acceptance verification data and process implementation record data; the second control and assessment result is obtained by performing difference analysis on the process implementation record data and initial geological survey data;

[0045] Infrastructure management and control unit 5 is used to determine the infrastructure management and control plan for the target infrastructure area based on the first management and control assessment result, when the management and control adaptability of the target infrastructure area is greater than that of the target infrastructure area in the second management and control assessment result.

[0046] It should be noted that the geological data obtained is from before the infrastructure work begins. This data describes the geological characteristics of the target area, such as soil type, groundwater level, and rock structure. This information is crucial for subsequent construction work. For example, suppose a pumped-storage power station is to be built in a mountainous area. Before construction begins, the geological survey team conducts detailed tests on the soil, rock strata, and groundwater level of the area to obtain initial geological data. This data is used to determine whether the foundation needs to be reinforced or the construction plan needs to be changed.

[0047] It acquires data after the implementation of infrastructure construction procedures during actual construction; it records how these procedures change the initial geological conditions, and these changes are the effects after the pre-set procedures; for example, suppose that in the foundation work of building a power station in a mountainous area, an earthwork excavation and foundation reinforcement procedure is carried out; after these procedures are completed, the system will record relevant data, such as soil compaction and changes in groundwater level; at this time, the construction team has made adjustments based on the previous geological survey information, improving the original geological conditions;

[0048] After the completion of infrastructure construction procedures, data is used for acceptance and verification of the effectiveness of the procedures. It helps to verify whether the completed infrastructure construction procedures have met the expected construction standards. For example, after the construction team completes the foundation reinforcement and earthwork, the relevant departments will inspect these modified infrastructure areas, such as testing the stability of the new soil and changes in groundwater flow. This data will be entered into the system as acceptance records to evaluate the effectiveness of the modifications.

[0049] The preliminary assessment results are compared. The first assessment result is derived from the difference analysis between acceptance data and process implementation data, while the second assessment result is derived from the difference analysis between implementation record data and initial exploration data. Comparing these two results can help determine whether the current infrastructure work is suitable for the current geological conditions. For example: First assessment result: By comparing the differences between process implementation record data and acceptance data, the suitability after construction can be determined; if the process implementation does not meet the expected standards and the difference is large, the suitability is low. Second assessment result: By comparing the differences between initial geological data and process implementation records, the assessment determines whether the geological modification after construction has achieved the expected results; if the initial data and implementation data differ significantly, it indicates that further construction adjustments may be necessary.

[0050] By comparing the evaluation results, it is determined whether a new infrastructure management plan needs to be formulated. Specifically, if the fit in the first evaluation result is greater than the fit in the second evaluation result, it indicates that the construction effect is good, and further infrastructure plans can be formulated based on the first evaluation result. For example, suppose the first evaluation analysis shows that the area after construction is very close to the preset standard, with a high fit; while the second evaluation shows that the geological transformation of the area has not yet met the expected requirements, with a low fit. In this case, the infrastructure management system will prioritize the first evaluation result to formulate subsequent infrastructure management plans, such as strengthening the support in certain areas or carrying out further environmental optimization.

[0051] In an optional embodiment, the method for calculating the control adaptability of the target infrastructure area includes:

[0052] The control and management assessment of the target infrastructure area is carried out based on a pre-trained infrastructure control and management assessment model to obtain the control and management suitability of the target infrastructure area.

[0053] In an optional embodiment, determining an infrastructure management and control scheme for the target infrastructure area based on the results of a first management and control assessment includes:

[0054] Compare the control indicator parameters of the target infrastructure area in the first control assessment results with the control indicator parameters of the target infrastructure area in the second control assessment results;

[0055] If there is a discrepancy between the control indicator parameters of the target infrastructure area in the first control assessment result and the control indicator parameters of the target infrastructure area in the second control assessment result, the first control assessment result shall be corrected based on the second control assessment result.

[0056] The revised first control assessment result was determined as the infrastructure control plan for the target infrastructure area.

[0057] It should be noted that the first control assessment result is based on process acceptance verification data and process implementation record data. Process acceptance verification data refers to the acceptance record data of the transformation effect of the target area after the completion of each infrastructure process. Process implementation record data refers to the record data of the implementation status of each process during the infrastructure construction process. Through these two types of data, difference analysis is performed to assess the gap between the completed infrastructure processes and the preset infrastructure standards. The result of this difference analysis is to determine the control adaptability of the target area, reflecting whether the target area meets the expected requirements after the completion of the infrastructure processes.

[0058] The second control assessment result is based on the process implementation record data and the initial geological survey data. The initial geological survey data is a preliminary record of the geological characteristics of the target area before infrastructure construction. This data includes soil properties, topography, groundwater level, etc. The process implementation record data is the geological condition changes recorded during the actual construction process, reflecting the impact of infrastructure construction on the target area. After performing differential analysis on these data, the second control assessment result can be obtained to assess whether the implementation of infrastructure construction processes has improved the original geological conditions as expected.

[0059] In the control and assessment unit, the control and assessment fit of the target infrastructure area in the first control and assessment result is compared with the control and assessment fit of the target infrastructure area in the second control and assessment result. The first control and assessment result assesses whether the construction standards have been met after the infrastructure construction procedures in the target area. The second control and assessment result assesses whether the geological conditions have been effectively improved before the implementation of the infrastructure construction procedures in the target area. The purpose of the comparison is to find deviations, that is, whether there are significant differences in the control indicators of the target area between the first control and assessment and the second control and assessment. If there are deviations, it means that the first control and assessment result does not fully reflect the actual infrastructure situation of the target area, or the data provided by the second control and assessment result is outdated or incomplete.

[0060] If there is a significant deviation between the control indicator parameters in the first control assessment result and the control indicator parameters in the second control assessment result (for example, the geological transformation effect reflected by the first control assessment result fails to meet the standard), then the first control assessment result needs to be revised. The revision is based on the actual data provided by the second control assessment result. In other words, by revising the original assessment result, it is made more consistent with the actual effect of the transformation of the target area after actual construction. For example, if the second control assessment finds that some transformation areas do not meet the standards, the first control assessment can be adjusted based on these newly discovered data, and certain parameters or processes in the infrastructure construction process can be revised to ensure that the infrastructure project is managed according to actual needs.

[0061] Ultimately, after the initial control assessment results were revised, the infrastructure control unit adopted them as the infrastructure control plan for the target infrastructure area. This plan, derived from the revised initial control assessment results, consists of specific renovation measures and management schemes for the target infrastructure area to ensure the smooth completion of all infrastructure construction processes and the achievement of expected quality standards. For example, the revised plan may include specific measures such as adjusting construction techniques, adding new environmental monitoring points, and optimizing construction schedules to ensure that the project can be completed on time and with high quality.

[0062] A concrete example: Suppose a pumped-storage power station is under construction, requiring some special infrastructure procedures, such as groundwater level adjustment. The first control assessment indicates that after the adjustment, the groundwater level drops to below the expected 50 meters, and the geological conditions meet the design requirements. The second control assessment, however, indicates that the actual water level drop during construction did not reach the expected level, only dropping to 55 meters, and the geological conditions did not improve as expected. A comparison reveals a 5-meter discrepancy between the first and second control assessments, indicating that the actual construction did not proceed as planned. Therefore, the revised assessment results will necessitate adjustments to the groundwater management plan, such as strengthening water level control measures and adding pumping equipment. Ultimately, the revised infrastructure control plan will be optimized based on the second control assessment results to ensure the project meets the final design requirements and avoids deviations in environmental impact and construction schedule.

[0063] In an optional embodiment, revising the first control assessment result based on the second control assessment result includes:

[0064] The first deviation location in the target infrastructure area is identified based on the second control assessment results; wherein, the first deviation location is the location where the process implementation form of the target infrastructure area is abnormal; the abnormal form includes at least one of the following: pouring thickness deviation, support structure misalignment, and excavation slope deviation;

[0065] Based on the standard parameters of the target infrastructure area at the first deviation position in the second control assessment result, the first deviation position of the target infrastructure area in the first control assessment result is corrected to obtain the corrected first control assessment result.

[0066] It should be noted that the first deviation location refers to the specific location where anomalies occur during the construction process of the target infrastructure area. These locations typically exhibit abnormalities, which may include: Pouring thickness deviation: During concrete pouring, the concrete thickness in the target area is not performed according to design requirements, resulting in a pour layer that is too thin or too thick, affecting the structure's load-bearing capacity; Support structure misalignment: Support structures (such as pillars and frames) are not installed according to design positions, resulting in misalignment or tilting, potentially leading to insufficient stability of the infrastructure area; Excavation slope deviation: During excavation, the slope of the earthwork or rock layer does not meet design requirements, resulting in terrain that does not conform to predetermined standards, affecting subsequent construction quality and safety. The occurrence of these abnormalities will cause construction quality to deviate from expectations, potentially affecting the stability and safety of the entire infrastructure project. Based on the second control assessment results, the actual geological changes during construction will first be assessed. By comparing the process implementation record data with the initial geological survey data, deviations will be identified in the actual construction process. For example, it may be found that the excavation slope in certain areas exceeds the design standard, or that the installation of the support structure is misaligned.

[0067] Once the location of the first deviation is identified, the first control assessment result needs to be corrected based on the standard parameters in the second control assessment result. The second control assessment result includes the changes in actual geological conditions during the infrastructure construction process, and the standard parameters may include the optimal values ​​of various construction standards, such as pouring thickness, support structure location, excavation slope, etc.

[0068] Based on the standard parameters in the second control assessment results, the specific correction steps are as follows: If the first deviation occurs in the pouring thickness deviation, the correction can be made by comparing the thickness value recorded during actual construction with the design standard to ensure that the concrete pouring thickness in this area meets the requirements; if the first deviation occurs in the misalignment of the support structure, the position of the support structure is corrected according to the standard data in the second control assessment results to ensure that the support structure is installed accurately according to the design requirements; if the first deviation occurs in the excavation slope deviation, the slope of the excavation area is adjusted according to the standard excavation slope parameters to ensure that the earthwork operation meets the design standards; these corrections can ensure the smooth progress of the infrastructure project and avoid potential safety hazards and quality problems.

[0069] After revision, the first control assessment results will be more accurate and able to reflect the actual construction quality in the target area; this revised assessment result will be used as part of the infrastructure control plan to ensure continuous tracking and improvement throughout the infrastructure process;

[0070] Specific example: Suppose that during the construction of a pumped-storage power station, excavation work is being carried out in a certain area. According to the preliminary construction records, it was found that the excavation slope deviated, that is, the slope was 10 degrees lower than the design requirement. The specific steps are as follows: In the second control assessment results, by comparing the process implementation record data with the initial geological survey data, the system found that the slope deviation of the excavation area was significant, exceeding the design requirement by 10 degrees. According to the standard parameters in the second control assessment results, the design slope of this area is 30 degrees, but the actual slope during construction is 40 degrees. Therefore, after correction, the slope needs to be adjusted to return to its original position. The system reverts to design standards; by adjusting excavation equipment and construction processes, it ensures that subsequent construction meets standard slope requirements; the revised first control assessment results will be updated to reflect the new slope standards, and the system can record this change and track whether subsequent construction complies with the new standards; thus, the construction quality of the target infrastructure area is improved and meets design requirements; based on the revised first control assessment results, the system will propose an adjusted infrastructure control plan, including optimization of construction techniques, improvement of equipment scheduling, and key areas for subsequent acceptance; these measures will ensure that the infrastructure project is successfully completed in accordance with the new requirements.

[0071] In an optional embodiment, revising the first control assessment result based on the second control assessment result further includes:

[0072] The second deviation location in the target infrastructure area is identified based on the second control assessment results; the second deviation location is the location in the target infrastructure area where there is excessive material accumulation or oversight in process implementation;

[0073] Based on the standard parameters of the target infrastructure area at the second deviation position in the second control assessment result, the second deviation position of the target infrastructure area in the first control assessment result is corrected to obtain the corrected first control assessment result.

[0074] It should be noted that the second deviation location refers to some non-standardized situations in the construction of the target infrastructure area. Specifically, this may include: excessive material accumulation: that is, on the construction site, some materials (such as steel bars, concrete, sand and gravel) are not stacked properly according to the plan. Excessive accumulation will affect the construction process and even affect the safety of the surrounding environment and the construction progress; omissions in process implementation: that is, during the construction process, some processes or steps are not executed according to the design standards, which may be due to the negligence of the operators, resulting in the process not being fully followed.

[0075] In the second control assessment results, the second deviation location can be identified by analyzing various construction data and actual conditions. For example, if too much material is found in a certain area during construction, making it impossible for workers to carry out subsequent construction normally, or if a certain process is not fully implemented due to technical or management oversights, it will be identified as the second deviation location.

[0076] Once the location of the second deviation is determined, it needs to be corrected based on the standard parameters in the second control assessment results; the standard parameters include the specific requirements for material accumulation, process steps, etc. in the design and construction specifications.

[0077] Correction for excessive material accumulation: If the second deviation is caused by excessive material accumulation, corrective measures may include: reallocating material storage areas: rationally planning and adjusting the material storage areas on the construction site according to design requirements to avoid excessive accumulation or hindering construction progress; strengthening material management: ensuring that materials are not excessively accumulated through timely material scheduling, supplying materials as needed, and avoiding waste of space and materials; Correction for process implementation omissions: If the second deviation is caused by process omissions, corrective measures may include: strengthening process training and inspection: ensuring that construction personnel understand and strictly follow the design process, especially complex construction processes; establishing quality monitoring points: increasing monitoring of key processes, and promptly identifying and correcting problems in process implementation; The first control assessment result after correction is that the construction quality and progress have become more in line with the predetermined standards after correcting the second deviation; these corrected assessment results will be reflected in subsequent infrastructure control plans, thereby avoiding the impact of material problems or process omissions on project progress and quality;

[0078] Specific examples: Suppose that during tunnel construction, an excessive amount of building materials (such as steel bars and concrete blocks) is piled up in a certain area of ​​the tunnel construction site. These materials are stacked on both sides of the passage, preventing workers from entering and completing subsequent construction steps, such as the installation of support structures or the pouring of concrete. The second control assessment results show that the amount of material piled up in this area exceeds the design requirements, causing delays in construction progress and creating difficulties for subsequent work. According to standard parameters, the amount of material piled up on site should be adjusted according to the construction progress, and the amount piled up in each area should be kept within a reasonable range. The material pile locations will be adjusted, and an effective material management process will be established to avoid excessive material accumulation. After the adjustment, the material pile situation at the construction site has improved, workers can smoothly carry out subsequent process operations, and the construction progress has been restored. The revised first control assessment results show that the material pile problem has been resolved, the construction environment is more orderly, and progress can proceed according to the predetermined plan. During the support construction of a certain section of the tunnel, due to improper operation by construction personnel, the support structure installation process was not carried out according to the correct design steps, resulting in insufficient support stability.

[0079] The second control assessment revealed oversights during the installation of the support structure. Some supports were not installed according to the design standards in terms of position and angle, posing a potential stability risk. To correct this, retraining of construction personnel was necessary to ensure they understood and strictly followed the design requirements. Additionally, monitoring points were added to ensure each process underwent rigorous inspection. The installed support structure also needed to be re-inspected and adjusted as necessary. After readjusting the support structure, construction quality improved, and the stability of the support structure was guaranteed. The revised first control assessment results showed that the support structure was installed according to design requirements, construction quality was restored, and construction progress was accelerated.

[0080] In an optional embodiment, the initial geological survey data, process implementation record data, and process acceptance verification data are all collected using high-precision infrastructure monitoring equipment; the high-precision infrastructure monitoring equipment includes a three-dimensional geological scanner, a stress monitor, and a material density tester.

[0081] In an optional embodiment, the training method for the pre-trained infrastructure management and assessment model includes:

[0082] Create an initial infrastructure management and control assessment model;

[0083] Obtain an infrastructure management and control sample dataset, which includes multiple infrastructure management and control sample pairs with management and control adaptability annotation information. Any target infrastructure management and control sample pair among the multiple infrastructure management and control sample pairs includes initial geological sample data and process implementation sample data. The process implementation sample data is generated by process planning based on the geological feature parameters in the initial geological sample data.

[0084] The initial infrastructure management assessment model was iteratively trained based on the infrastructure management sample dataset;

[0085] If the preset convergence conditions are met, the initial infrastructure management and control assessment model will be determined as the pre-trained infrastructure management and control assessment model.

[0086] It should be noted that the initial infrastructure management and assessment model refers to the model built at the beginning, which usually uses some basic rules, algorithms, or simplified parameter settings. It has not been trained, so its accuracy and effectiveness may be low, and it needs to be optimized and improved through subsequent training. For example, suppose we are building a model to assess the construction quality during the tunnel construction process. The initial model may only consider some basic input parameters, such as construction quality, construction progress, and material stacking, but it has not been trained with a large amount of data, so it cannot comprehensively assess various problems in the infrastructure process.

[0087] The infrastructure management sample dataset refers to a collection of data containing multiple infrastructure management sample pairs with labeled information; each sample pair includes two key parts: initial geological sample data, which reflects the geological conditions before the start of the infrastructure project, such as soil type, rock distribution, groundwater level, etc.

[0088] Process implementation sample data: This describes the execution status of a specific process during the construction of an infrastructure project. This data is typically generated based on initial geological sample data; for example, support structure design or construction sequence planned based on geological characteristics. Control and adaptation information refers to the target results for each sample pair, such as quality scores during construction, and the adaptation of construction progress. This information is used to supervise the training process of the monitoring model. For example, in tunnel engineering: initial geological sample data might include geological reports of the tunnel construction area, describing whether the soil is soft or hard, and whether there is groundwater; process implementation sample data might contain specific construction plans developed based on these geological conditions, such as the need for earthwork excavation, rockburst treatment, or reinforcement of support structures at specific locations; control and adaptation information might be based on feedback from actual construction, such as the quality score (e.g., 0 to 100 points) of a section of the tunnel during construction, or whether it was completed within the predetermined timeframe.

[0089] Iterative training refers to repeatedly using the training dataset to adjust the model's parameters, making the model's predictions more accurate. In each training iteration, the model makes predictions based on the sample dataset, compares them with the actual labeled information, calculates the error, and adjusts the model. Through continuous iterative training, the model gradually "learns" how to extract useful features from the initial geological data and process implementation data, thereby improving the evaluation accuracy. For example, in a tunnel construction project, suppose multiple different construction areas and different geological data samples are used when training the model. Initially, the model may not be able to accurately predict the suitability of each process and may over-rely on some irrelevant geological features, leading to inaccurate evaluation results. Through iterative training, the model will gradually adjust its weights and parameters, pay more attention to key factors affecting construction quality and progress (such as the relationship between groundwater level and soil type), and correct previous errors.

[0090] The convergence condition refers to the condition during training when the model's error drops below a certain threshold, or when the model's performance no longer improves significantly after multiple iterations. At this point, the training process can be considered complete. After training, the initial infrastructure management and assessment model is optimized into a pre-trained model, which can then stably and accurately assess various management and control tasks in infrastructure projects.

[0091] Example: The initial model may be inaccurate in evaluating the support structure under different geological conditions in the first few iterations of training; after multiple iterations, the model's error gradually decreases, and the final prediction error of the model is less than the preset convergence threshold, such as less than 5%; at this point, the training process ends, and the model is a pre-trained infrastructure management and assessment model that can provide accurate management and assessment in the actual construction process.

[0092] In an optional embodiment, the initial infrastructure management assessment model includes a management feature extraction layer, a management difference analysis layer, and a management fit evaluation layer;

[0093] The control feature extraction layer is used to receive the input initial geological sample data and process implementation sample data, and extract the control feature matrix of the initial geological sample data and the control feature matrix of the process implementation sample data, respectively.

[0094] The control difference analysis layer is used to determine the control difference information between the initial geological sample data and the process implementation sample data based on the control feature matrix of the initial geological sample data and the control feature matrix of the process implementation sample data.

[0095] The control fit evaluation layer is used to determine the control fit result of the process implementation sample data based on the control difference information between the initial geological sample data and the process implementation sample data.

[0096] It should be noted that the task of the control feature extraction layer is to extract valuable control feature matrices from the input sample data. The input data is divided into two parts: initial geological sample data: describing the geological conditions before the start of the infrastructure project; and process implementation sample data: describing the execution of processes during the construction of the infrastructure project. The control feature matrix is ​​an abstract representation of each sample data, containing key numerical or attribute information, which forms the basis for subsequent model analysis and evaluation. For example, suppose we want to evaluate a tunnel construction project. The model inputs the following two types of data: initial geological sample data may include: soil type (such as clay, sand, rock, etc.), etc. Groundwater level (e.g., 0-10 meters, 10-20 meters), rock strata type (e.g., soft rock, hard rock), and process implementation sample data may include: excavation method (e.g., mechanical excavation, manual excavation), support method (e.g., shotcrete support, steel support), construction speed (e.g., 3 meters per day). The control feature matrix will extract specific values ​​from these data to form a structured matrix. For example, for initial geological sample data, information such as soil type and groundwater level can be extracted to form a matrix containing these features. Similarly, process implementation sample data will also be transformed into another matrix containing features such as construction method and speed.

[0097] The task of the control difference analysis layer is to analyze and calculate the control difference information between the initial geological sample data and the process implementation sample data. Specifically, it compares the differences between the initial geological conditions and the construction procedures to identify factors that may affect the construction quality or schedule. Difference analysis can assess whether there are potential risks or control mismatches by comparing geological conditions and construction methods. Through this analysis, the model can identify which aspects require special attention during construction and which procedures are incompatible with geological conditions, potentially leading to construction problems. For example, the initial geological sample data shows that the tunnel area is in soft soil with a high groundwater level; the process implementation sample data indicates that mechanical excavation and shotcrete support are used. In this case, the control difference analysis layer will analyze whether the soft soil and high groundwater level will affect the efficiency and safety of mechanical excavation; whether shotcrete support is suitable for this high water level and soft soil environment, and whether additional reinforcement measures are needed. This difference analysis can help the model discover potential construction difficulties, such as soft soil potentially causing instability in the support structure, and a high groundwater level potentially causing excavation difficulties.

[0098] Based on the results of the control difference analysis, the control fit evaluation layer determines the control fit result of the process implementation sample data. That is, based on the information from the difference analysis, it assesses the degree of matching between the process implementation plan and the initial geological conditions, and provides a control fit score or evaluation result. The evaluation result may be a numerical score, representing the fit of the current process implementation plan under given geological conditions. A low score means that the construction plan is not compatible with the geological conditions, which may lead to construction risks. A high score indicates that the construction plan is adapted to the geological conditions and is relatively safe. For example, soft soil and high groundwater levels are not a perfect match for mechanical excavation, which may lead to slow construction progress or even construction accidents (such as landslides, ground subsidence, etc.). Shotcrete support may be ineffective in soft soil and high water levels, posing a risk of insufficient support. In the control fit evaluation layer, the model may provide a fit score, such as 70 points, indicating that the fit of this construction plan under these geological conditions is low and carries certain risks. If the fit score is very low, the model may suggest adopting a more suitable construction plan for these geological conditions, such as switching to manual excavation or strengthening the support structure.

[0099] In an optional embodiment, the control difference analysis layer is further configured to extract the global control feature vector of the initial geological sample data from the control feature matrix of the initial geological sample data, extract the global control feature vector of the process implementation sample data from the control feature matrix of the process implementation sample data, and determine the control difference information between the initial geological sample data and the process implementation sample data based on the global control feature vector of the initial geological sample data and the global control feature vector of the process implementation sample data.

[0100] It should be noted that the control feature matrix is ​​a matrix that contains all the features of the sample data. Each data sample will have a row in the matrix, and the row contains multiple feature values. For example, for a construction project, the sample data may include multiple features such as soil type, groundwater level, and construction method.

[0101] Extracting a "global control feature vector" from the control feature matrix means extracting a vector representing global features from the features of all samples. This vector can summarize the main characteristics of the entire dataset and reflect a global and comprehensive control status. The global control feature vector of the initial geological sample data is extracted from the geological sample data and reflects the global characteristics of geological conditions, such as the distribution of soil types and the range of groundwater level changes. The global control feature vector of the process implementation sample data is extracted from the data actually executed during construction and reflects the global characteristics of the construction process, such as construction methods, progress, and equipment usage.

[0102] Based on the extracted global control feature vector, the difference between the two vectors can be calculated to determine the control difference information between the initial geological sample data and the process implementation sample data. The control difference is usually represented by calculating the degree of difference between the two vectors, and mathematical methods such as Euclidean distance and cosine similarity can be used to quantify this difference. If the difference is large, it may mean that the construction plan is not suitable for the current geological conditions and there is a certain risk.

[0103] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A multi-dimensional management system of AI leading during the construction period of a pumped storage power station, characterized in that, The method comprises the following steps: A first acquisition unit is configured to acquire initial geological survey data, wherein the initial geological survey data is the record data of the geological characteristics of the target infrastructure region before the infrastructure process is carried out; A second acquisition unit is configured to acquire process implementation record data, wherein the process implementation record data is the process record data of the initial geological condition after the preset infrastructure process is completed after the target infrastructure region carries out the preset infrastructure process; A third acquisition unit is configured to acquire process acceptance verification data, wherein the process acceptance verification data is the acceptance record data of the target infrastructure region after the preset infrastructure process is completed; A control evaluation unit is configured to compare the control adaptation degree of the target infrastructure region in the first control evaluation result with the control adaptation degree of the target infrastructure region in the second control evaluation result; the first control evaluation result is obtained by difference analysis on the process acceptance verification data and the process implementation record data; the second control evaluation result is obtained by difference analysis on the process implementation record data and the initial geological survey data; An infrastructure control unit is configured to determine the infrastructure control scheme of the target infrastructure region based on the first control evaluation result when the control adaptation degree of the target infrastructure region in the first control evaluation result is greater than the control adaptation degree of the target infrastructure region in the second control evaluation result.

2. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 1, characterized in that, The method for calculating the control adaptation degree of the target infrastructure region comprises: Controlling and evaluating the target infrastructure region according to the pre-trained infrastructure control evaluation model to obtain the control adaptation degree of the target infrastructure region.

3. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 2, characterized in that, Determining the infrastructure control scheme of the target infrastructure region based on the first control evaluation result comprises: Comparing the control index parameters of the target infrastructure region in the first control evaluation result with the control index parameters of the target infrastructure region in the second control evaluation result; In the case that there is deviation between the control index parameters of the target infrastructure region in the first control evaluation result and the control index parameters of the target infrastructure region in the second control evaluation result, correcting the first control evaluation result based on the second control evaluation result; Determining the corrected first control evaluation result as the infrastructure control scheme of the target infrastructure region.

4. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 3, characterized in that, Correcting the first control evaluation result based on the second control evaluation result comprises: Identifying the first deviation position of the target infrastructure region in the first control evaluation result based on the second control evaluation result; wherein the first deviation position is the position where the process implementation form of the target infrastructure region is abnormal; the abnormal form includes at least one of the following: pouring thickness deviation, supporting structure dislocation, and excavation slope deviation; Correcting the first deviation position of the target infrastructure region in the first control evaluation result based on the standard parameters of the target infrastructure region in the first deviation position in the second control evaluation result to obtain the corrected first control evaluation result.

5. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 4, characterized in that, Correcting the first control evaluation result based on the second control evaluation result further comprises: identify a second deviation position of the target infrastructure region in the first control evaluation result based on the second control evaluation result, wherein the second deviation position is a position in the target infrastructure region where there is an excess of material accumulation or a process implementation oversight; correct the second deviation position of the target infrastructure region in the first control evaluation result based on a standard parameter of the target infrastructure region at the second deviation position in the second control evaluation result, to obtain a corrected first control evaluation result.

6. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 5, characterized in that, The initial geological survey data, the process implementation record data, and the process acceptance verification data are all collected using high-precision infrastructure monitoring equipment; the high-precision infrastructure monitoring equipment includes a three-dimensional geological scanner, a stress monitor, and a material density detector.

7. The AI-led multi-dimensional management system for the construction period of a pumped storage power station according to claim 6, characterized in that, The training method of the pre-trained infrastructure control evaluation model includes: creating an initial infrastructure control evaluation model; obtaining an infrastructure control sample data set, wherein the infrastructure control sample data set includes a plurality of infrastructure control sample pairs with control adaptation degree annotation information, wherein any one target infrastructure control sample pair in the plurality of infrastructure control sample pairs includes initial geological sample data and process implementation sample data, and the process implementation sample data is generated based on a geological feature parameter in the initial geological sample data; iteratively training the initial infrastructure control evaluation model based on the infrastructure control sample data set; if a preset convergence condition is met, the initial infrastructure control evaluation model is determined as the pre-trained infrastructure control evaluation model.

8. The pumped storage power station construction period AI main multi-dimensional management system according to claim 7, characterized by, The initial infrastructure control evaluation model includes a control feature extraction layer, a control difference analysis layer, and a control adaptation degree evaluation layer; The control feature extraction layer is configured to receive the input initial geological sample data and process implementation sample data, and extract a control feature matrix of the initial geological sample data and a control feature matrix of the process implementation sample data, respectively; The control difference analysis layer is configured to determine control difference information between the initial geological sample data and the process implementation sample data based on the control feature matrix of the initial geological sample data and the control feature matrix of the process implementation sample data; The control adaptation degree evaluation layer is configured to determine a control adaptation degree result of the process implementation sample data based on the control difference information between the initial geological sample data and the process implementation sample data.

9. The pumped storage power station construction period AI main multi-dimensional management system according to claim 8, characterized by, The control difference analysis layer is further configured to extract a global control feature vector of the initial geological sample data from the control feature matrix of the initial geological sample data, extract a global control feature vector of the process implementation sample data from the control feature matrix of the process implementation sample data, and determine the control difference information between the initial geological sample data and the process implementation sample data based on the global control feature vector of the initial geological sample data and the global control feature vector of the process implementation sample data.

Images