Drilling and blasting engineering carbon emission accounting and optimization method based on knowledge graph

Abstract

This invention relates to the field of carbon emission management technology in geotechnical engineering, specifically to a knowledge graph-based method for carbon emission accounting and optimization in drilling and blasting engineering. The method includes the following steps: S1: Constructing a knowledge graph of carbon emissions for the entire process; S2: Multi-source standardized data collection and storage; S3: Knowledge graph data matching and accuracy verification; S4: Basic carbon emission accounting for the entire process; S5: Multi-dimensional dynamic correction to obtain the actual carbon emissions; S6: Identification and analysis of high-carbon emission processes; S7: Big data AI carbon reduction path feedback and iterative optimization; S8: Outputting the optimal carbon reduction scheme, data feedback for continuous optimization, applying the optimized scheme to on-site construction, synchronously collecting optimized data, feeding it back to the knowledge graph database, and initiating the next round of optimization. This achieves accurate carbon emission accounting for the entire process, taking into account advanced geological exploration, stress-relieving blasting, complex environmental conditions, and equipment energy efficiency differences, and can effectively perform real-time calibration for different geological conditions.

Description

Technical Field

[0001] This invention relates to the field of carbon emission management technology in geotechnical engineering, and in particular to a method for carbon emission accounting and optimization in drilling and blasting engineering based on knowledge graphs. Background Technology

[0002] Currently, the scale of underground engineering construction in my country's transportation, water conservancy, energy, and mining sectors continues to expand, with a rising proportion of projects involving deep burial, high ground stress, and complex geological conditions. Due to its strong geological adaptability, flexible construction, and controllable costs, the drill-and-blast method remains the most mainstream excavation method for underground geotechnical engineering. According to authoritative industry statistics, carbon emissions during the construction phase of underground engineering account for 30% to 50% of the total carbon emissions throughout the entire life cycle. Drilling and blasting excavation and related auxiliary processes are the core sources of carbon emissions during the construction period. Major projects such as the Ya'an-Xiamen Hydropower underground cavern complex and the deep-buried tunnels of the Sichuan-Tibet Railway can generate tens of thousands of tons of carbon emissions annually for a single project. Conducting precise carbon emission accounting and intelligent carbon reduction research throughout the entire process is not only a necessary requirement in response to national strategies but also a key support for ensuring the green and compliant construction of major projects.

[0003] Underground drilling and blasting engineering has significant technical uniqueness: First, the process chain is lengthy and highly coupled, with carbon emissions occurring throughout the entire process, including drilling, blasting, excavation and loading, transportation, ventilation, support, advanced exploration, stress relief, and miscellaneous operations. Equipment energy consumption, material consumption, explosive decomposition, and auxiliary operations all generate substantial carbon emissions. Second, geological and environmental conditions vary extremely significantly. Complex working conditions such as high-altitude hypoxia, high ground stress, fractured surrounding rock, and extreme cold and heat significantly increase equipment energy consumption and material consumption, further exacerbating carbon emission intensity. This characteristic is particularly prominent in projects such as the Sichuan-Tibet Railway and the Ya'an-Xiamen Hydropower Project. Third, construction data is highly fragmented and isolated. Geological, equipment, process, energy consumption, and emission data are independent of each other, making it difficult to achieve collaborative analysis, unified access, and intelligent decision-making.

[0004] Existing carbon emission accounting and carbon reduction technologies suffer from significant shortcomings and industry pain points: First, the accounting dimensions are severely lacking, generally ignoring key auxiliary processes such as excavation, transportation, ventilation, and support, resulting in accounting results deviating from actual emissions by more than 40%, failing to meet the management requirements of major projects; second, there is a lack of dynamic correction mechanisms, making it impossible to perform real-time calibration based on geological conditions, environmental characteristics, and equipment energy efficiency; third, carbon reduction paths rely on manual experience for formulation, lacking the support of big data and artificial intelligence, making it difficult to achieve a balanced optimization of multiple objectives such as safety, quality, schedule, and carbon reduction; finally, the lack of knowledge graphs to achieve multi-source data fusion results in severely insufficient intelligent decision-making capabilities. To address these industry pain points, fill technological gaps, and meet the green, low-carbon, and intelligent construction needs of major projects and various underground engineering projects, this invention proposes a carbon emission accounting and carbon reduction path optimization method that covers all processes, performs high-precision accounting, deeply integrates data, and utilizes AI intelligent optimization, possessing significant theoretical and engineering value for promoting the low-carbon transformation of underground engineering. Summary of the Invention

[0005] The purpose of this invention is to provide a knowledge graph-based method for carbon emission accounting and optimization in drilling and blasting engineering. This method can systematically and standardly calculate carbon emissions for all processes and select carbon reduction pathways in underground engineering under complex geological conditions.

[0006] To achieve the above objectives, this invention provides a knowledge graph-based method for carbon emission accounting and optimization in drilling and blasting engineering, comprising the following steps: S1: Construct a knowledge graph of carbon emissions for the entire process, establish relationships between entities in four categories: geological environment, construction process, equipment energy consumption, and materials and emission factors, input historical engineering sample data, and improve the knowledge graph database; S2: Multi-source standardized data collection and storage, synchronously collecting on-site geological, construction, equipment, and energy consumption data, and after standardization processing, importing them into the knowledge graph database; S3: Knowledge graph data matching and accuracy verification. Execute the matching accuracy formula to calculate the data quality, and return unqualified data for re-collection and correction until the accuracy requirements are met. S4: Basic carbon emission accounting for the entire process, calculate the independent carbon emissions of each process, and sum them to obtain the total basic carbon emissions for the entire process; S5: Multi-dimensional dynamic correction to obtain the true carbon emissions, dynamically correct the carbon emission impact coefficient of advanced geological exploration and the carbon emission impact coefficient of stress relief blasting, and superimpose the basic accounting results to obtain the true carbon emissions. S6: Identification and analysis of high carbon emission processes; identify the processes with the highest carbon emission rate as high energy-consuming processes and optimize them. S7: Big Data AI Carbon Reduction Path Feedback Iterative Optimization, based on gradient descent algorithm and historical engineering database training, achieves adaptive iterative optimization of carbon reduction path through real-time error correction; S8: Output the optimal carbon reduction solution, continuously optimize the data feedback, apply the optimized solution to on-site construction, collect the optimized data synchronously, feed it back to the knowledge graph database, and start the next round of optimization.

[0007] The method of this invention constructs a knowledge graph covering the entire process of drilling, blasting, excavation and loading, which enables accurate accounting of carbon emissions for the entire process and also achieves standardized fusion of multi-source data.

[0008] Step S1, which involves constructing a knowledge graph of carbon emissions across the entire process, includes the following steps: Construct the logical structure of the knowledge graph, setting up entities for geological environment, construction procedures, equipment energy consumption, and materials and emission factors; Extract relevant data, and extract hidden geological features and energy consumption-related knowledge data from both structured and unstructured data; Knowledge fusion and correction: To address naming conflicts in the data, entity alignment and correction are performed by calculating the text similarity of entity attributes, ensuring the uniqueness and accuracy of the energy consumption data and emission factor standards entered into the database. Knowledge graph storage and dynamic updates store the extracted data in a database, breaking down data silos between various processes and energy consumption, and providing computational data for subsequent multi-dimensional dynamic correction and AI iterative optimization.

[0009] The geological environment entity includes the surrounding rock grade attribute and the exploration depth attribute; the construction procedure entity includes the drilling construction node, the blasting construction node and the excavation and loading construction node; the equipment energy consumption entity includes the rock drilling rig attribute, the excavator attribute, the fan and its power attribute and operating time attribute; and the material and emission factor entity includes the carbon emission factor of explosives, the carbon emission factor of diesel, and the carbon emission factor of electricity.

[0010] For structured data such as equipment ledgers and sensor energy consumption, and unstructured data such as geological survey reports and construction logs, the hidden geological features and energy consumption-related knowledge data are extracted.

[0011] Knowledge fusion and correction: To address naming conflicts in the data, entity alignment and correction are performed by calculating the text similarity of entity attributes, ensuring the uniqueness and accuracy of the energy consumption data and emission factor standards entered into the database.

[0012] Knowledge graph storage and dynamic updates. Extracted data is stored in a database, breaking down data silos between different processes and energy consumption, and providing computational data for subsequent multi-dimensional dynamic correction and AI iterative optimization.

[0013] The collected data is divided into five categories: geological environment data, construction procedure data, equipment energy consumption data, material consumption data, and emission factor data. The data collection methods include on-site monitoring, equipment ledgers, construction logs, smart sensor data collection, and third-party testing data.

[0014] Furthermore, the verification formula for knowledge graph data matching and accuracy verification in step S3 is as follows: In the formula, S match This refers to the accuracy of entity-relation matching in a knowledge graph, which is unitless and ranges from 0 to 1. ω1 is the process entity weighting coefficient; ω2 is the emission factor weighting coefficient; N correct It matches the correct number of entities; N totalIt represents the total number of entities in the knowledge graph; M factor It is about matching the correct number of emission factors; M total It is the number of total emission factors used in the calculation.

[0015] The process is as follows: data entry - entity matching - relationship verification - precision calculation - successful data entry / abnormal data return and correction. S match ω1 is the knowledge graph entity and relation matching accuracy, unitless, ranging from 0 to 1. The verification standard is: ≥0.9 for routine projects and ≥0.95 for major projects. Data matching below this standard requires re-matching. ω2 is the process entity weight coefficient, unitless, ranging from 0.5 to 0.7. The value is determined by increasing the complexity of the process; 0.6 for routine projects and 0.7 for major projects. ω3 is the emission factor weight coefficient, unitless, ranging from 0.3 to 0.5. The value is complementary to the process entity weight coefficient. For conventional projects, the value is 0.4; for major projects, the value is 0.3. N correct This is the number of correctly matched entities, in units of: individuals. The method of obtaining this value is a combination of manual verification and intelligent matching with statistics. N total This represents the total number of entities in the knowledge graph, measured in units of 1. The method of measurement is based on categories such as geology, construction, and equipment. For routine projects, the number of entities is 40-60, and for major projects, it is 60-80. M factor This is the number of correctly matched emission factors, in units of: numbers, determined by statistical verification against standard values / detected values. M total The total number of emission factors used in the calculation is expressed in units of "number". The method of value is based on energy, materials and process categories, with 20-30 for conventional projects and 30-40 for major projects.

[0016] Furthermore, in step S4, the derivation formula for the total basic carbon emissions of the entire process is as follows: In the formula, E total This represents the total basic carbon emissions from the entire drilling and blasting process, expressed in kgCO2e. E drill This refers to carbon emissions from the drilling process, measured in kgCO2e. E expel This refers to carbon emissions from the blasting process, measured in kgCO2e. E load This refers to carbon emissions from the excavation and loading process, measured in kgCO2e. E transCarbon emissions from transportation processes, unit: kgCO2e. E vent Carbon emissions from the ventilation process, unit: kgCO2e. E support This refers to carbon emissions from the support process, measured in kgCO2e. E explore This refers to carbon emissions from advanced geological exploration processes, measured in kgCO2e. E explore This refers to carbon emissions from advanced geological exploration processes, measured in kgCO2e. E stress This refers to carbon emissions from the stress relief blasting process, measured in kgCO2e. E other This refers to carbon emissions from minor processes, measured in kgCO2e.

[0017] E total It is the total basic carbon emission of the entire process of underground rock and soil drilling and blasting engineering, in kgCO2e, and the method of taking the value is to sum up the carbon emissions of each process. E drill It refers to carbon emissions from the drilling process, generated by the energy consumption of drilling equipment such as rock drilling rigs and handheld drills. The unit is kgCO2e, and the value is calculated by multiplying the power consumption of the drilling equipment by the power emission factor. For conventional projects, the value is 100-200 kgCO2e per cycle, and for major projects, it is 200-300 kgCO2e. E expel It is the carbon emission of the blasting process, which is generated by the chemical decomposition of explosives, the energy consumption of the detonation equipment and drilling auxiliary energy. The unit is kgCO2e. The value is calculated as explosive consumption × explosive emission factor + carbon emission of detonation equipment energy consumption. The value is 300-500 kgCO2e for conventional projects and 500-800 kgCO2e for major projects. E load Etrans represents carbon emissions from the excavation and loading process, generated by the energy consumption of excavators, loaders, and other equipment used for transporting construction waste. The unit is kgCO2e, and the value is calculated as the electricity consumption (or fuel consumption × diesel emission factor) of the excavating and loading equipment. For conventional projects, the value per cycle is 200-300 kgCO2e, and for major projects, it is 300-400 kgCO2e. Etrans represents carbon emissions from the transportation process, generated by the energy consumption of dump trucks and other vehicles used for transporting construction waste. The unit is kgCO2e, and the value is calculated as the fuel consumption of the transport vehicle × diesel emission factor × number of transport trips. For conventional projects, the value per cycle is 2000-3000 kgCO2e, and for major projects, it is 3000-5000 kgCO2e. E ventCarbon emissions from ventilation processes are generated by the long-term energy consumption of tunnel axial flow fans and jet fans. The unit is kgCO2e, and the value is calculated as fan power × running time × power emission factor. For conventional projects, the value is 500-600 kgCO2e for a single cycle, and for major projects, it is 600-800 kgCO2e. E support It refers to the carbon emissions of the support process, which are generated by the production and construction energy consumption of anchor bolts, shotcrete, and steel arch frame materials. The unit is kgCO2e. The value is calculated as the consumption of various support materials × the corresponding emission factor + the carbon emissions from the energy consumption of construction equipment. The value for a single cycle of conventional projects is 4000-5000 kgCO2e, and for major projects it is 5000-7000 kgCO2e. E explore It is the carbon emission of advanced geological exploration process, generated by the energy consumption of ground-penetrating radar and advanced drilling equipment. The unit is kgCO2e. The value is calculated by the energy consumption of the exploration equipment × the corresponding emission factor. The value is 100-150 kgCO2e for a single cycle in conventional projects and 150-200 kgCO2e for major projects. E stress It is the carbon emission of the stress relief blasting process, generated by the energy consumption of special blasting construction and auxiliary processes, with the unit: kgCO2e. The method of value is the same as that of blasting process. The value of a single cycle in conventional projects is 200-250 kgCO2e, and that in major projects is 250-300 kgCO2e. E other These are carbon emissions from minor processes, generated by construction lighting, drainage, auxiliary facilities, temporary operations, etc. The unit is kgCO2e. The method of taking the value is energy consumption of various minor equipment × corresponding emission factor. The value for a single cycle of conventional projects is 80-120 kgCO2e, and for major projects it is 120-150 kgCO2e.

[0018] Furthermore, in step S5, the formula for the carbon emission impact coefficient of the advanced geological exploration is: In the formula, f explore It is the carbon emission impact coefficient of advanced geological exploration; k 1 It is a correction factor for exploration methods; l geo It is the actual geological complexity coefficient; l 0 It is the standard geological complexity coefficient; L explore This is the actual exploration depth; L 0 It is the standard exploration depth; n explore It refers to the frequency of exploration.

[0019] In step S5, the formula for calculating the actual carbon emissions is as follows: In the formula, E real This is the corrected actual carbon emissions, in kgCO2e. f env It is the environmental correction factor; f alt It is the equipment energy efficiency correction factor. f sup It is the support strength correction factor. f explore is the carbon emission impact coefficient of advanced geological exploration, which has no unit and ranges from 0.8 to 1.5. The more complex the geology and the more frequent the exploration, the larger the value. k1 is the exploration method correction coefficient, which has no unit and is determined as follows: 1.0 for drilling technology, 0.6 to 0.8 for geophysical technology, and 1.1 to 1.2 for hybrid exploration technology. l geo λ0 is the actual geological complexity coefficient, which has no unit. The value is determined according to the surrounding rock grade and the degree of rock mass fragmentation: 0.8 for Grade I surrounding rock, 1.0 for Grade II, 1.2 for Grade III, 1.5 for Grade IV, and 2.0 for Grade V. λ0 is the standard geological complexity coefficient, which has no unit. The value is determined by the national standard recommended benchmark value of 1.0 (corresponding to Grade II surrounding rock). L explore L1 is the actual exploration depth, in meters (m). The method of measurement is actual measurement; for conventional projects, it is 20-30m, and for major projects, it is 30-50m. L2 is the standard exploration depth, in meters (m). The method of measurement is to take the average value of 20m for conventional projects. n explore This refers to the exploration frequency, measured in times per construction cycle. The value is calculated as follows: for conventional projects, once every 3-5 cycles (value 0.2-0.33); for major projects, once every 2-3 cycles (value 0.33-0.5).

[0020] E real This is the corrected actual carbon emissions, in kgCO2e, and is calculated by multiplying the base total carbon emissions by the correction factors. fenv This is an environmental correction factor, without units. The value is determined according to environmental conditions: 1.0 for normal temperature and pressure engineering, 1.1-1.2 for plateau engineering (altitude 2000-3000m), 1.2-1.3 for high-altitude engineering (above 3000m), 1.15-1.25 for frigid engineering (temperature ≤ -10℃), and 1.1-1.2 for high-temperature engineering (temperature ≥ 35℃). f altThis is the equipment energy efficiency correction factor, which has no unit. The method of determining the value is based on the age and energy consumption level of the equipment: 0.9-0.95 for new equipment (service life ≤ 3 years), 0.95-1.05 for medium-aged equipment (service life 3-8 years), 1.05-1.15 for old equipment (service life ≥ 8 years), and 0.05-0.1 for energy-saving equipment. f sup It is the support strength correction factor, which has no unit. The method of determining the value is based on the ratio of the support strength to the standard value. For conventional support, it is 1.0; for reinforced support, it is 1.05-1.15; and for simple support, it is 0.9-0.95.

[0021] Furthermore, in step S5, the formula for the carbon emission impact coefficient of the ground stress relief blasting is: In the formula, f stress It is the carbon emission impact coefficient of stress relief blasting; k 2 It is the blasting process correction coefficient; s max This is the measured maximum ground stress value; s 0 It is the standard ground stress reference value; q stress This is the actual explosive consumption per unit during ground stress relief blasting; q 0 This is the standard explosive consumption per unit. K v It is the surrounding rock integrity coefficient.

[0022] f stress It is the carbon emission impact coefficient of blasting for stress relief, without units, and its value ranges from 1.0 to 2.5. The greater the ground stress and the more fractured the surrounding rock, the larger the value. k 2 This is a blasting process correction factor, without units. The value is determined as follows: 1.0 for continuous charge blasting, 1.1-1.2 for interval charge blasting, and 1.2-1.3 for pre-splitting blasting. s max This is the measured maximum ground stress value, in MPa, obtained by actual measurement using a borehole stress gauge. s 0 It is the standard ground stress reference value, unit: MPa, and the method of taking the value is: take the average value of conventional engineering, 20MPa; q stressq0 is the actual explosive consumption per unit for stress relief blasting, in kg / m³, determined by actual measurement; 0.6-0.8 kg / m³ for conventional projects and 0.8-1.0 kg / m³ for major projects. q0 is the standard explosive consumption per unit, in kg / m³, determined by the average value of 0.7 kg / m³ for conventional projects. K v It is the rock mass integrity coefficient, which has no unit. The value is determined by the rock mass quality index Q or BQ classification. For rock mass of grades I-II, the value is 0.8-1.0; for grade III, the value is 0.6-0.8; and for grades IV-V, the value is 0.4-0.6.

[0023] Furthermore, in step S7, the iterative optimization formula is as follows: In the formula, E opt(i) It represents the carbon emissions after the i-th iteration of optimization; It is the AI ​​learning rate; W data ΔP is the big data weighting coefficient; ΔP is the optimization amount of drilling and blasting parameters; ΔT is the optimization amount of construction technology. It is the amount of low-carbon material substitution optimization; It is the optimization of excavation and loading transportation routes and efficiency; It refers to the duration of the ventilation system and the optimization of air volume; It is the amount of support parameter optimization; It is the real-time error correction coefficient; e i ε is the prediction error value of the i-th prediction; ε is the basic convergence threshold.

[0024] E opt(i) α is the carbon emissions after the i-th iteration optimization, in kgCO2e, and is calculated by combining the previous optimization value with the iteration formula; α is the AI ​​learning rate, which is dimensionless and ranges from 0.05 to 0.3. The method of setting the learning rate is: 0.1-0.2 for routine projects and 0.15-0.25 for major projects. The higher the learning rate, the faster the iteration speed, but it should be controlled within 0.3 to avoid iteration divergence. W data This is a big data weighting coefficient, without units. The value is determined by the sample size of historical projects: 0.7-0.8 for ≤100 samples, 0.8-0.9 for 100-200 samples, and 0.9-0.95 for ≥200 samples. Databases containing samples from major projects can be increased by an additional 0.05. ΔP It is the optimization amount of drilling and blasting parameters, without unit, with a value range of 0.05-0.2. The method of value selection is: determined according to the optimization range of explosive consumption and hole spacing. For conventional projects, it is 0.08-0.15, and for major projects, it is 0.1-0.2. ΔTThis is the amount of optimization in construction technology, without units, with a value range of 0.05-0.15. The method of determining the value is based on the construction cycle efficiency and the optimization range of process connection. For conventional projects, it is taken as 0.08-0.12, and for major projects, it is taken as 0.1-0.15. DM It is the optimized amount of low-carbon material substitution, without units, with a value range of 0.03-0.1. The method of determining the value is based on the low-carbon material substitution ratio, with 0.05-0.08 for conventional projects and 0.07-0.1 for major projects. ΔL It is the optimization quantity of excavation and loading transportation routes and efficiency, without units, with a value range of 0.1-0.2. The method of value determination is based on the reduction of transportation distance and empty load rate. For conventional projects, it is taken as 0.12-0.18, and for major projects, it is taken as 0.15-0.2. ΔV This refers to the optimized duration and air volume of the ventilation system. It has no unit and ranges from 0.1 to 0.2. The method of determining the value is based on the reduction in ineffective ventilation time. For routine projects, the value is 0.13-0.17, and for major projects, it is 0.15-0.2. ΔS δ is the optimized support parameter value, which is dimensionless and ranges from 0.05 to 0.15. The value is determined based on the material savings of the support. For conventional projects, it is 0.08 to 0.12, and for major projects, it is 0.1 to 0.15. δ is the real-time error correction coefficient, which is dimensionless and ranges from 0.5 to 0.8. The value is determined by the larger the error. For conventional projects, it is 0.6 to 0.7, and for major projects, it is 0.7 to 0.8. e i This is the prediction error value for the i-th time, without units, and the method for determining its value is as follows: For routine projects, the threshold value should be controlled within 0.05, and for major projects, it should be controlled within 0.03. ε is the basic convergence threshold, with the unit being kgCO2e. The value is determined by setting it to 10 kgCO2e, but for major projects, it can be adjusted to 5 kgCO2e to improve convergence accuracy.

[0025] This invention presents a knowledge graph-based method for carbon emission accounting and optimization in drilling and blasting engineering. It achieves accurate carbon emission accounting across the entire process, constructing a knowledge graph covering drilling, blasting, excavation, and loading, and realizing standardized fusion of multi-source data. It comprehensively considers advanced geological exploration, stress-relieving blasting, complex environmental conditions, and equipment energy efficiency differences, enabling effective real-time calibration for different geological conditions and achieving multi-variable optimization calculations on the basis of basic accounting. This method changes the previous reliance on manual experience for carbon reduction, providing an efficient and low-carbon transformation solution for underground geotechnical engineering in complex environments such as high altitude, deep burial, and high ground stress. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the carbon emission knowledge graph architecture for the entire process of underground rock and soil drilling and blasting engineering according to the present invention.

[0028] Figure 2 This invention provides a flowchart for intelligent carbon emission accounting and carbon reduction path feedback optimization in underground rock and soil drilling and blasting engineering. Detailed Implementation

[0029] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0030] Please see Figure 1 to Figure 2 This invention provides a knowledge graph-based method for carbon emission accounting and optimization in drilling and blasting engineering, comprising the following steps: S1: Construct a knowledge graph of carbon emissions for the entire process, establish relationships between entities in four categories: geological environment, construction process, equipment energy consumption, and materials and emission factors, input historical engineering sample data, and improve the knowledge graph database; S2: Multi-source standardized data collection and storage, synchronously collecting on-site geological, construction, equipment, and energy consumption data, and after standardization processing, importing them into the knowledge graph database; S3: Knowledge graph data matching and accuracy verification. Execute the matching accuracy formula to calculate the data quality, and return unqualified data for re-collection and correction until the accuracy requirements are met. S4: Basic carbon emission accounting for the entire process, calculate the independent carbon emissions of each process, and sum them to obtain the total basic carbon emissions for the entire process; S5: Multi-dimensional dynamic correction to obtain the true carbon emissions, dynamically correct the carbon emission impact coefficient of advanced geological exploration and the carbon emission impact coefficient of stress relief blasting, and superimpose the basic accounting results to obtain the true carbon emissions. S6: Identification and analysis of high carbon emission processes; identify the processes with the highest carbon emission rate as high energy-consuming processes and optimize them. S7: Big Data AI Carbon Reduction Path Feedback Iterative Optimization, based on gradient descent algorithm and historical engineering database training, achieves adaptive iterative optimization of carbon reduction path through real-time error correction; S8: Output the optimal carbon reduction solution, continuously optimize the data feedback, apply the optimized solution to on-site construction, collect the optimized data synchronously, feed it back to the knowledge graph database, and start the next round of optimization.

[0031] The specific implementation method is as follows: The underground main powerhouse cavern of a high-altitude hydropower project (referencing the construction characteristics of the underground cavern of the Yaxia Hydropower Project) was used as the verification object. The cavern has a cross-sectional dimension of 18m × 22m, an excavated cross-sectional area of ​​396m², an axial length of 210m, a burial depth of 900m, an altitude of 3260m, a measured in-situ stress of 28.6MPa, a surrounding rock grade of Class II, and a surrounding rock integrity coefficient Kv = 0.82. The cavern was excavated in layers using the drill-and-blast method, with an advance of 3.0m per cycle. The construction procedures fully cover drilling, blasting, excavation and loading, transportation, ventilation, support, advanced exploration, stress relief blasting, and miscellaneous operations, fully matching the full-process accounting system of this invention and complying with the carbon emission control requirements for major projects.

[0032] Example of basic parameter values Based on the parameter value range / method of this invention and in conjunction with engineering practice (major hydropower projects on plateaus), the following parameter values ​​are determined: 1. Drilling process: The power of the rock drilling rig is 132kW (range 100-150kW), the running time is 1.5h, the power consumption is 198kWh, and the power emission factor is 0.585kgCO2e / kWh; 2. Blasting procedure: 528 kg of explosives used, explosive emission factor of 0.75 kg CO2e / kg (recommended value), and explosive consumption of 0.85 kg / m³. 3. Excavation and loading process: Excavator power 220kW (range 180-250kW), running time 2.0h, power consumption 440kWh; 4. Transportation process: The fuel consumption of the dump truck is 28L / km, the transportation distance is 1.2km (the value range for major projects is 2-5km, this project is on-site transportation, so we take 1.2km), a total of 36 transportation trips are made, and the diesel emission factor is 2.63kgCO2e / L; 5. Ventilation process: Fan power 160kW, single-cycle ventilation time 6h (6-8h range for major projects); 6. Support procedure: Shotcrete 32m 3 120 anchor bolts (3m in length, ranging from 2-5m), 3 steel arch frames, and a shotcrete emission factor of 120kgCO2e / m³. 3 The emission factor of anchor bolts is 6.8 kg CO2e / bolt, and the emission factor of steel arch frames is 125 kg CO2e / frame; 7. Advanced exploration: Exploration depth 30m (30-50m for major projects), exploration once every 3 cycles (value 0.33), exploration method is mixed exploration, k1=1.15; 8. Ground stress relief blasting: Explosive consumption 0.85 kg / m³ 3 The blasting process uses interval charging, with k2=1.15; 9. Correction factors: Environmental correction factor fenv = 1.22 (for high-altitude projects, the value range is 1.2-1.3), equipment energy efficiency correction factor. f alt =0.94 (for new equipment, the value range is 0.9-0.95), support strength correction factor f sup =1.06 (Reinforced support, value range 1.05-1.15); 10. AI optimization parameter: α=0.15 (value range of 0.15-0.25 for major projects). W data =0.92 (sample size ≥200), δ=0.75 (value range 0.7-0.8 for major projects), ε=10kgCO2e.

[0033] (II) Calculation of basic carbon emissions for the entire process The full-process accounting formula is adopted: The calculations are as follows: 1. Carbon emissions from drilling: E drill =198 × 0.585 = 115.83 kg CO2e 2. Carbon emissions from explosions: E expl =528 × 0.75 = 396.00 kg CO2e 3. Carbon emissions from excavation and loading: E load =440 × 0.585 = 257.40 kg CO2e 4. Carbon emissions from transportation: E trans =28×1.2×36×2.63=3171.07kgCO2e 5. Carbon emissions from ventilation: E vent =160×6×0.585=561.60kgCO2e 6. Support carbon emissions: E support =32×120+120×6.8+3×125=4741.00kgCO2e 7. Carbon emissions from exploration:E explore =128.60kgCO2e (calculated based on the energy consumption of exploration equipment, which is within the range for major projects) 8. Carbon emissions from ground-stress blasting: E stress =212.50kgCO2e (Complies with the value range for major engineering projects) 9. Carbon emissions from minor processes: E other =98.20kgCO2e (meets the range for major engineering projects) Total base carbon emissions: E total =115.83+396+257.4+3171.07+561.6+4741+128.6+212.5+98.2=9682.2kgCO2e Key Influence Coefficient Calculation Exploration Influence Coefficient (l geo =1.0, l 0 =1.0, L explore =30m, L 0 =20m, n explore =0.33): Influence coefficient of blasting for stress relief ( s max =28.6MPa, s 0 =20MPa q stress =0.85kg / m³, q0=0.7kg / m³, K v =0.82): Overall correction factor: Real carbon emissions calculation Knowledge graph matching accuracy calculation (ω1=0.7, ω2=0.3, N) correct =47, N total =50, M factor =28, M total =30): With an accuracy of 0.937, it is close to the standard of 0.95 for major projects, and can reach above 0.95 after slight data correction, thus meeting the requirements.

[0034] AI-based carbon reduction optimization calculations Substituting into the AI ​​optimization formula (ΔP=0.08, ΔT=0.12, ΔM=0.06, ΔL=0.18, ΔV=0.16, ΔS=0.11, e i =0.025): Implementation effect Field application showed that the error between the calculated results and the measured emissions of this invention was only 2.87%, far superior to the error of over 42% of traditional methods, and meeting the accuracy requirement of less than 5% for major projects. After optimization, ventilation energy consumption was reduced by 19.3%, transportation energy consumption by 16.8%, support material consumption by 11.5%, and drilling and blasting energy consumption by 8.6%. The project achieved safe, efficient, and low-carbon construction, verifying the scientific validity, reliability, and practicality of this invention.

[0035] The above description discloses only one preferred embodiment of the present invention, and should not be construed as limiting the scope of the present invention. Those skilled in the art will understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for calculating and optimizing carbon emissions of drilling and blasting engineering based on a knowledge graph, characterized in that, Includes the following steps: S1: Construct a knowledge graph of carbon emissions for the entire process, establish relationships between entities in four categories: geological environment, construction process, equipment energy consumption, and materials and emission factors, input historical engineering sample data, and improve the knowledge graph database; S2: Multi-source standardized data collection and storage, synchronously collecting on-site geological, construction, equipment, and energy consumption data, and after standardization processing, importing them into the knowledge graph database; S3: Knowledge graph data matching and accuracy verification. Execute the matching accuracy formula to calculate the data quality, and return unqualified data for re-collection and correction until the accuracy requirements are met. S4: Basic carbon emission accounting for the entire process, calculate the independent carbon emissions of each process, and sum them to obtain the total basic carbon emissions for the entire process; S5: Multi-dimensional dynamic correction to obtain the true carbon emissions, dynamically correct the carbon emission impact coefficient of advanced geological exploration and the carbon emission impact coefficient of stress relief blasting, and superimpose the basic accounting results to obtain the true carbon emissions. S6: Identification and analysis of high carbon emission processes; identify the processes with the highest carbon emission rate as high energy-consuming processes and optimize them. S7: Big Data AI Carbon Reduction Path Feedback Iterative Optimization, based on gradient descent algorithm and historical engineering database training, achieves adaptive iterative optimization of carbon reduction path through real-time error correction; S8: Output the optimal carbon reduction solution, continuously optimize the data feedback, apply the optimized solution to on-site construction, collect the optimized data synchronously, feed it back to the knowledge graph database, and start the next round of optimization.

2. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1, characterized in that, Step S1, which involves constructing a knowledge graph of carbon emissions across the entire process, includes the following steps: Construct the logical structure of the knowledge graph, setting up entities for geological environment, construction procedures, equipment energy consumption, and materials and emission factors; Extract relevant data, and extract hidden geological features and energy consumption-related knowledge data from both structured and unstructured data; Knowledge fusion and correction: To address naming conflicts in the data, entity alignment and correction are performed by calculating the text similarity of entity attributes, ensuring the uniqueness and accuracy of the energy consumption data and emission factor standards entered into the database. Knowledge graph storage and dynamic updates store the extracted data in a database, breaking down data silos between various processes and energy consumption, and providing computational data for subsequent multi-dimensional dynamic correction and AI iterative optimization.

3. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1 or 2, characterized in that, The verification formula for knowledge graph data matching and accuracy verification in step S3 is as follows: In the formula, S match is the knowledge graph entity and relationship matching accuracy, unitless, value range 0-1, ω1 is the process entity weight coefficient; 2 is the emission factor weight coefficient; N correct is the number of matched correct entities; N total is the total number of knowledge graph entities; M factor is the number of matched correct emission factors; M total is the total number of emission factors used in accounting.

4. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1, characterized in that, In step S4, the derivation formula for the total basic carbon emissions of the entire process is as follows: In the formula, E total This represents the total basic carbon emissions from the entire drilling and blasting process, expressed in kgCO2e. E drill This refers to carbon emissions from the drilling process, measured in kgCO2e. E expel This refers to carbon emissions from the blasting process, measured in kgCO2e. E load This refers to carbon emissions from the excavation and loading process, measured in kgCO2e. E trans Carbon emissions from transportation processes, unit: kgCO2e. E vent Carbon emissions from the ventilation process, unit: kgCO2e. E support This refers to carbon emissions from the support process, measured in kgCO2e. This refers to carbon emissions from advanced geological exploration processes, measured in kgCO2e. This refers to carbon emissions from advanced geological exploration processes, measured in kgCO2e. E stress This refers to carbon emissions from the stress relief blasting process, measured in kgCO2e. E other This refers to carbon emissions from minor processes, measured in kgCO2e.

5. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1 or 4, characterized in that, In step S5, the formula for the carbon emission impact coefficient of advanced geological exploration is: In the formula, f explore It is the carbon emission impact coefficient of advanced geological exploration; k 1 It is a correction factor for exploration methods; λ geo It is the actual geological complexity coefficient; λ It is the standard geological complexity coefficient; L explore This is the actual exploration depth; It is the standard exploration depth; explore It refers to the frequency of exploration.

6. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 5, characterized in that, In step S5, the formula for calculating the actual carbon emissions is as follows: In the formula, E real This is the corrected actual carbon emissions, in kgCO2e. f env It is the environmental correction factor; f alt It is the equipment energy efficiency correction factor. f sup It is the support strength correction factor.

7. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1, characterized in that, In step S5, the formula for the carbon emission impact coefficient of the stress relief blasting is: In the formula, f stress It is the carbon emission impact coefficient of stress relief blasting; k 2 It is the blasting process correction coefficient; σ max This is the measured maximum ground stress value; 0 It is the standard ground stress reference value; q stress This is the actual explosive consumption per unit during ground stress relief blasting; q 0 This is the standard explosive consumption per unit. K v It is the surrounding rock integrity coefficient.

8. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 1 or 2, characterized in that, In step S7, the iterative optimization formula is as follows: In the formula, E opt(i) α is the carbon emissions after the i-th iteration of optimization; α is the AI ​​learning rate. W data ΔP is the weighting coefficient for big data; ΔT is the optimization amount for drilling and blasting parameters; ΔM is the optimization amount for construction technology; ΔL is the optimization amount for low-carbon material substitution; ΔV is the optimization amount for excavation, loading, and transportation routes and efficiency; ΔV is the optimization amount for ventilation system duration and air volume; ΔS is the optimization amount for support parameters. δ is the real-time error correction coefficient; e i ε is the prediction error value of the i-th prediction; ε is the basic convergence threshold.

9. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 8, characterized in that, The convergence condition for the iteration is: 。 10. The knowledge graph-based carbon emission accounting and optimization method for drilling and blasting engineering as described in claim 2, characterized in that, The geological environment entity includes the surrounding rock grade attribute and the exploration depth attribute; the construction procedure entity includes the drilling construction node, the blasting construction node and the excavation and loading construction node; the equipment energy consumption entity includes the rock drilling rig attribute, the excavator attribute, the fan and its power attribute and operating time attribute; and the material and emission factor entity includes the carbon emission factor of explosives, the carbon emission factor of diesel, and the carbon emission factor of electricity.

Images