Mine enterprise digital transformation path generation and risk assessment method, device, equipment and medium

Abstract

The application discloses a mine enterprise digital transformation path generation and risk assessment method and device, equipment and medium, relates to the technical field of data processing, and comprises the following steps: forming an enterprise state vector through multi-source heterogeneous data fusion, constructing a digital twin model and dynamically simulating production and environmental impact; conflict coordination and collaborative fusion are completed by adopting a multi-agent model, and a global decision is output; the optimal transformation decision is obtained by combining multi-dimensional risk quantification and multi-objective optimization; and finally, an interpretable transformation path is intelligently generated in stages. In the process of digital transformation of mineral resource enterprises, multi-source data is collaboratively modeled and an interpretable transformation path in stages is generated, multiple risks are reduced, benefits and compliance are considered, and the landing and controllability of the digital transformation of the mine enterprise are enhanced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, equipment and medium for generating and assessing the digital transformation path of mining enterprises. Background Technology

[0002] Currently, most technologies related to digital transformation in mining enterprises adopt a localized, single-point application model, mainly relying on big data analysis, machine learning, and digital twins to optimize single aspects, such as monitoring equipment status, predicting energy consumption and carbon emissions, and simulating localized production processes. At the decision-making level, they still rely primarily on experience-based judgment, static report analysis, and single-objective cost-benefit accounting. Some studies attempt to introduce multi-objective optimization models to balance economic and environmental benefits, or to assess transformation investments through simple risk lists.

[0003] Existing technologies and decision-making methods generally have significant limitations, focusing primarily on single businesses or localized scenarios, and failing to form an enterprise-level systemic transformation framework. They rely on static models and experience-based judgments, making it impossible to integrate and dynamically extrapolate multi-source data such as production, energy consumption, emissions, finance, and policy. They lack multi-objective collaborative decision-making mechanisms, making it difficult to coordinate conflicts between different dimensions of objectives such as production, finance, carbon emissions, and risk. They have not established a multi-dimensional risk coupling and quantification system, making it impossible to dynamically assess risks such as technology adaptation, production disturbances, financial investment, and policy compliance. At the same time, transformation paths are mostly planned manually, lacking a phased, explainable, and optimizable intelligent generation mechanism, resulting in insufficient scientific rigor, poor implementation, and uncontrollable risks in transformation decisions.

[0004] Therefore, how to achieve intelligent generation of digital transformation paths for mining enterprises and dynamic quantitative assessment of multi-dimensional risks has become an urgent problem to be solved. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device, equipment and medium for generating and assessing the digital transformation path of mining enterprises, aiming to solve the technical problem of how to collaboratively model multi-source data and generate an interpretable phased transformation path during the digital transformation of mineral resource enterprises.

[0006] To achieve the above objectives, this application proposes a method for generating and assessing the digital transformation path of mining enterprises, including: Collect production and operation data, energy consumption data, carbon emission data, financial data, and environmental data from mining enterprises to obtain a multi-source heterogeneous dataset; The multi-source heterogeneous dataset is preprocessed, time-aligned, and feature-fused to obtain the enterprise state vector; A digital twin model is constructed based on the enterprise state vector, and the production process and environmental impact are dynamically simulated through the state transition function and environment mapping function of the digital twin model to obtain the simulation state sequence and environmental impact indicators. A multi-agent decision-making model is constructed, and the enterprise state vector and environmental impact indicators are input into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision result. Based on the enterprise state vector, environmental impact indicators and global decision results, multi-dimensional risk identification and quantification and multi-objective comprehensive evaluation and optimization are carried out to obtain the optimal transformation decision results and comprehensive risk level. Based on the optimal transformation decision results, enterprise state vector, environmental impact indicators, and comprehensive risk level, a phased path generation and optimization are performed to obtain an executable transformation path sequence and path interpretation results.

[0007] In one embodiment, the step of constructing a digital twin model based on the enterprise state vector, and dynamically simulating the production process and environmental impact through the state transition function and environment mapping function of the digital twin model to obtain the simulation state sequence and environmental impact indicators includes: Based on the enterprise state vector, a state space model of the enterprise production system is constructed to obtain the state representation space; A digital twin model is established based on the state representation space, and the state transition function is obtained; Based on the enterprise state vector, a mapping relationship between production state and environmental impact is constructed to obtain initial environmental impact indicators; The state evolution of the state representation space is deduced through the state transition function of the digital twin model to obtain the predicted state vector; Based on the predicted state vector and the initial environmental impact index, a multi-step iterative simulation is performed to obtain the simulation state sequence and environmental impact index.

[0008] In one embodiment, the step of performing multi-step iterative simulation based on the predicted state vector and initial environmental impact indicators to obtain the simulation state sequence and environmental impact indicators includes: The production control parameters for the current moment are determined based on the predicted state vector, and the control decision input is obtained. The predicted state vector at the next moment is obtained by performing state evolution calculation on the predicted state vector and the control decision input through the state transition function. The environmental impact index for the next moment is obtained by performing environmental impact mapping calculation on the predicted state vector at the next moment through the environmental mapping function. The predicted state vector at the next moment is used as the new predicted state vector, and the environmental impact index at the next moment is used as the new initial environmental impact index, until the preset simulation step size is reached, so as to obtain the predicted state vector and the set of environmental impact indexes at each moment. A simulation state sequence is constructed based on the predicted state vectors at each time point, and environmental impact indicators are obtained based on the set of environmental impact indicators at each time point.

[0009] In one embodiment, the step of constructing a multi-agent decision-making model and inputting the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain a global decision result includes: Construct a multi-agent set, wherein the multi-agent set includes a production optimization agent, a financial assessment agent, a carbon emission optimization agent, and a risk identification agent; The enterprise state vector and environmental impact index are input into each agent in the multi-agent set to obtain the initial state of each agent. The initial state of each agent is input into the decision policy function corresponding to each agent to obtain the decision output of each agent. A decision conflict metric function is constructed based on the decision outputs of each agent, and a weight vector is constructed based on the objective function of each agent to obtain the decision difference degree and weight coefficients. Based on the decision difference degree and weight vector, conflict adjustment and weighted fusion are performed on the decision outputs of each agent to obtain the global decision result.

[0010] In one embodiment, the step of performing multi-dimensional risk identification and quantification and multi-objective comprehensive evaluation and optimization based on the enterprise state vector, environmental impact indicators, and global decision results to obtain the optimal transformation decision result and comprehensive risk level includes: Construct a multi-dimensional risk indicator system, which includes technology risk indicators, financial risk indicators, production risk indicators, and compliance risk indicators; Based on the enterprise state vector, environmental impact indicators, and global decision-making results, a risk impact factor vector is constructed to obtain risk input characteristics; Based on the risk input characteristics and the multidimensional risk indicator system, risk quantification functions for various risks are established, and a risk coupling model is constructed based on the correlation between various risks to obtain the initial risk level; A set of candidate transformation solutions is generated based on the global decision results; A multi-objective evaluation function is constructed based on the enterprise state vector, environmental impact indicators, global decision-making results, and initial risk level. Based on the multi-objective evaluation function, the candidate transformation scheme set is optimized to obtain the Pareto optimal solution set. The optimal solution is selected by combining the Pareto optimal solution set with preset enterprise preference parameters to obtain the optimal transformation decision result and comprehensive risk level.

[0011] In one embodiment, the step of generating and optimizing a phased path based on the optimal transformation decision result, the enterprise state vector, environmental impact indicators, and comprehensive risk level to obtain an executable transformation path sequence and path interpretation results includes: Based on the optimal transformation decision results, a set of transformation tasks is extracted to obtain each transformation task; Based on the enterprise state vector, environmental impact indicators and comprehensive risk level, the priority of each transformation task is evaluated to obtain the task priority ranking result. Based on the task priority ranking results, each transformation task is sorted and grouped to obtain multiple stage task subsets; A phased transformation path is constructed based on the multiple phased task subsets, and the feasibility of the phased transformation path is verified to obtain an initial transformation path sequence. A path evaluation function is constructed based on the initial transformation path sequence, enterprise state vector, environmental impact indicators, and comprehensive risk level. Based on the path evaluation function, the initial transformation path sequence is optimized and adjusted, and key influencing factors are identified to obtain an executable transformation path sequence and path interpretation results.

[0012] In one embodiment, the step of optimizing and adjusting the initial transformation path sequence based on the path evaluation function and identifying key influencing factors to obtain an executable transformation path sequence and path interpretation results includes: The initial transformation path sequence is calculated based on the path evaluation function to obtain the evaluation value of each stage path; Based on the evaluation values ​​of each stage path, the initial transformation path sequence is optimized and adjusted to obtain an optimized set of stage paths. Based on the enterprise state vector, environmental impact indicators, and comprehensive risk level, a set of path impact factors is constructed to obtain the contribution of each impact factor; Based on the path evaluation function and the contribution of each influencing factor, a path interpretation function is constructed to obtain the contribution weight of each influencing factor to the path optimization adjustment. Based on the optimized set of stage paths and the contribution weights of each influencing factor to the path optimization adjustment, the paths are sorted to obtain an executable transformation path sequence and path interpretation results.

[0013] Furthermore, to achieve the above objectives, this application also proposes a device for generating and assessing the digital transformation path of mining enterprises, the device comprising: The data acquisition module is used to collect production and operation data, energy consumption data, carbon emission data, financial data and environmental data of mining enterprises to obtain multi-source heterogeneous datasets; The data processing module is used to preprocess, time-align, and fuse the multi-source heterogeneous dataset to obtain the enterprise state vector. The dynamic simulation module is used to construct a digital twin model based on the enterprise state vector, and to perform dynamic simulation of the production process and environmental impact through the state transition function and environment mapping function of the digital twin model, so as to obtain the simulation state sequence and environmental impact indicators. The collaborative decision-making module is used to construct a multi-agent decision-making model and input the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision-making result. The assessment and optimization module is used to perform multi-dimensional risk identification and quantification, and multi-objective comprehensive assessment and optimization based on the enterprise state vector, environmental impact indicators and global decision results, so as to obtain the optimal transformation decision results and comprehensive risk level. The path generation module is used to generate and optimize paths in stages based on the optimal transformation decision results, enterprise state vectors, environmental impact indicators and comprehensive risk levels, so as to obtain an executable transformation path sequence and path interpretation results.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the mining enterprise digital transformation path generation and risk assessment method described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the mining enterprise digital transformation path generation and risk assessment method described above.

[0016] This application uses multi-source heterogeneous data fusion to form an enterprise state vector, constructs a digital twin model, and dynamically simulates the impact of production and the environment. It employs a multi-agent model to achieve conflict coordination and collaborative integration, outputting global decisions. Combining multi-dimensional risk quantification and multi-objective optimization, it obtains the optimal transformation decision, ultimately intelligently generating a phased and interpretable transformation path. This enables collaborative modeling of multi-source data and generation of interpretable, phased transformation paths during the digital transformation of mineral resource enterprises, reducing multiple risks, balancing efficiency and compliance, and enhancing the feasibility and controllability of digital transformation in mining enterprises. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the method for generating and assessing the digital transformation path of mining enterprises in this application. Figure 2 This is a flowchart illustrating the second embodiment of the method for generating and assessing the digital transformation path of mining enterprises in this application. Figure 3 This is a schematic diagram of the module structure of the mining enterprise digital transformation path generation and risk assessment device in this application; Figure 4 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the method for generating and assessing the digital transformation path of mining enterprises in this application embodiment.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Currently, most technologies related to digital transformation in mining enterprises adopt a localized, single-point application model, mainly relying on big data analysis, machine learning, and digital twins to optimize single aspects, such as monitoring equipment status, predicting energy consumption and carbon emissions, and simulating localized production processes. At the decision-making level, they still rely primarily on experience-based judgment, static report analysis, and single-objective cost-benefit accounting. Some studies attempt to introduce multi-objective optimization models to balance economic and environmental benefits, or to assess transformation investments through simple risk lists.

[0023] Existing technologies and decision-making methods generally have significant limitations, focusing primarily on single businesses or localized scenarios, failing to form an enterprise-level systemic transformation framework. They rely on static models and experience-based judgments, unable to integrate and dynamically extrapolate multi-source data from production, energy consumption, emissions, finance, and policy sources. They lack multi-objective collaborative decision-making mechanisms, making it difficult to reconcile conflicts between different dimensions of objectives such as production, finance, carbon emissions, and risk. Furthermore, the absence of a multi-dimensional risk coupling and quantification system makes it impossible to dynamically assess risks related to technology adaptation, production disruptions, financial investment, and policy compliance. Simultaneously, transformation paths are mostly manually planned, lacking a phased, explainable, and optimizable intelligent generation mechanism, resulting in insufficient scientific rigor, poor implementation, and uncontrollable risks in transformation decisions. Therefore, how to achieve intelligent generation of digital transformation paths and dynamic quantitative assessment of multi-dimensional risks for mining enterprises has become an urgent problem to be solved.

[0024] Based on the above, this application also provides a method for generating and assessing the digital transformation path of mining enterprises, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for generating and assessing the digital transformation path of mining enterprises in this application.

[0025] In this embodiment, the method for generating and assessing the digital transformation path of mining enterprises includes steps S10 to S60: Step S10: Collect production and operation data, energy consumption data, carbon emission data, financial data, and environmental data of mining enterprises to obtain a multi-source heterogeneous dataset.

[0026] It should be noted that production operation data is obtained from the mining company's production equipment and control systems, and directly reflects the actual production status and operation of mining and processing. Energy consumption data is collected from the mining company's energy management system and records the consumption of various energy sources such as electricity and fuel throughout the production process. Carbon emission data is obtained based on emission monitoring information or calculation models and is used to characterize the level of greenhouse gas emissions in the mining production process. Financial data includes cost investment and revenue information, reflecting the financial status of the mining company related to digital transformation. Environmental data includes environmental policy requirements and production environment constraints, and is used to regulate mining production and transformation behavior.

[0027] Specifically, firstly, production operation data is obtained by reading equipment operating parameters, output data, and process indicators from the mining enterprise's production control system; secondly, energy consumption data is obtained by reading energy consumption records such as electricity, water, and gas from the energy management system; then, carbon emission data is obtained by reading emission data of greenhouse gases such as carbon dioxide from environmental monitoring equipment or carbon emission accounting systems; next, financial data is obtained by reading financial records such as costs, investments, and revenues from the enterprise resource planning system; finally, environmental data is obtained by reading constraints such as environmental policies and industry standards from public channels or industry databases. All of these data are then integrated into a multi-source heterogeneous dataset. This approach is necessary because the digital transformation of mining enterprises involves multiple dimensions, including production, energy, environment, finance, and policy. Only by uniformly collecting these data scattered across different systems can the overall operational status of the enterprise be fully reflected, providing a complete data foundation for subsequent state vector construction and digital twin modeling.

[0028] Step S20 involves preprocessing, temporal alignment, and feature fusion of the multi-source heterogeneous dataset to obtain the enterprise state vector.

[0029] Specifically, firstly, missing values ​​in the multi-source heterogeneous dataset are filled using interpolation, outliers exceeding the normal range are removed, and data of different dimensions are normalized to the same scale to obtain a standardized dataset. Secondly, production operation data, energy consumption data, carbon emission data, financial data, and environmental data are unified to the same time base using timestamp matching methods to obtain a synchronized data set. Then, key features are extracted from each type of data in the synchronized data set, such as output fluctuation features from production data, energy intensity features from energy consumption data, emission trend features from carbon emission data, cost structure features from financial data, and policy constraint intensity features from environmental data. Next, the extracted features are integrated into a unified feature space through concatenation or mapping to obtain a fused feature representation. Finally, the fused feature representation is constructed into a unified input data structure containing time dimension, feature dimension, and data type identifier to obtain the enterprise state vector. ,in Represents the feature fusion function. Indicates a time index. This represents production operation data after preprocessing, time alignment, and feature fusion. This represents the energy consumption data after preprocessing, time alignment, and feature fusion. This represents carbon emission data after preprocessing, time alignment, and feature fusion. This represents the financial data after preprocessing, time alignment, and feature fusion. This represents environmental data after preprocessing, time alignment, and feature fusion. This is necessary because the data collection frequency, storage format, and time base of various business systems within a mining enterprise differ. Without preprocessing and time alignment, data quality would be poor, time-discrepancies would occur, and the data would fail to accurately reflect the enterprise's operational status at any given moment. For example, a production system might record output every minute, while a financial system might calculate costs monthly. Directly concatenating these two sets of data would result in a time dimension mismatch, distorting the simulation results of the subsequent digital twin model. Feature fusion maps multi-source data to a unified space, making information from different dimensions comparable and calculable within the same coordinate system, providing standardized state input for the digital twin model.

[0030] Step S30: Construct a digital twin model based on the enterprise state vector, and use the state transition function and environment mapping function of the digital twin model to dynamically simulate the production process and environmental impact, thereby obtaining the simulation state sequence and environmental impact indicators.

[0031] It should be noted that step S30 includes: First, constructing a state-space model of the enterprise's production system based on the enterprise's state vectors to obtain the state representation space. It should be noted that the state-space model is based on the enterprise's state vectors and serves as a mathematical representation framework for comprehensively describing all operating states of the mining enterprise's production system. The state representation space is the space comprised of the entire set of states carried by the state-space model, that is, the set used to represent all possible operating states of the enterprise's production system. Specifically, based on the equipment operating parameters, output indicators, and process variables in the enterprise's state vectors, a high-dimensional state-space model containing all possible operating conditions is constructed to obtain the state representation space. ,in This indicates the duration of the data collection, i.e., the total duration of the data collection.

[0032] Secondly, a digital twin model is established based on the state representation space, and the state transition function is obtained. It should be noted that the state transition function describes the rules governing the transition of the enterprise's production system from the current state to the next state; that is, it is a function used to characterize the dynamic evolution of the production system. The digital twin model is a virtual model that maps the actual mine production system, i.e., a digital mapping model capable of reproducing the real production process and its dynamic changes. Specifically, the digital twin model is established based on the physical relationships and technological laws between adjacent states in the state representation space, and the state transition function is obtained. The specific function expression is as follows: in, This represents the enterprise state vector at time t+1, which is the simulation state sequence for the next time step. Represents the state transition function; This indicates the input for production control or scheduling decisions, such as equipment start / stop commands and process parameter adjustments. It clarifies which new state the system will evolve to after a certain time interval from its current operating state. For example, how will the energy consumption of the grinding mill change and how will the output be adjusted when the ore grade decreases?

[0033] Then, based on the enterprise state vector, a mapping relationship between production state and environmental impact is constructed to obtain the initial environmental impact index, the specific formula of which is as follows: in This indicates the initial environmental impact indicators, including carbon emissions and pollution emission levels; The function represents the mapping relationship.

[0034] Next, the state evolution of the state representation space is deduced through the state transition function of the digital twin model to obtain the predicted state vector. Finally, multi-step iterative simulation is performed based on the predicted state vector and the initial environmental impact indicators to obtain the simulation state sequence and environmental impact indicators. Specifically, the production control parameters at the current moment are determined based on the predicted state vector to obtain the control decision input; the state evolution calculation is performed on the predicted state vector and the control decision input through the state transition function to obtain the predicted state vector at the next moment; the environmental impact mapping calculation is performed on the predicted state vector at the next moment through the environmental mapping function to obtain the environmental impact indicators at the next moment; the predicted state vector at the next moment is used as the new predicted state vector, and the environmental impact indicators at the next moment are used as the new initial environmental impact indicators, until the preset simulation step size is reached to obtain the predicted state vectors and the set of environmental impact indicators at each moment; the simulation state sequence is constructed based on the predicted state vectors at each moment, and the environmental impact indicators are obtained based on the set of environmental impact indicators at each moment. Furthermore, firstly, key operating parameters such as equipment load rate, raw material ratio, and process temperature are extracted from the predicted state vector. These parameters are then converted into specific equipment start / stop commands and process adjustment values ​​according to preset production scheduling rules, yielding the control decision input. This is necessary because the state vector itself is only a numerical representation and must be translated into actual executable control actions to drive system evolution. Secondly, both the predicted state vector and the control decision input are substituted into the state transition function. The operating state of the system at the next moment is calculated according to the equipment dynamics equations and process constraints, yielding the predicted state vector for the next moment. For example, when the control decision input is to increase the mill speed, the state transition function will calculate the output and energy consumption values ​​at the next moment based on the mill's power curve and ore hardness parameters. Finally, the predicted state vector for the next moment is substituted into the environmental mapping function, based on the emission factor model and energy consumption carbon emission system... The simulation process involves calculating the corresponding carbon dioxide and pollutant emissions to obtain the environmental impact indicators for the next time step. Then, the predicted state vector for the next time step is used as the new current state, and the environmental impact indicators for the next time step are used as the new initial environmental impact indicators. This process of state evolution calculation and environmental impact mapping calculation is repeated until a preset simulation step size is reached, resulting in the predicted state vectors and sets of environmental impact indicators for each time step. This is done because mining production is a continuous dynamic process; single-step simulations can only reflect short-term changes, and only through multi-step iterations can medium- and long-term trends such as the cumulative effect of equipment aging and the transmission effect of inventory fluctuations be captured. Finally, the predicted state vectors for each time step are arranged in chronological order to form a simulation state sequence, and the environmental impact indicators for each time step are arranged in chronological order to form an environmental impact indicator sequence, resulting in the simulation state sequence and environmental impact indicators. The specific formulas are as follows: The simulated state sequence and environmental impact indicators for the next moment are obtained by evolving the current state through a state transition function and mapping the next moment's state through an environmental mapping function, respectively. This is done because the production process of mining enterprises is highly dynamic and time-delayed; changes in equipment status will have a chain reaction on energy consumption and emissions. Only by gradually deduce through the state transition function can this chain reaction be captured. For example, if future emissions are estimated statically based solely on the current state, the gradual process of efficiency decline caused by equipment aging will be ignored. However, multi-step iterative simulation can simulate the complete evolution chain of "equipment efficiency decreasing month by month → energy consumption increasing month by month → carbon emissions increasing month by month," enabling transformation decisions to be based on accurate predictions of future trends, rather than a one-sided assessment based solely on the current snapshot.

[0035] Step S40: Construct a multi-agent decision-making model, and input the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision result.

[0036] It should be noted that step S40 includes: constructing a multi-agent set; inputting the enterprise state vector and environmental impact indicators into each agent in the multi-agent set to obtain the initial state of each agent; inputting the initial state of each agent into the decision strategy function corresponding to each agent to obtain the decision output of each agent; constructing a decision conflict measurement function based on the decision output of each agent, and constructing a weight vector based on the objective function of each agent to obtain the decision difference degree and weight coefficient; and performing conflict adjustment and weighted fusion on the decision output of each agent based on the decision difference degree and weight vector to obtain the global decision result.

[0037] It's important to understand that the multi-agent ensemble includes production optimization agents, financial evaluation agents, carbon emission optimization agents, and risk identification agents. The production optimization agent is an independent decision-making unit aimed at improving production efficiency and stability; it's a dedicated unit responsible for optimizing calculations in the production process. The financial evaluation agent is an independent decision-making unit aimed at controlling costs and increasing profits; it's a dedicated unit responsible for financial calculations and evaluations. The carbon emission optimization agent is an independent decision-making unit aimed at reducing carbon emission levels; it's a dedicated unit responsible for optimizing calculations of low-carbon and environmental protection indicators. The risk identification agent is an independent decision-making unit aimed at identifying and mitigating transition risks; it's a dedicated unit responsible for risk monitoring and analysis.

[0038] Specifically, firstly, a multi-agent ensemble is constructed, configuring the production optimization agent to maximize output and equipment utilization, the financial evaluation agent to minimize costs and return on investment, the carbon emission optimization agent to minimize total carbon emissions, and the risk identification agent to minimize technological and compliance risks, thus obtaining the multi-agent ensemble. ,in, Indicates the first An intelligent agent. To determine the number of agents, each agent corresponds to a different decision-making objective, including production optimization agents, financial assessment agents, carbon emission optimization agents, and risk identification agents. This is because the digital transformation of mining enterprises involves multiple interdependent dimensions such as production, finance, environmental protection, and risk. A single agent cannot consider all objectives; therefore, agents in each specialized field must optimize their respective assigned objectives. Secondly, production operation data from the enterprise's state vector is input into the production optimization agent, energy consumption data and financial data into the financial assessment agent, carbon emission data into the carbon emission optimization agent, and environmental data into the risk identification agent. Simultaneously, environmental impact indicators are input to all four agents as constraints, resulting in the initial state of each agent. ,in, Indicates the first An intelligent agent at time Input status, This represents the state mapping function, which allows each agent to possess global information relevant to its decision rather than making isolated decisions. Then, a corresponding decision function is constructed for each agent, with the specific formula as follows: in, Indicates the first The decision output of an agent This represents the decision-making strategy function of the intelligent agents. Specifically, the production optimization agent outputs suggestions for expanding or reducing production based on the current output gap and equipment capacity margin; the financial evaluation agent outputs a funding allocation plan based on the investment budget and payback period requirements; the carbon emission optimization agent outputs the priority of technological transformation based on emission limits and emission reduction costs; and the risk identification agent outputs the risk warning level based on the policy tightening trend and equipment aging, thus obtaining the decision outputs of each agent. Next, the decision discrepancy between the production optimization agent and the carbon emission optimization agent, and between the financial evaluation agent and the risk identification agent, are calculated to construct a decision conflict measurement function, specifically expressed as: in Indicates the first The first agent and the second The degree of decision-making differences among individual agents This represents the decision difference measurement function. Weights are assigned based on the importance of each agent's objective to the overall success of the enterprise's transformation; for example, financial sustainability is assigned the highest weight, and compliance risk the second highest weight, resulting in the decision difference degree and weight coefficients. Finally, the decisions of agents with significant conflict are adjusted based on the decision difference degree to obtain the adjusted decision results for each agent. The specific formula is as follows: in, Indicates the first The adjusted decision results of each agent. This is the conflict adjustment coefficient, used to control the intensity of conflict correction. For example, when the production optimization agent demands increased production while the carbon emission optimization agent demands decreased production, the conflict adjustment coefficient is introduced to allow both to compromise. Then, the adjusted decisions of each agent are weighted and fused according to the weight coefficients to obtain the global decision result. The specific formula is expressed as follows: in This represents the globally optimal decision result. Simultaneously, the global decision result is also... Feedback is sent to each agent to update their decision policy functions. This is to achieve dynamic optimization of the multi-agent collaborative mechanism.

[0039] Step S50 involves multi-dimensional risk identification and quantification, and multi-objective comprehensive evaluation and optimization based on the enterprise state vector, environmental impact indicators, and global decision results, to obtain the optimal transformation decision result and comprehensive risk level.

[0040] It should be noted that step S50 includes: constructing a multi-dimensional risk indicator system; constructing a risk impact factor vector based on the enterprise state vector, environmental impact indicators, and global decision results to obtain risk input characteristics; establishing risk quantification functions for various risks based on the risk input characteristics and the multi-dimensional risk indicator system, and constructing a risk coupling model based on the correlation between various risks to obtain the initial risk level; generating a set of candidate transformation schemes based on the global decision results; constructing a multi-objective evaluation function based on the enterprise state vector, environmental impact indicators, global decision results, and initial risk level; performing multi-objective optimization on the set of candidate transformation schemes based on the multi-objective evaluation function to obtain a Pareto optimal solution set; and selecting the optimal scheme based on the Pareto optimal solution set combined with preset enterprise preference parameters to obtain the optimal transformation decision result and comprehensive risk level.

[0041] It's important to understand that the multi-dimensional risk indicator system is an assessment framework composed of various types of risk indicators. This set of indicators comprehensively covers all types of risks associated with the digital transformation of mining enterprises, including technological risk indicators, financial risk indicators, production risk indicators, and compliance risk indicators. Technological risk indicators measure the uncertainty in the application and adaptation of digital technologies, reflecting the difficulty and instability of technology implementation. Financial risk indicators measure the volatility of transformation investment costs and returns, reflecting the stability of capital investment and returns. Production risk indicators measure the likelihood of production disruptions and interruptions during the transformation process, reflecting the degree to which production continuity is affected. Compliance risk indicators measure the degree to which transformation activities conform to industry regulations, reflecting the degree of deviation from policy compliance. The risk impact factor vector is a vector composed of enterprise operation and decision-making-related characteristics, representing the core set of input parameters driving risk calculation. The risk quantification function is the calculation rule that transforms risk input characteristics into specific risk values, the core function that realizes the transformation of risk from qualitative description to quantitative representation. The risk coupling model is a model used to describe the mutual influence and correlation between different risks; that is, a mathematical model for comprehensive calculation of multiple types of risks. The initial risk level is the initial comprehensive risk value obtained through risk quantification and coupling calculation; that is, the calculation result reflecting the basic risk level of the transformation plan. The candidate transformation plan set is a set of multiple feasible transformation plans; that is, the overall set of alternative plans used to screen for the optimal decision. The multi-objective evaluation function is a comprehensive calculation function that simultaneously considers multiple evaluation objectives; that is, an evaluation tool used to uniformly measure the merits of plans. Multi-objective optimization is the computational process of finding the optimal plan under multiple objective constraints; that is, the solution operation that achieves multi-objective balance and plan selection. The Pareto optimal solution set is the set of non-dominated solutions that satisfy the multi-objective optimization conditions; that is, the overall set of optimal candidate solutions obtained from multi-objective optimization calculation. The preset firm preference parameter is a pre-set parameter reflecting the firm's decision-making tendency; that is, a personalized setting value used to guide the selection of the optimal plan.

[0042] Specifically, the technical risk indicator is set as the probability of implementation failure due to insufficient maturity of new technologies; the financial risk indicator is set as the possibility of investment exceeding the budget and a break in the capital chain; the production risk indicator is set as the probability of reduced production capacity and equipment downtime during the transformation period; and the compliance risk indicator is set as the possibility of penalties or rectification due to tightening environmental policies, thus constructing a multi-dimensional risk indicator system. ,in, Indicates the first Risk indicators include technical risk indicators, financial risk indicators, production risk indicators, and compliance risk indicators. This represents the number of risk categories. This is because the risks faced by mining companies undergoing digital transformation are not singular; errors in technology selection, insufficient funding, production interruptions, and failure to meet environmental standards can all lead to transformation failure, either individually or in combination. A comprehensive indicator framework is necessary for systematic identification. Secondly, data on equipment aging and technical personnel reserves are extracted from the enterprise status vector; data on the frequency of emissions exceeding standards is extracted from environmental impact indicators; and data on investment scale and technological complexity are extracted from global decision-making results. These data are combined into a risk impact factor vector to obtain the risk input characteristics, expressed by the following formula: in, Indicates risk input characteristics, This represents the risk feature mapping function. Represents the enterprise state vector. Indicates environmental impact indicators, This represents the overall decision result. Then, based on the corresponding dimension data in the risk input features, a corresponding risk quantification function is established. The specific formula is as follows: in Indicates the first Risks at all times The risk value, This represents the risk mapping function. Based on the cascading relationship between technological and production risks, and the transmission relationship between financial and compliance risks, a risk coupling model is constructed to obtain the initial risk level. The specific formula is as follows: in, The initial risk level is indicated because various risks do not exist in isolation. For example, immature technology can lead to frequent equipment failures, resulting in production risks; environmental penalties can increase additional expenditures, leading to financial risks. The risk coupling model can capture this chain reaction. Next, the transformation task in the overall decision result is decomposed into specific technological transformation plans, equipment upgrade plans, and process optimization plans, generating a set of candidate transformation plans, expressed by the following formula: in Indicates the first One candidate transformation plan For the number of schemes, This represents the set of candidate transformation solutions. Then, based on the current production capacity and cost baseline in the enterprise state vector, an economic benefit evaluation sub-function is constructed; based on the current emission status in the environmental impact indicators, an environmental impact evaluation function is constructed; and based on the initial risk level, a risk evaluation function is constructed. These three sub-functions are combined into a multi-objective evaluation function, expressed by the following formula: in Represents the economic return function, based on the firm's state vector. calculate; Represents the environmental impact function, based on environmental impact indicators. calculate; The risk function is based on the initial risk level. Calculation. Obtain a comprehensive evaluation standard that simultaneously measures benefits, environment, and risk.

[0043] Next, based on a multi-objective evaluation function, each option in the candidate transformation solution set is comprehensively scored in terms of economic benefits, environmental impact, and risk level. Through non-dominated ranking, the optimal solution that cannot be surpassed by other options in all three dimensions of benefits, environment, and risk is selected, resulting in the Pareto optimal solution set. Finally, based on the preset enterprise preference parameters of economic benefit weight, environmental impact weight, and risk tolerance threshold, the solution that best matches the enterprise's current strategic preferences is selected from the Pareto optimal solution set, resulting in the optimal transformation decision result and comprehensive risk level.

[0044] Step S60: Based on the optimal transformation decision results, enterprise state vector, environmental impact indicators and comprehensive risk level, a phased path generation and optimization is performed to obtain an executable transformation path sequence and path interpretation results.

[0045] Specifically, firstly, the technological transformation tasks, equipment upgrade tasks, and management optimization tasks in the optimal transformation decision are broken down one by one to form a set of transformation tasks containing task names, implementation content, and expected outputs, thus obtaining each transformation task. Secondly, current capacity utilization rate and remaining equipment life data are extracted from the enterprise state vector, emission margin and energy consumption baseline data are extracted from environmental impact indicators, and current level data of various risks are extracted from the comprehensive risk level. Based on this data, the priority of each transformation task is evaluated to obtain the task priority ranking result. For example, "replacing high-energy-consuming motors" is set as a high priority because the current energy consumption baseline is high and the remaining equipment life is short. Then, according to the task priority ranking result, high-priority and interdependent tasks are assigned to the first stage, medium-priority tasks that require previous results are assigned to the second stage, and low-priority tasks that can be implemented independently are assigned to the third stage, resulting in multiple stage task subsets. This is because mining enterprises cannot complete all transformation tasks at once and must proceed step by step according to urgency and interdependence. The process involves several steps: First, to avoid resource dispersion and disjointed implementation, a phased transformation path is constructed, including timelines, responsible parties, and acceptance criteria. The feasibility of this path is then verified based on the company's state vector (financial reserves and personnel allocation) and the risk threshold in the overall risk level. Tasks exceeding resource capacity or risk tolerance are eliminated, resulting in an initial transformation path sequence. Finally, a path evaluation function is built based on the initial transformation path sequence, the company's state vector, environmental impact indicators, and the overall risk level. This function assesses economic benefits, environmental impact, and risk levels, scoring and optimizing each phase of the initial transformation path sequence. For example, the investment scale for the first phase is lowered from the budget ceiling to the budget median to reduce financial risk. "Remaining equipment lifespan" and "emission margin" are identified as the key factors most influential on path optimization, and the contribution weight of each factor is recorded to obtain an executable transformation path sequence and path interpretation results. This approach is necessary because digital transformation of mining companies involves huge investments and long-term implementation. A one-time, comprehensive push based on the optimal decision could lead to a disruption in the transformation due to a broken cash flow or a concentrated outbreak of risks. For example, if a mining company simultaneously implements three high-investment tasks in the first phase—unmanned mining trucks, intelligent ore dressing systems, and remote monitoring platforms—while the long-term returns may be considerable, the short-term financial pressure is enormous and the technical risks are compounded. By generating and optimizing in stages, the unmanned mining trucks can be arranged in the first phase to verify the technical feasibility, the intelligent ore dressing system in the second phase to utilize the data accumulated in the first phase, and the remote monitoring platform in the third phase as a final integration. At the same time, the path interpretation results clearly indicate that "short remaining equipment lifespan" is the key reason for prioritizing the replacement of high-energy-consuming equipment, enabling management to understand the basis for their decision.

[0046] This embodiment uses multi-source heterogeneous data fusion to form an enterprise state vector, constructs a digital twin model, and dynamically simulates the impact of production and the environment. A multi-agent model is employed to achieve conflict coordination and collaborative integration, outputting a global decision. Combining multi-dimensional risk quantification and multi-objective optimization, the optimal transformation decision is obtained, ultimately generating a phased and interpretable transformation path. This enables collaborative modeling of multi-source data and generation of interpretable, phased transformation paths during the digital transformation of mineral resource enterprises, reducing multiple risks, balancing efficiency and compliance, and enhancing the feasibility and controllability of digital transformation in mining enterprises.

[0047] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 The method for generating and assessing the digital transformation path of mining enterprises, step S50, further includes steps S201 to S205: Step S201: Extract the set of transformation tasks based on the optimal transformation decision results to obtain each transformation task.

[0048] Specifically, firstly, the overall transformation goal in the optimal transformation decision is decomposed into independently executable specific task units, resulting in various transformation tasks. This is because the optimal transformation decision only provides a macro-level direction and must be broken down into clearly defined and responsible task units to allocate resources, schedule progress, and conduct performance evaluations. Secondly, each task is labeled with its technical dependencies and implementation prerequisites. Then, based on the technical constraints of each transformation task, tasks that are currently ready for implementation and those requiring preparatory work are selected. Tasks requiring preparatory work are categorized into a preliminary task list, while tasks that are ready for implementation are categorized into a core task list, resulting in categorized transformation tasks. Finally, each categorized transformation task is assigned a unique task identifier code, and a task attribute file is established, recording the expected investment scale, implementation cycle, responsible department, and expected output indicators for each task, resulting in a structured set of transformation tasks. This approach is necessary because the digital transformation of mining enterprises involves multi-departmental collaboration and significant resource investment. Without structured decomposition and attribute labeling of tasks, problems such as blurred task boundaries, shirking of responsibility, and loss of progress control can arise.

[0049] Step S202: Prioritize each transformation task based on the enterprise state vector, environmental impact indicators, and comprehensive risk level to obtain the task priority ranking result.

[0050] Specifically, firstly, data on the capacity utilization rate, equipment failure rate, and process bottleneck location of each production line are extracted from the enterprise state vector. Weak links with capacity utilization rates below a preset threshold (e.g., 70%) or equipment failure rates above a preset threshold (e.g., 15%) are identified, resulting in a production urgency score. This is done because neglecting to address bottleneck tasks will hinder the overall transformation effect. Secondly, carbon emission increment data and pollutant emission change data corresponding to each transformation task are extracted from environmental impact indicators. Tasks with emission increments exceeding the current emission margin (the difference between the upper limit allowed by environmental policy and the current emission amount) are marked as high environmental constraint priority, resulting in an environmental constraint score. Finally, technology maturity risk is extracted from the overall risk level. The system uses data on funding availability and policy change risks to calculate the risk exposure of each transformation task. Tasks with risk exposure exceeding a preset risk tolerance threshold (medium risk level) are appropriately downgraded in priority or require additional risk mitigation measures before reassessment, resulting in a risk adjustment coefficient. Next, production urgency scores, environmental constraints scores, and risk adjustment coefficients are input into a preset priority assessment function. These are weighted according to preset production weights (40%), environmental weights (30%), and risk weights (30%) to obtain a comprehensive priority value for each transformation task. Finally, the transformation tasks are ranked from highest to lowest based on their comprehensive priority values. Tasks with the same value are fine-tuned according to the principle of prioritizing those with shorter implementation cycles, resulting in a task priority ranking. This approach is necessary because mining companies have limited resources and cannot simultaneously advance all transformation tasks; therefore, dynamic prioritization based on the company's most pressing constraints is essential.

[0051] Step S203: Sort and group each transition task according to the task priority ranking result to obtain multiple stage task subsets.

[0052] Specifically, firstly, based on the task priority ranking, the top few tasks with the highest overall priority (preset to the maximum number of tasks in the first phase) are assigned to the first-phase task subset. The next highest priority tasks are assigned to the second-phase task subset, and the remaining tasks are assigned to the third-phase task subset, resulting in a preliminary grouping. This is because mining companies cannot invest all resources at once and must concentrate high-priority tasks in the preceding phases for quick results. Secondly, the technical dependencies between tasks in the preliminary grouping are checked. If subsequent phase tasks depend on the output of preceding phase tasks, the dependent tasks are moved to the preceding phase or later, resulting in a grouping result with adjusted dependencies. For example, if "intelligent mineral processing system deployment" is an example... Based on the results data from the "installation of online ore grade monitoring sensors," the latter is adjusted to the same or previous stage. Then, it is checked whether the total investment scale of each stage's task subset exceeds the company's current budget limit. If it does, non-critical tasks are moved to the next stage, resulting in a grouping result after budget constraint verification. Next, it is checked whether the implementation cycle of each stage's task subset is within the company's acceptable single-stage duration. If it does, tasks with low parallelism are split or moved later, resulting in a grouping result after duration constraint verification. Finally, the tasks of each stage, after triple verification of dependency, budget, and duration constraints, are solidified, and each stage's task subset is labeled with stage objectives, stage budget, and stage acceptance criteria, resulting in multiple stage task subsets. This is done because the digital transformation of mining companies needs to balance the goals of "rapid results" and "steady progress." If grouping is simply done by priority, problems may arise such as too many tasks in the first stage leading to resource dispersion, or implicit dependencies between tasks preventing the initiation of subsequent stages.

[0053] Step S204: Construct phased transformation paths based on multiple phased task subsets, and verify the feasibility of the phased transformation paths to obtain an initial transformation path sequence.

[0054] Specifically, firstly, a specific implementation sequence is determined for each task within the task subset of each stage. The task execution sequence is arranged according to the principle of infrastructure first, then application systems, and data collection first, then intelligent analysis. For example, in the first stage, "sensor network deployment" is arranged first, followed by "data collection platform launch," resulting in the task execution sequence for each stage. Secondly, based on the task execution sequence of each stage, the required human resources, equipment resources, and financial resources for each task are estimated, and a resource demand list for each stage is compiled. This list is then compared with the current resource reserve data in the enterprise state vector to obtain the resource gap analysis results. Then, based on the task execution sequence of each stage and the risk level data in the comprehensive risk level, high-risk events that may be triggered within each stage and their chain reaction paths are identified. For example, "new technologies" are identified. Supplier defaults could lead to equipment commissioning delays, which in turn could cause overall project delays, resulting in risk exposure maps for each stage. Next, based on resource gap analysis, tasks with insufficient resources are outsourced or procured in phases. Alternative solutions or insurance mechanisms are added for high-risk tasks based on the risk exposure maps. The adjusted task execution sequences for each stage are then rearranged to obtain optimized task execution sequences. Finally, the optimized task execution sequences are linked sequentially to form a phased transformation path that includes stage milestones, stage deliverables, and stage transition conditions. This phased transformation path undergoes a triple check of resource feasibility, technical feasibility, and risk controllability. Stages that do not meet the check criteria or require supplementary conditions are removed before inclusion, resulting in an initial transformation path sequence. This approach is necessary because a mining company's transformation path is not simply a collection of tasks, but rather an organic chain that considers resource integration, risk buffering, and stage transitions.

[0055] Step S205: Construct a path evaluation function based on the initial transformation path sequence, enterprise state vector, environmental impact indicators, and comprehensive risk level.

[0056] Specifically, firstly, current capacity baseline, cost baseline, and equipment health data are extracted from the enterprise state vector. The expected capacity increase and cost change at each stage after implementing the initial transformation path sequence are calculated. An economic benefit evaluation subfunction is constructed to obtain the economic benefit evaluation value for each stage of the path. This is done because the transformation path must quantify its contribution to financial performance; otherwise, it is impossible to determine whether the investment is worthwhile. Secondly, current carbon emission baseline and pollutant emission baseline are extracted from environmental impact indicators. The expected carbon emission reduction and pollutant emission reduction at each stage after implementing the initial transformation path sequence are calculated. An environmental impact evaluation subfunction is constructed to obtain the environmental impact evaluation value for each stage of the path. Then, technological risk, financial risk, and production risk are extracted from the comprehensive risk level. The current levels of production and compliance risks are assessed, and the changing trends of risk levels at each stage after implementing the initial transformation path sequence are evaluated. A risk level evaluation sub-function is constructed to obtain the risk level evaluation value for each stage, i.e., to determine whether the stage reduces risk or introduces new risk. Next, based on preset weights for economic benefits (40%), environmental impact (30%), and risk level (30%), the economic benefit evaluation value, environmental impact evaluation value, and risk level evaluation value are weighted and integrated to construct a comprehensive path evaluation function, obtaining the comprehensive evaluation value for each stage. Finally, the comprehensive evaluation value of each stage is compared with the preset path qualification threshold, and the stage paths that meet the qualification standard are selected, obtaining the effective output result of the path evaluation function. This is done because the transformation path of mining enterprises involves the trade-off of multi-dimensional objectives, and a single-dimensional evaluation will lead to one-sided decision-making.

[0057] Step S206: Optimize and adjust the initial transformation path sequence based on the path evaluation function and identify key influencing factors to obtain an executable transformation path sequence and path interpretation results.

[0058] It should be noted that step S206 includes: calculating the initial transformation path sequence based on the path evaluation function to obtain the evaluation value of each stage path; optimizing and adjusting each stage path in the initial transformation path sequence according to the evaluation value of each stage path to obtain the optimized stage path set; constructing a path impact factor set based on the enterprise state vector, environmental impact indicators, and comprehensive risk level to obtain the contribution of each impact factor; constructing a path interpretation function based on the path evaluation function and the contribution of each impact factor to the path optimization and adjustment to obtain the contribution weight of each impact factor to the path optimization and adjustment; and sorting the paths according to the optimized stage path set and the contribution weight of each impact factor to the path optimization and adjustment to obtain the executable transformation path sequence and path interpretation results.

[0059] Specifically, firstly, the first-stage path in the initial transformation path sequence is input into the path evaluation function to calculate the score of this stage in three dimensions: economic benefits, environmental impact, and risk level, thus obtaining the evaluation value of the first-stage path. Then, the same calculation is performed on the second and third-stage paths to obtain the evaluation values ​​for each stage. This is done because only by quantifying the comprehensive performance of each stage can we identify which stages need adjustment and which can be retained. Secondly, the difference between the evaluation value of each stage path and the preset optimization target value is compared. Stage paths with evaluation values ​​lower than the target value are marked as paths to be optimized, and the reasons for their low scores are analyzed. For example, if the second-stage path is lower than the target value due to "excessive investment scale leading to a low financial risk score," then the investment scale of this stage is reduced from the upper limit of the budget to the median budget, and the implementation period is extended to reduce financial pressure. Stage paths with evaluation values ​​reaching or exceeding the target value are directly retained, resulting in the optimized set of stage paths. Then, data on remaining equipment life and capacity utilization rate are extracted from the enterprise state vector, emission surplus and energy consumption base data are extracted from environmental impact indicators, and policy changes are extracted from the comprehensive risk level. Frequency and technology maturity data are used as candidate influencing factors. The sensitivity of each candidate influencing factor to the fluctuation of the path evaluation value is calculated. Factors with sensitivity exceeding a preset threshold (0.3) are included in the path influencing factor set to obtain the contribution of each influencing factor. For example, it is found that for every year the "remaining life of equipment" decreases, the corresponding path evaluation value decreases by 15%, so the factor has a high contribution. Next, the weight structure of the path evaluation function (economic benefit weight 40%, environmental impact weight 30%, risk level weight 30%) is correlated with the contribution of each influencing factor to construct a path interpretation function. This quantifies the driving role of each influencing factor in the path optimization and adjustment decision, and obtains the contribution weight of each influencing factor to the path optimization and adjustment. For example, it is found that "the contribution weight of short remaining life of equipment is 35%, which is the primary reason for prioritizing the replacement of high-energy-consuming equipment". Finally, the stages in the optimized stage path set are sorted according to the expected start time of each stage. The contribution weight of each influencing factor to the path optimization and adjustment is compiled into a decision basis explanation document and attached to the corresponding stage path to obtain the executable transformation path sequence and path interpretation results. This approach is necessary because mining company management needs to understand the reasons behind decisions to implement them effectively. Simply providing a "do A first, then B" path without explanation might lead to resistance or distorted execution by lower-level staff due to a lack of understanding. For example, a clear explanation stating that "the primary reason for prioritizing the replacement of high-energy-consuming motors is that the equipment has only two years of remaining lifespan and its current energy consumption exceeds the industry average by 40%. Failure to address this issue first will result in the double loss of equipment scrapping and energy consumption penalties next year" will convince management of the necessity of the decision. The equipment department will also understand that replacement is not about "following the trend" but about "avoiding risk," thus increasing willingness and accuracy in implementation.

[0060] This embodiment extracts transformation tasks based on the optimal transformation decision results. After priority evaluation, sorting, and grouping, a phased transformation path is constructed and its feasibility is verified. Then, the path evaluation function is used to optimize and adjust the path and identify key influencing factors. Finally, an executable transformation path sequence and path interpretation results are output. This enables the phased and structured deployment of transformation tasks, dynamically optimizes the path based on the company's actual situation and risk level, significantly improves the feasibility of transformation, reduces implementation disturbances, and clarifies the decision-making basis in an interpretable way, making the digital transformation of mining enterprises more scientific, controllable, and efficient.

[0061] Based on the first embodiment of this application, this application also provides a device for generating and assessing the digital transformation path of mining enterprises. Please refer to... Figure 3 The device includes: The data acquisition module 10 is used to collect production and operation data, energy consumption data, carbon emission data, financial data and environmental data of mining enterprises to obtain multi-source heterogeneous datasets.

[0062] The data processing module 20 is used to preprocess, time-align, and feature-fuse multi-source heterogeneous datasets to obtain enterprise state vectors.

[0063] The dynamic simulation module 30 is used to construct a digital twin model based on the enterprise's state vector, and to perform dynamic simulation of the production process and environmental impact through the state transition function and environment mapping function of the digital twin model, so as to obtain the simulation state sequence and environmental impact indicators.

[0064] The collaborative decision-making module 40 is used to construct a multi-agent decision-making model and input the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision result.

[0065] The evaluation and optimization module 50 is used to perform multi-dimensional risk identification and quantification, and multi-objective comprehensive evaluation and optimization based on the enterprise state vector, environmental impact indicators and global decision results, so as to obtain the optimal transformation decision results and comprehensive risk level.

[0066] The path generation module 60 is used to generate and optimize paths in stages based on the optimal transformation decision results, enterprise state vectors, environmental impact indicators and comprehensive risk levels, so as to obtain an executable transformation path sequence and path interpretation results.

[0067] The mining enterprise digital transformation path generation and risk assessment device provided in this application adopts the mining enterprise digital transformation path generation and risk assessment method in the above embodiments, which can solve the technical problem of how to collaboratively model multi-source data and generate interpretable phased transformation paths during the digital transformation of mineral resource enterprises. Compared with the prior art, the beneficial effects of the mining enterprise digital transformation path generation and risk assessment device provided in this application are the same as those of the mining enterprise digital transformation path generation and risk assessment method provided in the above embodiments, and other technical features in the mining enterprise digital transformation path generation and risk assessment device are the same as those disclosed in the above embodiments, and will not be repeated here.

[0068] This application provides a device for generating and assessing the digital transformation path of a mining enterprise. The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for generating and assessing the digital transformation path of a mining enterprise as described in Embodiment 1 above.

[0069] The following is for reference. Figure 4 The diagram illustrates a structural schematic of a mining enterprise digital transformation path generation and risk assessment device suitable for implementing embodiments of this application. The mining enterprise digital transformation path generation and risk assessment device in this application embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The mining enterprise digital transformation path generation and risk assessment equipment shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0070] like Figure 4As shown, the mining enterprise digital transformation path generation and risk assessment device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the mining enterprise digital transformation path generation and risk assessment device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the mining enterprise digital transformation path generation and risk assessment equipment to communicate wirelessly or wiredly with other devices to exchange data. Although various mining enterprise digital transformation path generation and risk assessment devices are shown in the figure, it should be understood that implementation or possession of all of them is not required. More or fewer may be implemented alternatively.

[0071] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0072] The mining enterprise digital transformation path generation and risk assessment device provided in this application adopts the mining enterprise digital transformation path generation and risk assessment method in the above embodiments, which can solve the technical problem of how to collaboratively model multi-source data and generate interpretable phased transformation paths during the digital transformation of mineral resource enterprises. Compared with the prior art, the beneficial effects of the mining enterprise digital transformation path generation and risk assessment device provided in this application are the same as those of the mining enterprise digital transformation path generation and risk assessment method provided in the above embodiments, and other technical features in the mining enterprise digital transformation path generation and risk assessment device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0073] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0074] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0075] This application provides a computer-readable medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the mining enterprise digital transformation path generation and risk assessment method in the above embodiments.

[0076] The computer-readable medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any combination thereof. More specific examples of computer-readable media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be executed by instructions, used by a device, or used in conjunction with it. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0077] The aforementioned computer-readable medium may be included in the mining enterprise digital transformation path generation and risk assessment equipment; or it may exist independently and not be installed in the mining enterprise digital transformation path generation and risk assessment equipment.

[0078] The aforementioned computer-readable medium carries one or more programs that, when executed by the mining enterprise digital transformation path generation and risk assessment device, enable the device to write computer program code for performing the operations of this application in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0079] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this application. In this regard, all blocks in the flowcharts or block diagrams may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that all blocks in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using dedicated hardware-based implementations that perform the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0080] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0081] The readable medium provided in this application is a computer-readable medium, which stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for generating and assessing the digital transformation path of mining enterprises. This method can solve the technical problem of how to collaboratively model multi-source data and generate an interpretable, phased transformation path during the digital transformation of mineral resource enterprises. Compared with the prior art, the beneficial effects of the computer-readable medium provided in this application are the same as those of the method for generating and assessing the digital transformation path of mining enterprises provided in the above embodiments, and will not be elaborated upon here.

[0082] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for generating and assessing the digital transformation path of mining enterprises.

[0083] The computer program product provided in this application can solve the technical problem of how to collaboratively model multi-source data and generate interpretable, phased transformation paths during the digital transformation of mineral resource enterprises. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the mining enterprise digital transformation path generation and risk assessment method provided in the above embodiments, and will not be repeated here.

[0084] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for generating and assessing the risks of digital transformation paths for mining enterprises, characterized in that, The method includes: Collect production and operation data, energy consumption data, carbon emission data, financial data, and environmental data from mining enterprises to obtain a multi-source heterogeneous dataset; The multi-source heterogeneous dataset is preprocessed, time-aligned, and feature-fused. The fused feature representation is constructed into a unified input data structure containing time dimension, feature dimension, and data type identifier to obtain the enterprise state vector. A digital twin model is constructed based on the enterprise state vector, and the production process and environmental impact are dynamically simulated through the state transition function and environment mapping function of the digital twin model to obtain the simulation state sequence and environmental impact indicators. A multi-agent decision-making model is constructed, and the enterprise state vector and environmental impact indicators are input into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision result. Based on the enterprise state vector, environmental impact indicators and global decision results, multi-dimensional risk identification and quantification and multi-objective comprehensive evaluation and optimization are carried out to obtain the optimal transformation decision results and comprehensive risk level. Based on the optimal transformation decision results, enterprise state vector, environmental impact indicators and comprehensive risk level, a phased path generation and optimization is performed to obtain an executable transformation path sequence and path interpretation results. The steps of constructing a multi-agent decision-making model and inputting the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain a global decision result include: Construct a multi-agent set, wherein the multi-agent set includes a production optimization agent, a financial assessment agent, a carbon emission optimization agent, and a risk identification agent; The enterprise state vector and environmental impact index are input into each agent in the multi-agent set to obtain the initial state of each agent. The initial state of each agent is input into the decision policy function corresponding to each agent to obtain the decision output of each agent. A decision conflict metric function is constructed based on the decision outputs of each agent, and a weight vector is constructed based on the objective function of each agent to obtain the decision difference degree and weight coefficients. Based on the decision difference degree and weight vector, conflict adjustment and weighted fusion are performed on the decision outputs of each agent to obtain the global decision result; The steps of performing multi-dimensional risk identification and quantification, and multi-objective comprehensive evaluation and optimization based on the enterprise state vector, environmental impact indicators, and global decision results to obtain the optimal transformation decision result and comprehensive risk level include: Construct a multi-dimensional risk indicator system, which includes technology risk indicators, financial risk indicators, production risk indicators, and compliance risk indicators; Based on the enterprise state vector, environmental impact indicators, and global decision-making results, a risk impact factor vector is constructed to obtain risk input characteristics; Based on the risk input characteristics and the multidimensional risk indicator system, risk quantification functions for various risks are established, and a risk coupling model is constructed based on the correlation between various risks to obtain the initial risk level; A set of candidate transformation solutions is generated based on the global decision results; A multi-objective evaluation function is constructed based on the enterprise state vector, environmental impact indicators, global decision-making results, and initial risk level. Based on the multi-objective evaluation function, the candidate transformation scheme set is optimized to obtain the Pareto optimal solution set. The optimal solution is selected by combining the Pareto optimal solution set with preset enterprise preference parameters to obtain the optimal transformation decision result and comprehensive risk level.

2. The method as described in claim 1, characterized in that, The steps of constructing a digital twin model based on the enterprise state vector, and dynamically simulating the production process and environmental impact using the state transition function and environment mapping function of the digital twin model to obtain the simulation state sequence and environmental impact indicators include: Based on the enterprise state vector, a state space model of the enterprise production system is constructed to obtain the state representation space; A digital twin model is established based on the state representation space, and the state transition function is obtained; Based on the enterprise state vector, a mapping relationship between production state and environmental impact is constructed to obtain initial environmental impact indicators; The state evolution of the state representation space is deduced through the state transition function of the digital twin model to obtain the predicted state vector; Based on the predicted state vector and the initial environmental impact index, a multi-step iterative simulation is performed to obtain the simulation state sequence and environmental impact index.

3. The method as described in claim 2, characterized in that, The step of performing multi-step iterative simulation based on the predicted state vector and initial environmental impact indicators to obtain the simulation state sequence and environmental impact indicators includes: The production control parameters for the current moment are determined based on the predicted state vector, and the control decision input is obtained. The predicted state vector at the next moment is obtained by performing state evolution calculation on the predicted state vector and the control decision input through the state transition function. The environmental impact index for the next moment is obtained by performing environmental impact mapping calculation on the predicted state vector at the next moment through the environmental mapping function. The predicted state vector at the next moment is used as the new predicted state vector, and the environmental impact index at the next moment is used as the new initial environmental impact index, until the preset simulation step size is reached, so as to obtain the predicted state vector and the set of environmental impact indexes at each moment. A simulation state sequence is constructed based on the predicted state vectors at each time point, and environmental impact indicators are obtained based on the set of environmental impact indicators at each time point.

4. The method as described in claim 1, characterized in that, The step of generating and optimizing a phased path based on the optimal transformation decision result, the enterprise state vector, environmental impact indicators, and comprehensive risk level to obtain an executable transformation path sequence and path interpretation results includes: Based on the optimal transformation decision results, a set of transformation tasks is extracted to obtain each transformation task; Based on the enterprise state vector, environmental impact indicators and comprehensive risk level, the priority of each transformation task is evaluated to obtain the task priority ranking result. Based on the task priority ranking results, each transformation task is sorted and grouped to obtain multiple stage task subsets; A phased transformation path is constructed based on the multiple phased task subsets, and the feasibility of the phased transformation path is verified to obtain an initial transformation path sequence. A path evaluation function is constructed based on the initial transformation path sequence, enterprise state vector, environmental impact indicators, and comprehensive risk level. Based on the path evaluation function, the initial transformation path sequence is optimized and adjusted, and key influencing factors are identified to obtain an executable transformation path sequence and path interpretation results.

5. The method as described in claim 4, characterized in that, The steps of optimizing and adjusting the initial transformation path sequence based on the path evaluation function and identifying key influencing factors to obtain an executable transformation path sequence and path interpretation results include: The initial transformation path sequence is calculated based on the path evaluation function to obtain the evaluation value of each stage path; Based on the evaluation values ​​of each stage path, the initial transformation path sequence is optimized and adjusted to obtain an optimized set of stage paths. Based on the enterprise state vector, environmental impact indicators, and comprehensive risk level, a set of path impact factors is constructed to obtain the contribution of each impact factor; Based on the path evaluation function and the contribution of each influencing factor, a path interpretation function is constructed to obtain the contribution weight of each influencing factor to the path optimization adjustment. Based on the optimized set of stage paths and the contribution weights of each influencing factor to the path optimization adjustment, the paths are sorted to obtain an executable transformation path sequence and path interpretation results.

6. A device for generating and assessing the digital transformation path of mining enterprises, characterized in that, The apparatus is applied to the method for generating and assessing the digital transformation path of mining enterprises as described in any one of claims 1-5, and the apparatus comprises: The data acquisition module is used to collect production and operation data, energy consumption data, carbon emission data, financial data and environmental data of mining enterprises to obtain multi-source heterogeneous datasets; The data processing module is used to preprocess, time-align, and fuse the multi-source heterogeneous dataset to obtain the enterprise state vector. The dynamic simulation module is used to construct a digital twin model based on the enterprise state vector, and to perform dynamic simulation of the production process and environmental impact through the state transition function and environment mapping function of the digital twin model, so as to obtain the simulation state sequence and environmental impact indicators. The collaborative decision-making module is used to construct a multi-agent decision-making model and input the enterprise state vector and environmental impact indicators into the multi-agent decision-making model for conflict coordination and collaborative fusion to obtain the global decision-making result. The assessment and optimization module is used to perform multi-dimensional risk identification and quantification, and multi-objective comprehensive assessment and optimization based on the enterprise state vector, environmental impact indicators and global decision results, so as to obtain the optimal transformation decision results and comprehensive risk level. The path generation module is used to generate and optimize paths in stages based on the optimal transformation decision results, enterprise state vectors, environmental impact indicators and comprehensive risk levels, so as to obtain an executable transformation path sequence and path interpretation results.

7. A device for generating and assessing the digital transformation path of mining enterprises, characterized in that, The device includes: a memory, a processor, and a mining enterprise digital transformation path generation and risk assessment program stored on the memory and running on the processor, wherein the mining enterprise digital transformation path generation and risk assessment program is configured to implement the steps of the mining enterprise digital transformation path generation and risk assessment method as described in any one of claims 1-5.

8. A storage medium, characterized in that, The storage medium stores a program for generating and assessing the digital transformation path of mining enterprises. When the processor executes the program, it implements the steps of the method for generating and assessing the digital transformation path of mining enterprises as described in any one of claims 1-5.

Images