Electricity-coal performance management and control method and system for realizing three-level application by coal mine management department

Abstract

The invention discloses an electricity-coal performance management and control method and system for realizing three-level application by a coal mine management department, and relates to the technical field of coal supply chain management. According to the invention, the problems of insufficient data processing capability and poor real-time performance are solved; through a distributed database, an intelligent contract and a block chain technology, a trust mechanism of multi-party collaboration is enhanced, task distribution is optimized, and the defect of a traditional method in collaboration efficiency is made up; meanwhile, resource scheduling optimization and dynamic adjustment are realized by using linear programming and time sequence analysis technologies, so that the flexibility and adaptability of the supply chain are improved, the accuracy of performance evaluation and the timeliness of contract execution are further guaranteed, efficient, stable and accurate management of the coal supply chain is realized, and the coal supply chain management efficiency is improved. And powerful support is provided for energy guarantee and enterprise economic benefit improvement.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal supply chain management, in particular to an electric coal performance management and control method and system realized by a three-level application of a coal mine management department. BACKGROUND

[0002] Coal supply chain management plays a key role in energy security and economic stability, and its efficient operation is directly related to the continuous supply of the power industry and the economic benefits of enterprises. However, the traditional coal supply chain management method has significant shortcomings in data processing, collaboration efficiency and dynamic adjustment capability. Many systems rely on manual operation and offline data, making it difficult to monitor production and inventory status in real time, resulting in delayed performance and wasted resources. In addition, the existing method lacks an effective trust mechanism and data sharing method in multi-party collaboration, limiting the overall optimization of the supply chain. In this context, real-time data synchronization and multi-party collaboration have become core challenges that need to be addressed. If coal production and inventory data cannot be synchronized at the millisecond level, the accuracy of performance evaluation will decrease, thereby affecting the timeliness of contract execution. SUMMARY

[0003] The purpose of the present application is to provide an electric coal performance management and control method and system realized by a three-level application of a coal mine management department, which solves the problems of insufficient data processing, low collaboration efficiency and weak dynamic adjustment capability in traditional coal supply chain management, and realizes efficient, stable and accurate coal supply chain management.

[0004] The purpose of the present application can be achieved by the following technical solutions: The present application provides an electric coal performance management and control method realized by a three-level application of a coal mine management department, comprising the following steps: Real-time data acquisition of coal production and inventory data is performed by sensors and Internet of Things devices to generate a first data set containing time stamps, production volume and inventory volume. The data with a time stamp interval exceeding a preset millisecond-level threshold is compressed and preprocessed by an edge computing node to obtain a first processed data set meeting the millisecond-level synchronization requirement; According to the first processed data set, a distributed database technology is used to store production and inventory information to generate a second data set containing real-time status. An intelligent contract technology is used to calculate a performance evaluation score in real time for abnormal data with an inventory volume below a preset threshold or a production volume fluctuation exceeding a preset range to obtain a first performance evaluation result; A blockchain technology is used to verify and share the first performance evaluation result to generate a third data set containing a credit score. When the credit score in the third data set is lower than a preset credit threshold, a reinforcement learning algorithm is used to adjust the task allocation weight to obtain a first task allocation scheme; Based on the first task allocation scheme, a linear programming algorithm is used to optimize resource scheduling and generate a fourth dataset containing scheduling priorities. When the scheduling priorities in the fourth dataset change, adjustment instructions are pushed to relevant nodes through message queue technology to obtain the first execution result. Based on the first execution result, time series analysis technology is used to predict future inventory and production trends, generating a fifth dataset for iterative optimization.

[0005] Furthermore, a first processing dataset that meets the millisecond-level synchronization requirements is obtained, specifically including: When the timestamp interval of the first dataset exceeds the preset threshold, the timestamp sequence is obtained through the edge computing node, the difference is calculated, the timestamp pairs exceeding the threshold are determined, the corresponding data segments are extracted using the sliding window method, and data compression is performed to obtain the compressed dataset. If the timestamp interval of the compressed dataset still exceeds the preset threshold, the timestamp sequence is adjusted by a linear interpolation algorithm to generate a calibration dataset. Then, the mean filtering method is used to preprocess the data features to obtain a smooth dataset. When the synchronization error of the smoothed dataset is lower than a preset threshold, the data is sharded and stored through edge computing nodes to generate the first processing dataset. Based on the timestamp sequence of the first processing dataset, a verification algorithm is used to detect the synchronization consistency and determine the final synchronized dataset.

[0006] Furthermore, a second dataset containing real-time states is generated, specifically including: Production and inventory data are obtained from the first dataset and stored in a structured manner through a distributed database to obtain an initial data table. When the recording time of production and inventory data in the initial data table is later than a preset threshold, a real-time status update is triggered to generate an intermediate dataset containing status identifiers. Data processing algorithms are used to deduplicate and standardize the format of the intermediate dataset to obtain a standard dataset. Then, based on the state identifiers in the standard dataset, a time series analysis algorithm is used to extract real-time state features and generate a state feature set. When the feature values ​​in the state feature set exceed the preset range, a correlation analysis is performed on the production data and inventory data to determine data consistency and obtain a consistency result set. By merging the consistency result set with the standard dataset, a second dataset containing real-time status is generated.

[0007] Furthermore, the first performance evaluation results were obtained, specifically including: Inventory and production data are obtained from the second dataset. Data cleaning techniques are used to remove outliers to obtain a cleaned dataset. When the inventory in the cleaned dataset is lower than a preset threshold, the inventory is marked as abnormal by threshold judgment techniques to obtain an inventory abnormality label. When the production volume in the cleaned dataset fluctuates beyond the preset range, the fluctuation amplitude is calculated using fluctuation detection technology to obtain a fluctuation anomaly marker. Based on inventory anomaly markers and fluctuation anomaly markers, a logistic regression algorithm is used to predict the probability of performance risk and obtain a performance risk score. Then, through smart contract technology, performance evaluation parameters are obtained from the performance risk score, the score calculation logic is executed, and the first performance evaluation result is obtained. The performance evaluation score is extracted from the first performance evaluation result, recorded on the blockchain using data storage technology, and a blockchain storage identifier is obtained. Then, the query index of the performance evaluation score is obtained, a performance evaluation query key is generated, and the query key result is obtained.

[0008] Furthermore, a third dataset containing credit scores is generated, specifically including: Initial performance data is obtained from the performance evaluation system, a data fingerprint is generated using a hash algorithm, and stored in the blockchain network to obtain the first data hash value. The first data hash value is then verified by multiple parties through verification nodes in the blockchain network. When the consistency of the verification nodes reaches a preset threshold, a verification pass identifier is generated. Based on the verification pass identifier, key fields are extracted from the initial performance data, and a weighted average algorithm is used to calculate the credit score to obtain the first credit score set. Then, through the smart contract of the blockchain network, the first credit score set is bound with the verification pass identifier to generate the encrypted second dataset. Credit scores and associated performance data are obtained from the second dataset. A data consistency check algorithm is used to determine the data integrity and obtain the consistency verification result. Then, through the sharing mechanism of the blockchain network, the second dataset is distributed to authorized nodes to generate a third dataset containing credit scores. Credit scores and performance data are extracted from the third dataset, and data access logs are recorded through the automatic update mechanism of smart contracts to obtain shared data tracking records.

[0009] Furthermore, the first task allocation scheme is obtained, which specifically includes: When the credit score of the third dataset is lower than the preset credit threshold, the data filtering module extracts the subset of data below the threshold to obtain the first subset of data. Based on the credit score distribution in the first subset of data, the statistical analysis method is used to calculate the initial weight allocation of each task to obtain the first weight set. The first weight set is iteratively optimized by the reinforcement learning algorithm, and the task allocation weights are adjusted to obtain the second weight set. When the weight values ​​in the second weight set meet the preset convergence condition, the task allocation scheme is generated according to the second weight set to obtain the first task allocation scheme. The task allocation module maps the first task allocation scheme to specific task nodes to obtain the first task execution sequence. Then, the scheduling algorithm is used to optimize the task execution order to obtain the second task execution sequence. By executing the second task execution sequence, the final task allocation result is generated, and the first task allocation record is obtained.

[0010] Furthermore, a fourth dataset containing scheduling priorities is generated, specifically including: The linear programming algorithm is used to obtain the task execution order and allocation constraints from the task allocation scheme, obtain the optimization objective function, and then calculate the resource utilization and scheduling execution efficiency to determine the resource scheduling optimization scheme. When the resource scheduling optimization scheme meets the allocation constraints, a priority ranking is generated according to the task execution order to obtain the initial scheduling priority. The priority ranking is then iteratively adjusted to determine whether it meets the scheduling execution efficiency requirements, and the adjusted scheduling priority is obtained. By adjusting the scheduling priority, relevant data is extracted from the task allocation scheme to generate a fourth dataset. Then, a data verification algorithm is used to determine whether the generated dataset meets the allocation constraints, and the final dataset is obtained. Using the final dataset, a priority sorting algorithm is employed to verify the scheduling priorities and determine the final scheduling priorities.

[0011] Furthermore, the first execution result is obtained, specifically including: When the scheduling priority of the fourth dataset changes, the dataset content is parsed through a preset message queue protocol to determine the priority adjustment instruction. The priority adjustment instruction is then pushed to the relevant nodes through the message queue, and the nodes receive confirmation. Once the node receives the confirmation, it updates the node task scheduling configuration according to the adjustment instruction, obtains the node configuration update result, extracts the execution feedback data from the node configuration update result, and uses the consistent hashing algorithm to allocate the feedback data to the storage cluster to determine the data storage location. The system retrieves feedback data based on the storage location, determines whether the feedback data meets the preset priority adjustment threshold, obtains the verification result, and pushes a successful execution command to the control center through the message queue to obtain the first execution result. Update the priority record of the fourth dataset based on the first execution result to determine the final scheduling state.

[0012] Furthermore, a fifth dataset is generated, specifically including: Historical inventory and production data are obtained from the first execution result. Data cleaning techniques are used to remove outliers and missing values ​​to obtain a standardized dataset. Then, time series analysis techniques are used to build an ARIMA model to fit the historical data and obtain a trend prediction model. By using trend prediction models, future inventory change trends are predicted, generating an inventory trend dataset; similarly, future production demand change trends are predicted, generating a production trend dataset. By employing a merging technique, the inventory trend dataset and the production trend dataset are integrated to generate a fifth dataset. When the prediction error of the fifth dataset exceeds a preset threshold, an iterative optimization mechanism is used to adjust the ARIMA model parameters and regenerate the fifth dataset, resulting in an optimized fifth dataset. The optimized fifth dataset is used to update inventory management and production planning data, generating an adjusted management dataset.

[0013] This application provides a coal mine management department's three-tiered coal supply contract management system, which includes methods for implementing three-tiered coal supply contract management by coal mine management departments. The data acquisition and preprocessing module collects coal production and inventory data in real time through sensors and IoT devices, generating a first dataset containing timestamps, production volume and inventory volume. Then, for data whose timestamp interval exceeds a preset threshold, edge computing nodes are used to compress and preprocess the data to form the first processed dataset. The performance evaluation calculation module, based on the first processing dataset, uses distributed database technology to store production and inventory information, generates a second dataset containing real-time status, analyzes abnormal data such as inventory levels below a preset threshold or production fluctuations exceeding the standard through smart contract technology, predicts the probability of performance risk by combining logistic regression algorithm, calculates the performance evaluation score, and obtains the first performance evaluation result. The credit assessment and task allocation module uses blockchain technology to verify and share the first performance evaluation results from multiple parties, generating a third dataset containing credit scores. When the credit score is lower than a preset threshold, the module uses reinforcement learning algorithms to iteratively optimize the task allocation weights, generates a first task allocation scheme based on the optimized weights, maps the scheme to specific task nodes, optimizes the task execution order, and determines the final task allocation result. The resource scheduling optimization module optimizes resource scheduling using a linear programming algorithm based on the first task allocation scheme, generates a fourth dataset containing scheduling priorities, and pushes adjustment instructions to relevant nodes through message queue technology when scheduling priorities change, and updates the scheduling configuration based on node feedback. The trend prediction and cycle optimization module collects and cleans historical inventory and production data based on the first execution result of resource scheduling, builds an ARIMA model to predict future inventory and production trends, and generates a fifth dataset. When the prediction error exceeds the threshold, the model parameters are adjusted and the prediction is re-made. The optimized dataset is then used to update inventory management and production plans.

[0014] The beneficial effects of this invention are as follows: This invention collects coal production and inventory data in real time through sensors and IoT devices, and uses edge computing nodes for data compression and preprocessing to achieve millisecond-level data synchronization. This solves the problems of insufficient data processing capabilities and difficulty in real-time monitoring of production and inventory status in traditional methods, providing an accurate and timely data foundation for the efficient operation of the supply chain and ensuring the stability of energy supply. By using a distributed database to store data and combining smart contracts and blockchain technology to verify and share performance evaluation results among multiple parties, trust among all participants in the supply chain is enhanced, task allocation is optimized, and the shortcomings of traditional methods in multi-party collaboration, such as the lack of effective trust mechanisms and data sharing methods, are solved. This improves the operational efficiency and resource utilization efficiency of the supply chain, and enhances the economic benefits and market competitiveness of enterprises. By optimizing resource scheduling through linear programming algorithms and using time series analysis techniques to predict future inventory and production trends, dynamic adjustment and cyclical optimization of the supply chain are achieved. This solves the shortcomings of traditional methods in terms of dynamic adjustment capabilities, enables timely responses to market changes and production uncertainties, further improves the accuracy of performance evaluation, ensures the timeliness of contract execution, and achieves efficient and stable operation of the coal supply chain. Attached Figure Description

[0015] To better understand and implement this application, the technical solution is described in detail below with reference to the accompanying drawings.

[0016] Figure 1 A flowchart illustrating the coal mine management department's method for implementing three-level application of coal performance control in Embodiment 1 of this application; Figure 2 A flowchart illustrating the process of obtaining a first processing dataset that meets millisecond-level synchronization requirements for the coal mine management department's three-level application coal performance control method provided in Embodiment 1 of this application; Figure 3 A flowchart illustrating the process of generating a third dataset containing credit scores for the coal mine management department's three-level application coal performance control method provided in Embodiment 1 of this application; Figure 4 This is a schematic diagram of the structure of the coal mine management department's three-level application coal performance control system provided in Embodiment 2 of this application. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and systems consistent with some aspects of this application as detailed in the appended claims.

[0018] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0019] The following detailed description of the specific implementation methods, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided in detail.

[0020] Example 1

[0021] Please see Figures 1-3 This embodiment provides a method for coal mine management departments to implement three-level application of coal performance contract control, including the following steps: S1. Real-time data collection of coal production and inventory data is performed through sensors and IoT devices to generate a first dataset containing timestamps, production volume and inventory volume. Then, data with timestamp intervals exceeding a preset millisecond threshold is compressed and preprocessed through edge computing nodes to obtain a first processed dataset that meets the millisecond synchronization requirements. Furthermore, a first dataset containing timestamps, production quantities, and inventory quantities is generated, specifically including: Coal production and inventory data are acquired through IoT sensors. A first dataset containing production and inventory is generated by recording timestamps. If the first dataset has missing timestamps, the missing timestamps are supplemented through a data transmission mechanism to obtain a complete first dataset. Based on the complete first dataset, production and inventory are classified and stored using a data storage structure to determine the classified dataset. Then, the changing trends of production and inventory are obtained, and the K-means algorithm is used to cluster and analyze the trend features to obtain the trend clustering results. When the trend clustering results show that the deviation between production and inventory exceeds the preset threshold, the data collection frequency is adjusted through sensor equipment management to obtain an optimized dataset. Then, a linear regression algorithm is used to predict future production and inventory to determine the prediction dataset. By predicting the dataset, the matching relationship between production and inventory is obtained, a dynamic adjustment strategy is generated, and the first dataset containing timestamps, production and inventory is obtained.

[0022] Specifically, real-time coal production and inventory data are collected through IoT sensors, with a timestamp supplementation mechanism to ensure data integrity. K-means algorithm is used to analyze data trends, and linear regression algorithm is used to predict changes in production and inventory. This ultimately generates a dynamic adjustment strategy and a complete first dataset. This not only achieves real-time and complete collection of production and inventory data but also proactively identifies abnormal deviations in production and inventory based on data analysis. By adjusting the data collection frequency and predicting future trends, it provides accurate and dynamic data support for coal performance contract management, effectively avoiding performance risks caused by data lag or inaccuracy. The data transmission mechanism combines redundant transmission with verification, implementing dual verification at both the sending and receiving ends. When a timestamp is missing, it is quickly supplemented through redundant backup data. The data storage structure adopts a distributed columnar storage structure, classifying and storing production and inventory quantities according to dimensions such as production stage and inventory location.

[0023] Furthermore, a first processing dataset that meets the millisecond-level synchronization requirements is obtained, specifically including: S11. When the timestamp interval of the first dataset exceeds the preset threshold, the timestamp sequence is obtained through the edge computing node, the difference is calculated, the timestamp pairs exceeding the threshold are determined, the corresponding data segments are extracted using the sliding window method, and data compression is performed to obtain the compressed dataset. S12. When the timestamp interval of the compressed dataset still exceeds the preset threshold, the timestamp sequence is adjusted by a linear interpolation algorithm to generate a calibration dataset. Then, the mean filtering method is used to preprocess the data features to obtain a smooth dataset. S13. When the synchronization error of the smoothed dataset is lower than the preset threshold, the data is sharded and stored through edge computing nodes to generate the first processing dataset. Based on the timestamp sequence of the first processing dataset, a verification algorithm is used to detect the synchronization consistency and determine the final synchronization dataset.

[0024] Specifically, the first dataset was processed in multiple stages through edge computing nodes, achieving millisecond-level data synchronization. First, data compression using interpolation and sliding windows effectively reduced the data volume and improved transmission efficiency. Next, the timestamp sequence was calibrated using a linear interpolation algorithm, and data features were preprocessed using mean filtering to ensure the accuracy of the data's time dimension and the stability of the data itself. Finally, the application of sharded storage and verification algorithms not only improved data storage and retrieval efficiency but also ensured the consistency and reliability of the final synchronized dataset, providing a high-quality real-time data foundation for accurate analysis and decision-making in subsequent coal supply contract management.

[0025] The preset threshold can be determined by combining statistical analysis and expert experience based on historical fluctuations in coal production and inventory data and business needs. The difference calculation can be performed by subtracting adjacent timestamps and comparing the result with the preset threshold to determine whether the value exceeds the limit. The sliding window size is dynamically adjusted based on the data acquisition frequency and the allowable data delay. The linear interpolation algorithm uses common methods such as Lagrange interpolation or Newton interpolation to perform interpolation calculations based on known timestamps and corresponding data. The mean filtering method can select an appropriate odd-numbered window based on the data noise level, and the weights can be uniformly distributed. The verification algorithm uses common algorithms such as hash verification or CRC verification, and the rules for generating and comparing verification codes are clearly defined, making the data processing process clearer and more operable.

[0026] S2. Based on the first processing dataset, use distributed database technology to store production and inventory information, generate a second dataset containing real-time status, and use smart contract technology to calculate the performance evaluation score in real time for abnormal data such as inventory levels below a preset threshold or production fluctuations exceeding a preset range, and obtain the first performance evaluation result. Furthermore, a second dataset containing real-time states is generated, specifically including: Production and inventory data are obtained from the first dataset and stored in a structured manner through a distributed database to obtain an initial data table. When the recording time of production and inventory data in the initial data table is later than a preset threshold, a real-time status update is triggered to generate an intermediate dataset containing status identifiers. Data processing algorithms are used to deduplicate and standardize the format of the intermediate dataset to obtain a standard dataset. Then, based on the state identifiers in the standard dataset, a time series analysis algorithm is used to extract real-time state features and generate a state feature set. When the feature values ​​in the state feature set exceed the preset range, a correlation analysis is performed on the production data and inventory data to determine data consistency and obtain a consistency result set. By merging the consistency result set with the standard dataset, a second dataset containing real-time status is generated.

[0027] Specifically, by extracting production and inventory data from the first dataset, and utilizing distributed database structured storage, real-time status update mechanisms, data cleaning and feature extraction, and correlation analysis, a second dataset containing real-time status is generated. This enables real-time storage and dynamic updating of coal production and inventory data, quickly identifies data anomalies, and ensures data accuracy and reliability through deduplication, normalization, and consistency analysis. This provides timely and high-quality data support for the evaluation of coal performance contracts, helping coal mine management departments to grasp the production and inventory status in real time and adjust performance strategies accordingly.

[0028] Furthermore, the first performance evaluation results were obtained, specifically including: Inventory and production data are obtained from the second dataset. Data cleaning techniques are used to remove outliers to obtain a cleaned dataset. When the inventory in the cleaned dataset is lower than a preset threshold, the inventory is marked as abnormal by threshold judgment techniques to obtain an inventory abnormality label. When the production volume in the cleaned dataset fluctuates beyond the preset range, the fluctuation amplitude is calculated using fluctuation detection technology to obtain a fluctuation anomaly marker. Based on inventory anomaly markers and fluctuation anomaly markers, a logistic regression algorithm is used to predict the probability of performance risk and obtain a performance risk score. Then, through smart contract technology, performance evaluation parameters are obtained from the performance risk score, the score calculation logic is executed, and the first performance evaluation result is obtained. The performance evaluation score is extracted from the first performance evaluation result, recorded on the blockchain using data storage technology, and a blockchain storage identifier is obtained. Then, the query index of the performance evaluation score is obtained, a performance evaluation query key is generated, and the query key result is obtained.

[0029] Specifically, it enables a quantitative assessment of coal supply performance. Data cleaning removes outliers to ensure data quality; threshold judgment and fluctuation detection accurately identify inventory and production anomalies; logistic regression algorithms scientifically predict performance risks; and smart contract technology ensures the objectivity and transparency of the evaluation process. Finally, the performance evaluation score is stored on the blockchain and a query key is generated. This not only guarantees the credibility and immutability of the evaluation results but also facilitates subsequent traceability and querying, providing a strong basis for coal mine management departments to supervise coal supply performance and make decisions.

[0030] S3. The first performance evaluation result is verified and shared by multiple parties through blockchain technology to generate a third dataset containing credit scores. When the credit score in the third dataset is lower than the preset credit threshold, the task allocation weight is adjusted by reinforcement learning algorithm to obtain the first task allocation scheme. Furthermore, a third dataset containing credit scores is generated, specifically including: S31. Obtain initial performance data from the performance evaluation system, generate data fingerprints using a hash algorithm, store them in the blockchain network to obtain the first data hash value, and perform multi-party verification of the first data hash value through verification nodes in the blockchain network. When the consistency of the verification nodes reaches a preset threshold, a verification pass identifier is generated. S32. Based on the verification pass identifier, extract key fields from the initial performance data, calculate the credit score using a weighted average algorithm to obtain the first credit score set, and then bind the first credit score set with the verification pass identifier through the smart contract of the blockchain network to generate the encrypted second dataset. S33. Obtain the credit score and associated performance data from the second dataset, use a data consistency check algorithm to determine the data integrity, obtain the consistency verification result, and then distribute the second dataset to authorized nodes through the sharing mechanism of the blockchain network to generate a third dataset containing the credit score. S34. Extract credit scores and performance data from the third dataset, and record data access logs through the automatic update mechanism of smart contracts to obtain shared data tracking records.

[0031] Specifically, through a series of operations including hash algorithms, multi-party verification, weighted average calculation, smart contract binding, and data consistency checks, the scientific generation and reliable storage and sharing of credit scores are achieved. Hash algorithms and multi-party verification ensure the authenticity and immutability of initial performance data; the weighted average algorithm accurately calculates credit scores; smart contract binding and encryption enhance data security; consistency checks and sharing mechanisms ensure accurate data flow between authorized nodes; and data access logs enable full traceability of shared data. This process effectively constructs a transparent and reliable credit evaluation system, providing crucial credit basis for task allocation and resource scheduling in coal performance management.

[0032] Furthermore, the first task allocation scheme is obtained, which specifically includes: When the credit score of the third dataset is lower than the preset credit threshold, the data filtering module extracts the subset of data below the threshold to obtain the first subset of data. Based on the credit score distribution in the first subset of data, the statistical analysis method is used to calculate the initial weight allocation of each task to obtain the first weight set. The first weight set is iteratively optimized by the reinforcement learning algorithm, and the task allocation weights are adjusted to obtain the second weight set. When the weight values ​​in the second weight set meet the preset convergence condition, the task allocation scheme is generated according to the second weight set to obtain the first task allocation scheme. The task allocation module maps the first task allocation scheme to specific task nodes to obtain the first task execution sequence. Then, the scheduling algorithm is used to optimize the task execution order to obtain the second task execution sequence. By executing the second task execution sequence, the final task allocation result is generated, and the first task allocation record is obtained.

[0033] The preset credit threshold can be determined by combining historical performance data, industry average performance level, and business risk tolerance. The statistical analysis method uses descriptive statistics such as mean and standard deviation, and divides the intervals based on the credit score distribution to determine the initial weight of tasks in each interval. The reinforcement learning algorithm uses task completion rate and the degree of performance risk reduction as reward functions, and determines parameters such as learning rate and discount factor through experiments and parameter tuning. The preset convergence condition is set as the weight value changing less than a certain minimum value for multiple consecutive iterations. The scheduling algorithm uses genetic algorithm, ant colony algorithm, etc., and optimizes the fitness function according to factors such as task priority and resource consumption, thereby clarifying the key details of the task allocation process.

[0034] Specifically, by determining initial weights based on credit scores and filtering data, the task allocation weights are iteratively optimized using reinforcement learning algorithms, then mapped to specific nodes and the execution order is optimized to achieve dynamic and intelligent allocation of coal fulfillment tasks, thereby improving supply chain collaboration efficiency and fulfillment capabilities.

[0035] S4. Based on the first task allocation scheme, the linear programming algorithm is used to optimize resource scheduling and generate a fourth dataset containing scheduling priorities. When the scheduling priorities in the fourth dataset change, the adjustment instructions are pushed to the relevant nodes through message queue technology to obtain the first execution result. Furthermore, a fourth dataset containing scheduling priorities is generated, specifically including: The linear programming algorithm is used to obtain the task execution order and allocation constraints from the task allocation scheme, obtain the optimization objective function, and then calculate the resource utilization and scheduling execution efficiency to determine the resource scheduling optimization scheme. When the resource scheduling optimization scheme meets the allocation constraints, a priority ranking is generated according to the task execution order to obtain the initial scheduling priority. The priority ranking is then iteratively adjusted to determine whether it meets the scheduling execution efficiency requirements, and the adjusted scheduling priority is obtained. By adjusting the scheduling priority, relevant data is extracted from the task allocation scheme to generate a fourth dataset. Then, a data verification algorithm is used to determine whether the generated dataset meets the allocation constraints, and the final dataset is obtained. Using the final dataset, a priority sorting algorithm is employed to verify the scheduling priorities and determine the final scheduling priorities.

[0036] Specifically, by using linear programming algorithms to integrate the execution order and constraints in the task allocation scheme, resource utilization and scheduling efficiency are calculated to determine the optimization scheme. After priority iterative adjustment, data verification and priority validation, a fourth dataset containing scheduling priorities that is accurate and meets the constraints is generated, providing a reliable basis for efficient scheduling of coal supply resources.

[0037] Furthermore, the first execution result is obtained, specifically including: When the scheduling priority of the fourth dataset changes, the dataset content is parsed through a preset message queue protocol to determine the priority adjustment instruction. The priority adjustment instruction is then pushed to the relevant nodes through the message queue, and the nodes receive confirmation. Once the node receives the confirmation, it updates the node task scheduling configuration according to the adjustment instruction, obtains the node configuration update result, extracts the execution feedback data from the node configuration update result, and uses the consistent hashing algorithm to allocate the feedback data to the storage cluster to determine the data storage location. The system retrieves feedback data based on the storage location, determines whether the feedback data meets the preset priority adjustment threshold, obtains the verification result, and pushes a successful execution command to the control center through the message queue to obtain the first execution result. Update the priority record of the fourth dataset based on the first execution result to determine the final scheduling state.

[0038] Specifically, when the scheduling priority of the fourth dataset changes, the adjustment instruction is parsed and pushed through the message queue protocol. After the node confirms receipt, the scheduling configuration is updated. The consistent hashing algorithm is used to store the execution feedback data. After passing the threshold verification, the control center is notified of the successful execution, and the dataset priority record is updated synchronously. This achieves a closed loop of accurate transmission, reliable execution and dynamic update of the scheduling instructions for coal fulfillment resources.

[0039] S5. Based on the first execution result, use time series analysis technology to predict future inventory and production trends, and generate the fifth dataset for iterative optimization.

[0040] Furthermore, a fifth dataset is generated, specifically including: Historical inventory and production data are obtained from the first execution result. Data cleaning techniques are used to remove outliers and missing values ​​to obtain a standardized dataset. Then, time series analysis techniques are used to construct an ARIMA model. In the ARIMA model, p represents the autoregression order, d represents the difference order, and q represents the moving average order. This model is used to fit historical data to obtain a trend prediction model. By using trend prediction models, future inventory change trends are predicted, generating an inventory trend dataset; similarly, future production demand change trends are predicted, generating a production trend dataset. By employing a merging technique, the inventory trend dataset and the production trend dataset are integrated to generate a fifth dataset. When the prediction error of the fifth dataset exceeds a preset threshold, an iterative optimization mechanism is used to adjust the ARIMA model parameters and regenerate the fifth dataset, resulting in an optimized fifth dataset. The optimized fifth dataset is used to update inventory management and production planning data, generating an adjusted management dataset.

[0041] Specifically, by extracting and cleaning historical data from the execution results, an ARIMA model is constructed to accurately fit the data trends, predict future changes in inventory and production demand, and integrate them to form a fifth dataset. The model parameters are then iteratively optimized based on the prediction error to ensure the accuracy of data predictions. Finally, the optimized dataset is used to update inventory management and production plans, achieving forward-looking and dynamic control of thermal coal production and inventory, effectively improving the scientific nature of thermal coal performance management and the efficiency of resource allocation.

[0042] Example 2

[0043] Please see Figure 4 This embodiment provides a coal mine management department's coal performance contract control system for implementing three-level application, and a method for implementing coal performance contract control for three-level application by coal mine management departments, including: The data acquisition and preprocessing module collects coal production and inventory data in real time through sensors and IoT devices, generating a first dataset containing timestamps, production volume, and inventory volume. Then, for data with timestamp intervals exceeding a preset threshold, edge computing nodes are used to compress and preprocess the data. Through operations such as difference calculation, sliding window data compression, linear interpolation timestamp adjustment, and mean filtering, the data is ensured to meet millisecond-level synchronization requirements, forming the first processed dataset. The performance evaluation calculation module, based on the first processing dataset, uses distributed database technology to store production and inventory information, generates a second dataset containing real-time status, analyzes abnormal data such as inventory levels below a preset threshold or production fluctuations exceeding the standard through smart contract technology, predicts the probability of performance risk by combining logistic regression algorithm, calculates the performance evaluation score, obtains the first performance evaluation result, and records the score to the blockchain for subsequent verification and query. The credit assessment and task allocation module uses blockchain technology to verify and share the first performance evaluation results from multiple parties, generating a third dataset containing credit scores. When the credit score is lower than a preset threshold, the module uses reinforcement learning algorithms to iteratively optimize the task allocation weights, generates a first task allocation scheme based on the optimized weights, maps the scheme to specific task nodes, optimizes the task execution order, and determines the final task allocation result. The resource scheduling optimization module, based on the first task allocation scheme, uses a linear programming algorithm to comprehensively consider task execution order, allocation constraints, resource utilization, and scheduling efficiency to optimize resource scheduling and generate a fourth dataset containing scheduling priorities. When scheduling priorities change, adjustment instructions are pushed to relevant nodes through message queue technology, and scheduling configurations are updated based on node feedback to ensure efficient execution of resource scheduling. The trend prediction and cyclic optimization module collects and cleans historical inventory and production data based on the first execution result of resource scheduling, constructs an ARIMA model to predict future inventory and production trends, and generates a fifth dataset. When the prediction error exceeds the threshold, the model parameters are adjusted and the prediction is re-made. The optimized dataset is used to update inventory management and production plans, thereby achieving cyclic optimization of coal delivery control.

[0044] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for coal mine management departments to implement three-level application of coal performance contract control, characterized by: Includes the following steps: Real-time data collection of coal production and inventory data is performed by sensors and IoT devices to generate a first dataset containing timestamps, production volume and inventory volume. Then, edge computing nodes compress and preprocess data with timestamp intervals exceeding a preset millisecond threshold to obtain a first processed dataset that meets millisecond-level synchronization requirements. Based on the first processing dataset, distributed database technology is used to store production and inventory information, and a second dataset containing real-time status is generated. Smart contract technology is used to calculate the performance evaluation score in real time for abnormal data such as inventory levels below a preset threshold or production fluctuations exceeding a preset range, and the first performance evaluation result is obtained. The first performance evaluation result is verified and shared by multiple parties using blockchain technology, generating a third dataset containing credit scores. When the credit score in the third dataset is lower than the preset credit threshold, a reinforcement learning algorithm is used to adjust the task allocation weights to obtain the first task allocation scheme. Based on the first task allocation scheme, a linear programming algorithm is used to optimize resource scheduling and generate a fourth dataset containing scheduling priorities. When the scheduling priorities in the fourth dataset change, adjustment instructions are pushed to relevant nodes through message queue technology to obtain the first execution result. Based on the first execution result, time series analysis technology is used to predict future inventory and production trends, generating a fifth dataset for iterative optimization.

2. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: The first processing dataset that meets the millisecond-level synchronization requirements is obtained, specifically including: When the timestamp interval of the first dataset exceeds the preset threshold, the timestamp sequence is obtained through the edge computing node, the difference is calculated, the timestamp pairs exceeding the threshold are determined, the corresponding data segments are extracted using the sliding window method, and data compression is performed to obtain the compressed dataset. If the timestamp interval of the compressed dataset still exceeds the preset threshold, the timestamp sequence is adjusted by a linear interpolation algorithm to generate a calibration dataset. Then, the mean filtering method is used to preprocess the data features to obtain a smooth dataset. When the synchronization error of the smoothed dataset is lower than a preset threshold, the data is sharded and stored through edge computing nodes to generate the first processing dataset. Based on the timestamp sequence of the first processing dataset, a verification algorithm is used to detect the synchronization consistency and determine the final synchronized dataset.

3. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: Generate a second dataset containing real-time status, specifically including: Production and inventory data are obtained from the first dataset and stored in a structured manner through a distributed database to obtain an initial data table. When the recording time of production and inventory data in the initial data table is later than a preset threshold, a real-time status update is triggered to generate an intermediate dataset containing status identifiers. Data processing algorithms are used to deduplicate and standardize the format of the intermediate dataset to obtain a standard dataset. Then, based on the state identifiers in the standard dataset, a time series analysis algorithm is used to extract real-time state features and generate a state feature set. When the feature values ​​in the state feature set exceed the preset range, a correlation analysis is performed on the production data and inventory data to determine data consistency and obtain a consistency result set. By merging the consistency result set with the standard dataset, a second dataset containing real-time status is generated.

4. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: The first performance evaluation result was obtained, specifically including: Inventory and production data are obtained from the second dataset. Data cleaning techniques are used to remove outliers to obtain a cleaned dataset. When the inventory in the cleaned dataset is lower than a preset threshold, the inventory is marked as abnormal by threshold judgment techniques to obtain an inventory abnormality label. When the production volume in the cleaned dataset fluctuates beyond the preset range, the fluctuation amplitude is calculated using fluctuation detection technology to obtain a fluctuation anomaly marker. Based on inventory anomaly markers and fluctuation anomaly markers, a logistic regression algorithm is used to predict the probability of performance risk and obtain a performance risk score. Then, through smart contract technology, performance evaluation parameters are obtained from the performance risk score, the score calculation logic is executed, and the first performance evaluation result is obtained. The performance evaluation score is extracted from the first performance evaluation result, recorded on the blockchain using data storage technology, and a blockchain storage identifier is obtained. Then, the query index of the performance evaluation score is obtained, a performance evaluation query key is generated, and the query key result is obtained.

5. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: Generate a third dataset containing credit scores, specifically including: Initial performance data is obtained from the performance evaluation system, a data fingerprint is generated using a hash algorithm, and stored in the blockchain network to obtain the first data hash value. The first data hash value is then verified by multiple parties through verification nodes in the blockchain network. When the consistency of the verification nodes reaches a preset threshold, a verification pass identifier is generated. Based on the verification pass identifier, key fields are extracted from the initial performance data, and a weighted average algorithm is used to calculate the credit score to obtain the first credit score set. Then, through the smart contract of the blockchain network, the first credit score set is bound with the verification pass identifier to generate the encrypted second dataset. Credit scores and associated performance data are obtained from the second dataset. A data consistency check algorithm is used to determine the data integrity and obtain the consistency verification result. Then, through the sharing mechanism of the blockchain network, the second dataset is distributed to authorized nodes to generate a third dataset containing credit scores. Credit scores and performance data are extracted from the third dataset, and data access logs are recorded through the automatic update mechanism of smart contracts to obtain shared data tracking records.

6. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: The first task allocation plan is obtained, which includes: When the credit score of the third dataset is lower than the preset credit threshold, the data filtering module extracts the subset of data below the threshold to obtain the first subset of data. Based on the credit score distribution in the first subset of data, the statistical analysis method is used to calculate the initial weight allocation of each task to obtain the first weight set. The first weight set is iteratively optimized by the reinforcement learning algorithm, and the task allocation weights are adjusted to obtain the second weight set. When the weight values ​​in the second weight set meet the preset convergence condition, the task allocation scheme is generated according to the second weight set to obtain the first task allocation scheme. The task allocation module maps the first task allocation scheme to specific task nodes to obtain the first task execution sequence. Then, the scheduling algorithm is used to optimize the task execution order to obtain the second task execution sequence. By executing the second task execution sequence, the final task allocation result is generated, and the first task allocation record is obtained.

7. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: Generate a fourth dataset containing scheduling priorities, specifically including: The linear programming algorithm is used to obtain the task execution order and allocation constraints from the task allocation scheme, obtain the optimization objective function, and then calculate the resource utilization and scheduling execution efficiency to determine the resource scheduling optimization scheme. When the resource scheduling optimization scheme meets the allocation constraints, a priority ranking is generated according to the task execution order to obtain the initial scheduling priority. The priority ranking is then iteratively adjusted to determine whether it meets the scheduling execution efficiency requirements, and the adjusted scheduling priority is obtained. By adjusting the scheduling priority, relevant data is extracted from the task allocation scheme to generate a fourth dataset. Then, a data verification algorithm is used to determine whether the generated dataset meets the allocation constraints, and the final dataset is obtained. Using the final dataset, a priority sorting algorithm is employed to verify the scheduling priorities and determine the final scheduling priorities.

8. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: The first execution result is obtained, specifically including: When the scheduling priority of the fourth dataset changes, the dataset content is parsed through a preset message queue protocol to determine the priority adjustment instruction. The priority adjustment instruction is then pushed to the relevant nodes through the message queue, and the nodes receive confirmation. Once the node receives the confirmation, it updates the node task scheduling configuration according to the adjustment instruction, obtains the node configuration update result, extracts the execution feedback data from the node configuration update result, and uses the consistent hashing algorithm to allocate the feedback data to the storage cluster to determine the data storage location. The system retrieves feedback data based on the storage location, determines whether the feedback data meets the preset priority adjustment threshold, obtains the verification result, and pushes a successful execution command to the control center through the message queue to obtain the first execution result. Update the priority record of the fourth dataset based on the first execution result to determine the final scheduling state.

9. The method for coal mine management departments to implement three-level application of coal performance management according to claim 1, characterized in that: The fifth dataset was generated, specifically including: Historical inventory and production data are obtained from the first execution result. Data cleaning techniques are used to remove outliers and missing values ​​to obtain a standardized dataset. Then, time series analysis techniques are used to build an ARIMA model to fit the historical data and obtain a trend prediction model. By using trend prediction models, future inventory change trends are predicted, generating an inventory trend dataset; similarly, future production demand change trends are predicted, generating a production trend dataset. By employing a merging technique, the inventory trend dataset and the production trend dataset are integrated to generate a fifth dataset. When the prediction error of the fifth dataset exceeds a preset threshold, an iterative optimization mechanism is used to adjust the ARIMA model parameters and regenerate the fifth dataset, resulting in an optimized fifth dataset. The optimized fifth dataset is used to update inventory management and production planning data, generating an adjusted management dataset.

10. A coal mine management department's three-level application coal performance management system, used to implement the coal mine management department's three-level application coal performance management method as described in any one of claims 1-9, characterized in that: include: The data acquisition and preprocessing module collects coal production and inventory data in real time through sensors and IoT devices, generating a first dataset containing timestamps, production volume and inventory volume. Then, for data whose timestamp interval exceeds a preset threshold, edge computing nodes are used to compress and preprocess the data to form the first processed dataset. The performance evaluation calculation module, based on the first processing dataset, uses distributed database technology to store production and inventory information, generates a second dataset containing real-time status, analyzes abnormal data such as inventory levels below a preset threshold or production fluctuations exceeding the standard through smart contract technology, predicts the probability of performance risk by combining logistic regression algorithm, calculates the performance evaluation score, and obtains the first performance evaluation result. The credit assessment and task allocation module uses blockchain technology to verify and share the first performance evaluation results from multiple parties, generating a third dataset containing credit scores. When the credit score is lower than a preset threshold, the module uses reinforcement learning algorithms to iteratively optimize the task allocation weights, generates a first task allocation scheme based on the optimized weights, maps the scheme to specific task nodes, optimizes the task execution order, and determines the final task allocation result. The resource scheduling optimization module optimizes resource scheduling using a linear programming algorithm based on the first task allocation scheme, generates a fourth dataset containing scheduling priorities, and pushes adjustment instructions to relevant nodes through message queue technology when scheduling priorities change, and updates the scheduling configuration based on node feedback. The trend prediction and cycle optimization module collects and cleans historical inventory and production data based on the first execution result of resource scheduling, builds an ARIMA model to predict future inventory and production trends, and generates a fifth dataset. When the prediction error exceeds the threshold, the model parameters are adjusted and the prediction is re-made. The optimized dataset is then used to update inventory management and production plans.

Images