Coal-fired power plant material storage digital management platform

Abstract

The invention discloses a digital management platform for material storage of a coal-fired power plant, and belongs to the technical field of data management. The platform comprises a material identification code endowing module, a material data acquisition module, a material storage layout module and a purchase scheme generation module. Wherein the purchase scheme generation module is used for constructing a purchase prediction model based on ITransform and outputting a purchase suggestion list; the material storage and layout module is used for constructing a fitness function by taking a space utilization rate and a taking convenience value as income items, taking weight overrun, volume overrun, goods allocation conflicts, channel blockage and hazardous chemical isolation as hard constraint penalty items and taking a material shelf life as a soft constraint penalty item, continuously iterating a population through selection, intersection and mutation operations, and storing a material storage and layout result; and finding an optimal storage layout scheme. According to the invention, the technical effects of realizing real-time information of coal-fired power plant material warehouse management, improving the warehouse space utilization rate and reducing the purchase period can be realized.

Description

Technical Field

[0001] This application relates to the field of data management technology, and more specifically, to a digital management platform for material storage in coal-fired power plants. Background Technology

[0002] Material storage management for coal-fired power plants involves data management activities throughout the entire lifecycle of materials (such as coal, lubricating oil, spare parts, etc.) required for daily operation of coal-fired power plants, including procurement, acceptance, warehousing, storage, issuance and disposal. It is one of the most critical logistical links to ensure the continuous, safe and economical operation of the units, and aims to ensure stable power generation, reduce operating costs and improve management efficiency.

[0003] In existing technologies, information flow between various stages of material warehousing management in coal-fired power plants (procurement, transportation, quality inspection, warehousing, inventory, and disposal, etc.) is inefficient. Data silos between these stages create information islands, leading to delays in the transmission of information such as arrival and demand. This results in problems such as inventory backlog, delayed warehousing, inefficient inventory management, and insufficient decision-making optimization. In addition, warehouse layout and procurement plans rely heavily on subjective decisions, resulting in low decision accuracy, low utilization of warehouse space, and extended procurement cycles. Summary of the Invention

[0004] The main purpose of this application is to provide a digital management platform for the material storage of coal-fired power plants, in order to solve the technical problems of untimely information transmission, low utilization of storage space, and extended procurement cycle in the existing material storage management of coal-fired power plants, and to achieve the technical effects of real-time information, improved storage space utilization, and reduced procurement cycle.

[0005] To achieve the above objectives, this application proposes a digital management platform for material storage in coal-fired power plants, comprising: The material identification code assignment module is used to assign a unique identification code to materials so that the unique identification code can be identified by scanning devices at each stage of the platform, and the material data generated at each stage of the platform can be associated with the unique identification code. The material data acquisition module is used to automatically collect material data generated at each stage of the platform. The procurement plan generation module is used to acquire historical data, generate training and testing sets based on the historical data, construct a procurement prediction model based on ITransformer, train the procurement prediction model based on the training set, and validate the procurement prediction model based on the testing set to obtain a trained procurement prediction model. Current data is input into the trained procurement prediction model, and a procurement suggestion list is output. The historical data includes historical material procurement data, historical material inventory data, historical material consumption data, historical market price data, historical supplier data, and historical plan data. The current data includes current material inventory data, current market price data, current material consumption data, current supplier data, and current plan data. The procurement suggestion list includes a unique identifier for each material, material name, predicted demand quantity, predicted demand range, suggested procurement quantity, suggested procurement time, and a list of recommended suppliers. ITransformer is an inverted self-attention Transformer structure. The material storage layout module is used to randomly generate an initial population based on available storage locations. It constructs a fitness function using space utilization and ease of access as benefits, weight exceeding limits, volume exceeding limits, storage location conflicts, passageway blockage, and hazardous materials isolation as hard constraints, and material shelf life as a soft constraint. Through selection, crossover, and mutation operations, the population is iterated continuously, retaining individuals with high fitness. After multiple generations of evolution, the optimal storage layout scheme is found. The initial population consists of multiple chromosomes, each representing a storage layout scheme. Each chromosome is composed of multiple gene segments, and each gene segment consists of a unique identifier, storage location, and material quantity.

[0006] In one possible implementation, the material storage layout module includes: The initial population generation submodule is used to obtain an idle storage space pool carrying storage space attribute information; for any of the aforementioned materials, based on the material attribute information of the material and the storage space attribute information of the idle storage spaces in the idle storage space pool, an idle storage space is selected from the idle storage space pool to generate a candidate storage space pool for the material; all types of materials are randomly sorted, and the real-time available pool for each type of material is obtained in a random order, and an idle storage space is selected from the real-time available pool as an occupied storage space; a chromosome is generated based on the occupied storage spaces of all types of materials, wherein the real-time available pool is the intersection of the candidate storage space pool of the current sequence of materials and the current storage space pool, and the current storage space pool is the storage space pool of the idle storage space pool after deleting the occupied storage spaces of all preceding materials; the idle storage space pool is copied, and multiple chromosomes are regenerated to form the initial population.

[0007] In one possible implementation, the material storage layout module includes: The revenue calculation submodule is used to: for any chromosome, calculate the ratio of the sum of the volumes of all materials on the chromosome to the sum of the volumes of all occupied storage locations on the chromosome, as the space utilization rate; for any gene segment, calculate the Manhattan distance from the storage location corresponding to the gene segment to all entrances and exits, and take the minimum value as the first distance; sum the first distance and the second distance to obtain the single picking cost of the gene segment, where the second distance is the product of the climbing equivalence coefficient and the climbing level; obtain the annual outbound frequency and the quantity stored in the storage location of the materials corresponding to the gene segment, and multiply the single picking cost, the annual outbound frequency, and the quantity stored to obtain the annual total picking distance of the gene segment; sum the annual total picking distance of all gene segments on the same chromosome, and then normalize it to obtain the retrieval convenience value; perform a weighted summation of the space utilization rate and the retrieval convenience value to obtain the revenue value; and / or, The penalty value calculation submodule is used to determine, for any of the aforementioned chromosomes, whether there is at least one of the following: location conflict, volume exceeding limit, weight exceeding limit, passage blockage, or hazardous chemical isolation; if so, the chromosome is eliminated; otherwise, the hard constraint penalty value is set to 0. For any of the aforementioned chromosomes, the remaining shelf life ratio of each material on the chromosome is calculated. If the remaining shelf life ratio is not less than a preset ratio, the corresponding sub-soft constraint penalty value is set to 0; if the remaining shelf life ratio is less than the preset ratio, the corresponding sub-soft constraint penalty value is calculated according to the first formula, which is: In the formula, Let the sub-soft constraint penalty value be the j-th material. Let j be the remaining shelf life of the j-th material, where j = 1, 2, ..., M, and M is the total number of material types on the chromosome. Both j and M are positive integers. The preset ratio is used; the sub-soft constraint penalty values ​​of all materials are summed and multiplied by the soft constraint coefficient to obtain the soft constraint penalty value; the hard constraint penalty value and the soft constraint penalty value are summed to obtain the penalty value.

[0008] In one possible implementation, the material storage layout module further includes: The fitness function construction submodule is used to construct the fitness function based on the difference between the reward value and the penalty value.

[0009] In one possible implementation, the objective function of the procurement forecasting model in the procurement scheme generation module is to minimize the following composite loss:

[0010]

[0011]

[0012]

[0013] In the formula, θ represents all training parameters; This is the mean squared error loss for point prediction, i.e., the point prediction loss; , This is the quantile loss for interval prediction, i.e., the interval prediction loss; α and β are hyperparameters used to balance the importance of the point prediction loss and the interval prediction loss in the overall objective; i = 1, 2, ..., N, where N is the number of samples in the training set, and i and N are both positive integers; This represents the historical material consumption of the i-th sample. This is the predicted point value output for the i-th sample, i.e., the predicted demand.

[0014] In one possible implementation, the digital management platform for coal-fired power plant material storage also includes: The inbound / outbound management module is used to deploy and manage automated inbound / outbound equipment.

[0015] In one possible implementation, the digital management platform for coal-fired power plant material storage also includes: The report generation module is used to generate material reports based on preset templates. The material reports include inventory balance reports, material turnover rate reports, procurement cost reports, and tool usage frequency reports.

[0016] In one possible implementation, the digital management platform for coal-fired power plant material storage also includes: The material quality assurance traceability module is used to establish a material quality assurance information database that stores quality assurance information of various materials and strongly associates it with unique identification codes. When it is necessary to trace the quality of materials, the unique identification code is scanned and the quality assurance information corresponding to the unique identification code is obtained by querying the material quality assurance information database. The quality assurance information includes quality assurance documents, production batches, raw material sources, inspection reports, and quality certification certificates.

[0017] In one possible implementation, the digital management platform for coal-fired power plant material storage also includes: The warehouse environment control module includes an environmental sensor and an environmental control device. When the environmental parameters collected by the environmental sensor exceed the preset range, it determines the abnormal environmental area, activates the corresponding environmental control device to adjust the environmental parameters of the abnormal environmental area, generates a list of materials affected by the environmental anomaly, and issues an early warning message containing the list of materials. The environmental sensors include temperature and humidity sensors, smoke detectors, and vibration sensors; the environmental control equipment includes ventilation equipment, dehumidification equipment, and cooling equipment.

[0018] In one possible implementation, the digital management platform for coal-fired power plant material storage also includes: The positioning alarm module is used to monitor the location of materials in real time using positioning technology, and to issue an alarm signal when the location of materials changes abnormally.

[0019] The technical solutions provided by the embodiments of this application may include the following beneficial effects: By assigning unique identification codes to materials, the entire process of procurement, warehousing, storage, outbound delivery, and consumption can be traced by scanning the codes, enabling traceability of material origin, batch, shelf life, and storage location, achieving one code per item and full visibility throughout the process. Based on the ITransformer procurement forecasting model, which integrates multi-dimensional data such as historical procurement, inventory, consumption, price, suppliers, and plans, it predicts future demand trends and automatically outputs a procurement suggestion list of "predicted demand + range + recommended suppliers," thereby optimizing procurement strategies, enabling on-demand procurement, precise replenishment, shortening procurement cycles, and improving response speed. With space utilization and ease of access as benefits, it automatically optimizes the allocation of storage locations, improves warehouse volume utilization and operational efficiency, and ensures that heavy goods are placed low, hazardous materials are isolated, and passageways are unobstructed through hard constraints (weight, volume, aisle, and hazardous materials isolation). It automatically detects illegal layouts, reducing the risk of safety accidents. Through iterative optimization using genetic algorithms, it supports dynamic adjustments (seasonal, maintenance period), ensuring that the inventory structure is always in an optimal or near-optimal state, reducing slow-moving and expired materials, optimizing material warehousing strategies, and improving inventory turnover. Attached Figure Description

[0020] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings: Figure 1 The structural block diagram of the digital management platform for material storage in coal-fired power plants provided in this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0022] Figure 1 This is a structural block diagram of a digital management platform 10 for material storage in a coal-fired power plant, provided in this embodiment. Figure 1 As shown, the digital management platform 10 for material storage in the coal-fired power plant mainly includes: a material identification code assignment module 100, a material data acquisition module 200, a material storage layout module 300, and a procurement plan generation module 400.

[0023] The Material Identification Code Assignment Module 100 is used to assign a unique identifier to each material, enabling the scanning of this unique identifier at each stage of the platform and linking the material data generated at each stage with the unique identifier. QR code or RFID technology is used to uniformly encode all materials, assigning each item (set) a globally unique identifier. This unique identifier is used throughout the entire lifecycle of the material, from purchase order to final disposal. By parsing the unique identifier, data from the entire lifecycle of the material is obtained. During the procurement process, the identifier is bound to the purchase order information; upon warehousing, the material warehousing information is quickly registered and inventory is updated using a scanning device; during borrowing and returning, the operation information and time are recorded by scanning; during maintenance, repair, and inspection, the corresponding maintenance records are associated with the identifier; during inventory checks, the actual status of the material is quickly verified against the system records by scanning; and upon disposal, the reason for disposal and the time are recorded, ensuring that the status of the material is traceable throughout its entire lifecycle. The material data acquisition module 200 is used to automatically collect material data generated in various stages of the platform. Material data includes, but is not limited to, data such as inbound and outbound records, inventory changes, maintenance history, and procurement costs.

[0024] In this embodiment, the material storage layout module 300 is used to randomly generate an initial population based on available storage locations; a fitness function is constructed using space utilization and ease of access as benefit items, weight exceeding limits, volume exceeding limits, storage location conflicts, channel blockage, and hazardous chemical isolation as hard constraint penalty items, and material shelf life as a soft constraint penalty item; the population is iterated continuously through selection, crossover, and mutation operations, and individuals with high fitness are selected for retention. After multiple generations of evolution, the optimal storage layout scheme is found; wherein, the initial population consists of multiple chromosomes, each chromosome represents a storage layout scheme, each chromosome is spliced ​​together from multiple gene segments, and each gene segment consists of a unique identifier, storage location, and material quantity.

[0025] In some alternative embodiments, such as Figure 1 As shown, the material storage layout module 300 includes: The initial population generation submodule 301 is used to obtain an idle storage space pool carrying storage space attribute information. For any material, based on the material attribute information and the storage space attribute information of the idle storage spaces in the idle storage space pool, an idle storage space is selected from the idle storage space pool to generate a candidate storage space pool for the material. All types of materials are randomly sorted, and the real-time available pool for each type of material is obtained in a random order. An idle storage space is selected from the real-time available pool as an occupied storage space. A chromosome is generated based on the occupied storage spaces of all types of materials. The real-time available pool is the intersection of the candidate storage space pool of the current sequence of materials and the current storage space pool. The current storage space pool is the storage space pool of the idle storage space pool after deleting the occupied storage spaces of all previous materials. The idle storage space pool is copied, and multiple chromosomes are regenerated to form the initial population.

[0026] In this embodiment, the location attribute information includes coordinates, capacity, load-bearing capacity, and hazardous chemical classification; the material attribute information includes volume, weight, and hazardous chemical classification. The quantity of each type of material to be stored is pre-set. This quantity can be the same as the actual quantity to be stored, or it can be randomly sampled within a range of ±10% or ±20% of the actual quantity. For example, if the bolt inventory is 5000, the chromosome might be set to 4500, 5200, 4800, etc. When generating the candidate storage location pool, the required number of storage locations for each type of material is calculated based on its volume and weight. The required number of storage locations for each material is calculated separately for volume and weight, and the maximum value is taken as the final required number of storage locations. For example, a motor has a volume of 1.4 cubic meters, and the volume of a single storage location is 0.6 cubic meters. 1.4 / 0.6 < 3, so based on volume, one motor requires 3 storage locations. However, a motor weighs 210 kg, and the maximum load capacity of a single storage location is 140 kg. 210 / 140 < 2, so based on weight, one motor requires 2 storage locations. Therefore, one motor requires 3 storage locations.

[0027] In addition, the selection of a hazardous storage area is determined based on the hazardous chemical properties of the materials. If the materials are overweight, the low-load-bearing layers in the available storage space pool should be filtered out; if the materials are high-frequency items (high usage frequency), available storage spaces near the entrance / exit should be prioritized.

[0028] After generating the candidate storage location pool, a random allocation method can be used to assign storage locations to each type of material. This involves randomly selecting storage locations without replacement from the filtered available pool. Specifically, first, all types of materials are randomly sorted. For the material ranked first, the intersection of its candidate storage location pool and available storage location pool is calculated. This intersection is used as the real-time available pool for that material. Available storage locations are selected from this pool as the occupied storage locations for that material. At this point, the occupied storage location for that material is removed from the initial available pool, and the remaining available storage locations form the current storage location pool. Then, the storage locations for the second-ranked material are selected. The intersection of its candidate storage location pool and current storage location pool is calculated again, and this intersection is used as the real-time available pool for that material. Available storage locations are selected from this pool as the occupied storage locations for that material. The current storage location pool is then updated, i.e., the occupied storage location for that material is removed from the current storage location pool. This ensures that the occupied storage locations for the second-ranked material and the first-ranked material do not overlap, thus avoiding conflicts. This process is repeated to select the storage space for all subsequent materials.

[0029] After obtaining the occupied storage locations of all types of materials, the gene segments of each material are first constructed based on its unique identifier, storage location (warehouse area, shelf, layer, position) and storage quantity. Then, all gene segments are arranged in a preset order, and the first and last segments of these gene segments are connected to obtain a chromosome.

[0030] It should be noted that the fixed coding order of each gene segment must be the same. For example, they can be spliced ​​together in the order of unique identifier, storage location, and storage quantity.

[0031] Duplicate the initial idle storage pool and regenerate multiple chromosomes following the steps described above. The method is the same and will not be repeated here. Furthermore, the gene segments corresponding to all resources are arranged in the same order on each chromosome.

[0032] It is important to note that although the probability of generating two identical chromosomes in practice is very low, deduplication checks should still be implemented. For example, each newly generated chromosome should be compared with the previously generated chromosomes. If they are duplicates, they should be discarded, or slightly perturbed (e.g., randomly selecting two gene segments from different materials on the same chromosome, swapping their storage locations, and keeping everything else unchanged) before reuse, so that multiple chromosomes in the initial population are different from each other.

[0033] By randomly generating the initial population in the above manner, the genetic algorithm can have a clean and diverse starting point, avoiding falling into the quagmire of illegal solutions from the very beginning.

[0034] In some alternative embodiments, such as Figure 1 As shown, the material storage layout module 300 includes: The revenue calculation submodule 302 is used to calculate the space utilization rate as the ratio of the sum of the volumes of all materials on the chromosome to the sum of the volumes of all occupied storage locations on the chromosome for any chromosome; for any gene segment, it calculates the Manhattan distance from the storage location corresponding to the gene segment to all entrances and exits, and takes the minimum value as the first distance; it sums the first distance and the second distance to obtain the single picking cost of the gene segment, where the second distance is the product of the climbing equivalence coefficient and the climbing level; it obtains the annual outbound frequency and the quantity stored in the storage location of the materials corresponding to the gene segment, and calculates the annual total picking distance of the gene segment by multiplying the single picking cost, the annual outbound frequency, and the quantity stored; it sums the annual total picking distance of all gene segments on the same chromosome, and then performs normalization to obtain the retrieval convenience value; it performs a weighted summation of the space utilization rate and the retrieval convenience value to obtain the revenue value. The penalty value calculation submodule 303 is used to determine, for any chromosome, whether there is at least one of the following: location conflict, volume exceeding limit, weight exceeding limit, passage blockage, or hazardous chemical isolation; if so, the chromosome is eliminated; otherwise, the hard constraint penalty value is set to 0. For any chromosome, the remaining shelf life ratio of each material on the chromosome is calculated. If the remaining shelf life ratio is not less than a preset ratio, the corresponding sub-soft constraint penalty value is set to 0; if the remaining shelf life ratio is less than the preset ratio, the corresponding sub-soft constraint penalty value is calculated according to the first formula, which is: In the formula, Let the sub-soft constraint penalty value be the j-th material. Let j represent the remaining shelf life of the j-th item, where j = 1, 2, ..., M, and M is the total number of item types on the chromosome. Both j and M are positive integers. The soft constraint penalty values ​​of all materials are summed and multiplied by the soft constraint coefficient to obtain the soft constraint penalty value; the hard constraint penalty value and the soft constraint penalty value are summed to obtain the penalty value. The fitness function construction submodule 304 is used to construct a fitness function based on the difference between the reward value and the penalty value.

[0035] In this embodiment, the cost of picking a single item includes not only the distance from the location to the entrance / exit, but also the round-trip distance for staff to climb the shelves. The climbing equivalence coefficient here refers to the distance for each floor climbed. For example, if the location is on the third floor of the shelf, the number of floors climbed is 2, and the climbing equivalence coefficient is 2 meters / floor. Therefore, the second distance is 2×2=4 meters.

[0036] The number of annual outbound shipments can be directly retrieved from the outbound details table of the ERP / WMS (Enterprise Resource Planning / Warehouse Management System), and then grouped and counted by material code.

[0037] To calculate the remaining shelf life, first calculate the difference between the expiration date and the current date, then calculate the ratio of this difference to the total shelf life. This ratio is the remaining shelf life. The preset ratio is set manually, for example, 30%.

[0038] For the soft constraint coefficient, you can first run a batch (e.g., 1000-2000) of random legal chromosomes, calculate their payoff values, and take the 95th percentile value to obtain the upper limit of the payoff item. Multiply the upper limit of the payoff item by the maximum reduction ratio to obtain the soft constraint coefficient, which can be set to 20%, 30%, etc. Of course, the soft constraint coefficient can be fine-tuned in practice.

[0039] In this embodiment, the procurement plan generation module 400 is used to acquire historical data, generate training and test sets based on the historical data, construct a procurement prediction model based on ITransformer, train the procurement prediction model based on the training set, and verify the procurement prediction model based on the test set to obtain a trained procurement prediction model. Current data is then input into the trained procurement prediction model, and a procurement suggestion list is output. The historical data includes historical material procurement data, historical material inventory data, historical material consumption data, historical market price data, historical supplier data, and historical plan data. The current data includes current material inventory data, current market price data, current material consumption data, current supplier data, and current plan data. The procurement suggestion list includes the unique identifier of the material, the material name, the predicted demand quantity, the predicted demand range, the suggested procurement quantity, the suggested procurement time, and a list of recommended suppliers.

[0040] Optionally, the objective function of the procurement forecasting model in the procurement scheme generation module 400 is to minimize the following composite loss:

[0041]

[0042]

[0043]

[0044] In the formula, θ represents all training parameters; This is the mean squared error loss for point prediction, i.e., the point prediction loss; , This is the quantile loss for interval prediction, i.e., the interval prediction loss; α and β are hyperparameters used to balance the importance of point prediction loss and interval prediction loss in the overall objective; i = 1, 2, ..., N, where N is the number of samples in the training set, and i and N are both positive integers; This represents the historical material consumption of the i-th sample. This is the predicted point value output for the i-th sample, i.e., the predicted demand.

[0045] By integrating with market price databases, supplier management systems, and power plant production planning systems, the system obtains information on material market price fluctuations, supplier supply capacity and credit data, and power plant production planning arrangements. The procurement plan generation module 400 provides data support for procurement decisions and inventory management, such as developing procurement plans in advance based on material consumption patterns to avoid stockouts and overstocking. Based on forecast results and inventory status, the module determines the corresponding procurement time and quantity.

[0046] In some alternative embodiments, such as Figure 1 As shown, the digital management platform 10 for coal-fired power plant material storage also includes an inbound / outbound management module 500, used to deploy and manage automated inbound / outbound equipment, such as automated guided vehicles (AGVs), automated storage and retrieval systems (AS / RS), and intelligent sorting systems. Combined with barcode scanning technology, when materials arrive and are put into storage, the AGVs automatically transport the materials to the designated storage location, and the system automatically updates the inbound information. When materials are taken out of storage, according to the outbound instructions, the AGVs quickly and accurately move the materials to the shipping area. The entire process requires no manual intervention, significantly improving the efficiency and accuracy of inbound / outbound operations. In some alternative embodiments, such as Figure 1 As shown, the digital management platform 10 for material storage in coal-fired power plants also includes a positioning alarm module 600, which uses positioning technology to monitor the location of materials in real time and issues an alarm signal when the location of materials changes abnormally to prevent material loss.

[0047] Specifically, the location of various materials in the warehouse can be monitored in real time using UWB or RFID technology; preferably, UWB tags can be affixed to core materials and RFID tags to general materials to achieve hybrid positioning, thus balancing cost and positioning effectiveness.

[0048] Define an electronic fence for each item, which is its authorized storage area. As soon as the location signal of the item leaves the corresponding electronic fence and there is no associated legitimate process (such as no corresponding outbound slip or borrowing slip), the platform will immediately trigger an alarm and notify the relevant personnel.

[0049] In some alternative embodiments, such as Figure 1As shown, the digital management platform 10 for material storage in coal-fired power plants also includes a warehouse environment control module 700 that uses a sensor network. This module includes environmental sensors and environmental control equipment. When the environmental parameters collected by the environmental sensors exceed the preset range, it identifies an abnormal environmental area, activates the corresponding environmental control equipment to adjust the environmental parameters in the abnormal area, generates a list of materials affected by the environmental anomaly, and issues an early warning message containing the list of materials. The environmental sensors include temperature and humidity sensors, smoke detectors, and vibration sensors; the environmental control equipment includes ventilation equipment, dehumidification equipment, and cooling equipment.

[0050] Specifically, for physical quantities that are continuously distributed in space (such as temperature and humidity), the distribution of the entire space can be estimated using measurements from a finite number of discrete points (sensors) and algorithms (such as inverse distance weighting or Kriging interpolation). Taking temperature and humidity sensors as an example, when a few sensors alarm, the system combines the readings of normal sensors in the surrounding area to estimate the temperature and humidity field of the entire warehouse, thereby more accurately delineating the range and boundaries of the abnormal area.

[0051] For environmental regulation, the platform will prioritize activating environmental regulation equipment in areas with abnormal environmental conditions or closest to such areas to achieve energy conservation and precise control. In complex situations, the platform will activate equipment in tandem; for example, in hot and humid weather, the platform may simultaneously activate cooling and dehumidifying equipment.

[0052] In some alternative embodiments, such as Figure 1 As shown, the digital management platform 10 for coal-fired power plant material storage also includes a material quality assurance traceability module 800, which is used to establish a material quality assurance information database that stores quality assurance information of various materials and strongly associates it with unique identification codes. When it is necessary to trace the quality of materials, the unique identification code can be scanned, and the quality assurance information corresponding to the unique identification code can be obtained by querying the material quality assurance information database. This allows for the quick retrieval of the complete quality chain information of the material, providing a strong basis for quality problem investigation and responsibility determination. The quality assurance information includes, but is not limited to, quality assurance documents, production batches, raw material sources, inspection reports, and quality certification certificates.

[0053] In some alternative embodiments, such as Figure 1 As shown, the digital management platform 10 for material storage in coal-fired power plants also includes a user permission management module 900, which is used to build a role-based access control (RBAC) model. By introducing the intermediate layer of "roles", users and permissions are decoupled. It abstracts and defines roles according to the actual job system of the power plant, assigns precise operation permissions to roles, and finally assigns users to the corresponding roles, so that users can obtain all the permissions necessary for the job.

[0054] For example, purchasing staff can create and modify purchase orders and view supplier information; warehouse managers are responsible for material inbound and outbound operations, inventory counting, and warehouse management; production staff can submit material borrowing applications and view borrowing records; finance staff can view material costs and reimbursement records; and management has the highest authority, able to view all material information and analysis reports, and make overall decisions. Permission settings are flexible and can be adjusted at any time according to actual business needs.

[0055] In this embodiment, the digital management platform 10 for coal-fired power plant material storage adopts a distributed server architecture, distributing functional modules such as data storage, business logic processing, and user interface display across different servers to improve the system's concurrent processing capabilities, stability, and scalability. It is equipped with a comprehensive data backup and recovery system, regularly performing full and incremental backups of material management data, and storing the backup data in an off-site disaster recovery center. When the system fails or data is lost, it can quickly recover from the backup data, ensuring business continuity and data security.

[0056] In addition, applications for mobile devices such as smartphones and tablets have been developed, synchronizing data with the platform's core system in real time. Employees can use the mobile application to check material inventory information, submit tool borrowing requests, view approval progress, and receive system notifications anytime, anywhere. Warehouse managers can perform inventory checks and record inbound and outbound operations on their mobile devices. Management can view key material indicator reports and conduct emergency approvals through the mobile application, improving work flexibility and responsiveness. The system boasts strong stability and scalability. Its distributed architecture ensures stable operation under high concurrency scenarios, reducing failure rates, while also allowing for easy expansion of functional modules and server resources to meet the company's development needs and adapt to business growth.

[0057] This embodiment constructs a warehousing, procurement, and distribution decision-making system integrating artificial intelligence and automation technologies, industrial big data technologies, geographic information system technologies, RAID technologies, and Internet of Things technologies. This system achieves efficient and intelligent management from the submission of material requirements to delivery to the destination. Information transmission is real-time and accurate, breaking down information silos and enabling real-time sharing and synchronous updates of material information across departments. Departments can collaborate based on accurate real-time data, improving decision-making efficiency. Refined inventory management improves inventory accuracy and turnover, effectively reducing inventory backlog and stockouts, lowering capital tied up in inventory, and freeing up funds for other business development needs. Effective cost control is achieved through intelligent procurement assistance, precise inventory management, reduced material waste, and optimized transportation routes, lowering overall costs and enhancing profitability. Scientific decision-making allows management to access multi-dimensional, in-depth material management analysis reports in real time, providing more comprehensive decision-making support, improving accuracy, and avoiding economic losses due to decision-making errors. Quality data traceability effectively ensures tool quality, improving production safety and efficiency. It achieves efficient, precise, and intelligent management of the entire material warehousing process, optimizes inventory management, effectively controls costs, assists in scientific decision-making, and meets the needs of refined tool management, ultimately improving the overall operational efficiency and economic benefits of the power plant.

[0058] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0059] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0060] The platform embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0061] While numerous embodiments of the invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of the invention. The appended claims are intended to define the scope of protection of the invention and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A coal-fired power plant material warehousing digital management platform, characterized in that, include: The material identification code assignment module is used to assign a unique identification code to materials so that the unique identification code can be identified by scanning devices at each stage of the platform, and the material data generated at each stage of the platform can be associated with the unique identification code. The material data acquisition module is used to automatically collect material data generated at each stage of the platform. The procurement plan generation module is used to acquire historical data, generate training and testing sets based on the historical data, construct a procurement prediction model based on ITransformer, train the procurement prediction model based on the training set, and validate the procurement prediction model based on the testing set to obtain a trained procurement prediction model. Current data is input into the trained procurement prediction model, and a procurement suggestion list is output. The historical data includes historical material procurement data, historical material inventory data, historical material consumption data, historical market price data, historical supplier data, and historical plan data. The current data includes current material inventory data, current market price data, current material consumption data, current supplier data, and current plan data. The procurement suggestion list includes a unique identifier for each material, material name, predicted demand quantity, predicted demand range, suggested procurement quantity, suggested procurement time, and a list of recommended suppliers. ITransformer is an inverted self-attention Transformer structure. The material storage layout module is used to randomly generate an initial population based on available storage locations. It constructs a fitness function using space utilization and ease of access as benefits, weight exceeding limits, volume exceeding limits, storage location conflicts, passageway blockage, and hazardous materials isolation as hard constraints, and material shelf life as a soft constraint. Through selection, crossover, and mutation operations, the population is iterated continuously, retaining individuals with high fitness. After multiple generations of evolution, the optimal storage layout scheme is found. The initial population consists of multiple chromosomes, each representing a storage layout scheme. Each chromosome is composed of multiple gene segments, and each gene segment consists of a unique identifier, storage location, and material quantity.

2. The coal-fired power plant material warehousing digital management platform of claim 1, wherein, The material storage layout module includes: The initial population generation submodule is used to obtain an idle storage space pool carrying storage space attribute information; for any of the aforementioned materials, based on the material attribute information of the material and the storage space attribute information of the idle storage spaces in the idle storage space pool, an idle storage space is selected from the idle storage space pool to generate a candidate storage space pool for the material; all types of materials are randomly sorted, and the real-time available pool for each type of material is obtained in a random order, and an idle storage space is selected from the real-time available pool as an occupied storage space; a chromosome is generated based on the occupied storage spaces of all types of materials, wherein the real-time available pool is the intersection of the candidate storage space pool of the current sequence of materials and the current storage space pool, and the current storage space pool is the storage space pool of the idle storage space pool after deleting the occupied storage spaces of all preceding materials; the idle storage space pool is copied, and multiple chromosomes are regenerated to form the initial population.

3. The coal-fired power plant material warehousing digital management platform according to claim 1 or 2, characterized in that, The material storage layout module includes: The revenue calculation submodule is used to: for any chromosome, calculate the ratio of the sum of the volumes of all materials on the chromosome to the sum of the volumes of all occupied storage locations on the chromosome, as the space utilization rate; for any gene segment, calculate the Manhattan distance from the storage location corresponding to the gene segment to all entrances and exits, and take the minimum value as the first distance; sum the first distance and the second distance to obtain the single picking cost of the gene segment, where the second distance is the product of the climbing equivalence coefficient and the climbing level; obtain the annual outbound frequency and the quantity stored in the storage location of the materials corresponding to the gene segment, and multiply the single picking cost, the annual outbound frequency, and the quantity stored to obtain the annual total picking distance of the gene segment; sum the annual total picking distance of all gene segments on the same chromosome, and then normalize it to obtain the retrieval convenience value; perform a weighted summation of the space utilization rate and the retrieval convenience value to obtain the revenue value; and / or, The penalty value calculation submodule is configured to determine, for any chromosome, whether at least one of the following conditions exists: a storage location conflict, a volume overrun, a weight overrun, a passage blockage, and a hazardous isolation; if yes, eliminate the chromosome; otherwise, set the hard constraint penalty value to 0; for any chromosome, calculate the remaining shelf life proportion of each material on the chromosome; if the remaining shelf life proportion is not less than a preset proportion, set the corresponding sub-soft constraint penalty value to 0; if the remaining shelf life proportion is less than the preset proportion, calculate the corresponding sub-soft constraint penalty value according to a first formula, the first formula being: , wherein, is a sub-soft constraint penalty value of the jth material, is a remaining shelf life of the jth material, j = 1, 2, …, M, M is a total number of material types on the chromosome, j and M are positive integers, is the preset proportion; sum all the sub-soft constraint penalty values of the materials and multiply by a soft constraint coefficient to obtain a soft constraint penalty value; sum the hard constraint penalty value and the soft constraint penalty value to obtain a penalty value.

4. The coal-fired power plant material warehousing digital management platform of claim 3, wherein, The material storage layout module also includes: The fitness function construction submodule is used to construct the fitness function based on the difference between the reward value and the penalty value.

5. The coal-fired power plant material warehousing digital management platform according to claim 1 or 2, characterized in that, The objective function of the procurement prediction model in the procurement scheme generation module is to minimize the following composite loss: In the formula, θ represents all training parameters; a mean squared error loss for point predictions, i.e., point prediction loss; , Quantile loss for interval prediction, i.e., interval prediction loss; α and β are hyperparameters used to balance the importance of the point prediction loss and the interval prediction loss in the overall objective; i = 1, 2, ..., N, where N is the number of samples in the training set, and i and N are both positive integers; HistoricalMaterialConsumptioni for the ith sample; Output the point prediction value for the i-th sample, i.e., the predicted demand.

6. The coal-fired power plant material warehousing digital management platform of claim 1, wherein, The digital management platform for material storage in coal-fired power plants also includes: The inbound / outbound management module is used to deploy and manage automated inbound / outbound equipment.

7. The digital management platform for material storage in coal-fired power plants as described in claim 1 or 2, characterized in that, The digital management platform for material storage in coal-fired power plants also includes: The report generation module is used to generate material reports based on preset templates. The material reports include inventory balance reports, material turnover rate reports, procurement cost reports, and tool usage frequency reports.

8. The digital management platform for material storage in coal-fired power plants as described in claim 1 or 2, characterized in that, The digital management platform for material storage in coal-fired power plants also includes: The material quality assurance traceability module is used to establish a material quality assurance information database that stores quality assurance information of various materials and strongly associates it with unique identification codes. When it is necessary to trace the quality of materials, the unique identification code is scanned and the quality assurance information corresponding to the unique identification code is obtained by querying the material quality assurance information database. The quality assurance information includes quality assurance documents, production batches, raw material sources, inspection reports, and quality certification certificates.

9. The digital management platform for material storage in coal-fired power plants as described in claim 1 or 2, characterized in that, The digital management platform for material storage in coal-fired power plants also includes: The warehouse environment control module includes an environmental sensor and an environmental control device. When the environmental parameters collected by the environmental sensor exceed the preset range, it determines the abnormal environmental area, activates the corresponding environmental control device to adjust the environmental parameters of the abnormal environmental area, generates a list of materials affected by the environmental anomaly, and issues an early warning message containing the list of materials. The environmental sensors include temperature and humidity sensors, smoke detectors, and vibration sensors; the environmental control equipment includes ventilation equipment, dehumidification equipment, and cooling equipment.

10. The digital management platform for material storage in coal-fired power plants as described in claim 1 or 2, characterized in that, The digital management platform for material storage in coal-fired power plants also includes: The positioning alarm module is used to monitor the location of materials in real time using positioning technology, and to issue an alarm signal when the location of materials changes abnormally.

Images