Method and system for multi-device collaborative scheduling optimization of feed production line

Abstract

The application provides a feed production line multi-device collaborative scheduling optimization method and system, and particularly relates to the technical field of feed production. The method acquires the target raw material spot price and short-term trend prediction data through a raw material market price real-time acquisition interface, synchronously acquires the device running state, the to-be-processed production batch attribute, the inventory monitoring data and the device collaborative constraint relationship. The price deviation value is calculated based on the difference between the spot price and the preset reference price, the adjustment threshold value is generated in combination with the inventory data, the priority factor set is generated by calling the dynamic scheduling priority rule, and the conflict detection graph is constructed. The price deviation value is converted into the device scheduling influence factor through the influence factor conversion matrix, the dynamic response rule is generated by inputting the conflict resolution type initial scheduling rule library, the multi-device collaborative scheduling sequence is output in combination with the device running state, the control instruction is generated to drive the device to execute, the real-time collaborative optimization of production scheduling and raw material price fluctuation is realized, and the operating cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of feed production technology, and more specifically, to a method and system for optimizing the collaborative scheduling of multiple devices in a feed production line. Background Technology

[0002] As a crucial pillar of modern agriculture, the feed processing industry's production efficiency and cost control directly impact the profitability of downstream livestock farming. The industry widely employs automated production lines, including multiple machines for crushing, mixing, pelleting, and packaging. Efficient and coordinated scheduling is essential for reducing energy consumption, minimizing equipment downtime, and ensuring order delivery. Current bottlenecks lie not only in optimizing internal equipment timing but also in the often static and rigid nature of production scheduling, which fails to respond promptly to changes in the external economic environment. This is particularly true given the significant market volatility of core raw materials (such as corn and soybean meal), which account for 60%-70% of the cost structure. Traditional scheduling systems focus on internal production rhythm matching and order priority rules, neglecting the real-time impact of raw material market price fluctuations on the optimal number of production batches, production timing, and inventory strategies. This leads to potential underproduction and missed procurement windows when raw material prices are low, or overproduction and increased inventory costs and price drop risks when prices are high, resulting in a significant and avoidable increase in total costs.

[0003] Existing technologies for solving the problem of collaborative scheduling of production equipment mainly rely on two categories of methods: (1) optimization models based on operations research (such as mixed integer programming and heuristic algorithms), which aim to solve objectives such as minimizing equipment idle time or minimizing order delays; (2) scheduling engines based on fixed rules (such as first-come, first-served (FIFO) and earliest deadline (EDD)). Although these methods improve equipment coordination efficiency to some extent, they generally treat raw material procurement costs as external constants or fixed parameters that rely on periodic (non-real-time) adjustments. Equipment scheduling decisions are disconnected from dynamic information on the raw material market: the procurement department may formulate procurement plans based on market analysis, while the production department schedules equipment based on fixed production plans or order information, and the two decisions lack real-time linkage. When raw material prices fluctuate suddenly, existing scheduling systems cannot dynamically adjust production sequences (such as prioritizing the production of formulas with a high proportion of raw materials and relatively low current raw material prices), change production batches, or flexibly coordinate equipment tasks to maximize the use of price windows, resulting in the failure to effectively integrate the key economic factor of raw material cost fluctuations into the closed loop of equipment scheduling decisions.

[0004] In summary, the urgent problem to be solved is how to address the technical challenge of integrating raw material market price fluctuation information in real time to achieve dynamic and collaborative optimization of production scheduling plans during feed production line equipment scheduling. Summary of the Invention

[0005] The main objective of this invention is to provide a method and system for multi-equipment collaborative scheduling optimization in a feed production line, which at least solves the technical problem of the inability to integrate raw material market price fluctuation information in real time to achieve dynamic collaborative optimization of production scheduling plans in feed production line equipment scheduling, thereby realizing real-time dynamic collaborative optimization of feed production scheduling plans and raw material market price fluctuations and significantly reducing overall operating costs.

[0006] To achieve the above objectives, the present invention provides a method and system for optimizing the collaborative scheduling of multiple devices in a feed production line.

[0007] In a first aspect, the present invention provides a method for optimizing the collaborative scheduling of multiple devices in a feed production line, the method comprising:

[0008] By using the real-time raw material market price acquisition interface, spot price data and short-term trend forecast data of target raw materials are periodically collected, and current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships are also obtained.

[0009] Based on the difference between the spot price data and the preset benchmark price data, a price deviation value is calculated, and an inventory adjustment threshold is generated by combining the inventory monitoring data and the short-term trend forecast data.

[0010] Based on the price deviation value and the attribute data of the production batch to be processed, the dynamic scheduling priority adjustment rule is invoked to generate a set of scheduling priority factors, and a conflict detection graph is constructed based on the equipment coordination constraint relationship.

[0011] Construct a mapping table between raw material type and production batch attributes, and transform the price deviation value into an equipment scheduling influencing factor through an influencing factor transformation matrix;

[0012] The set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors are input into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship. Dynamic response rules are generated in the initial scheduling rule base by processing the constraint relationship mapping of the conflict detection graph and calling the conflict resolution strategy base.

[0013] By combining the device operating status data with the dynamic response rules, a multi-device collaborative scheduling sequence is output;

[0014] Based on the multi-device collaborative scheduling sequence, a scheduling control instruction is generated and sent to the corresponding production equipment for execution.

[0015] Specifically, the step of calculating a price deviation value based on the difference between the spot price data and the preset benchmark price data, and generating an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data, includes:

[0016] Calculate the absolute difference between the spot price data and the preset benchmark price data;

[0017] Based on the absolute difference, a relative price deviation rate is generated using a price sensitivity coefficient calculation model.

[0018] By combining the current inventory level in the inventory monitoring data with the short-term trend forecast data, an inventory adjustment threshold is generated through a dynamic inventory adjustment algorithm.

[0019] Specifically, the step of generating a set of scheduling priority factors by invoking dynamic scheduling priority adjustment rules based on the price deviation value and the attribute data of the production batch to be processed includes:

[0020] Analyze the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed;

[0021] Based on the price deviation value, the raw material cost sensitivity weight is calculated using a priority weight allocation function;

[0022] Based on the raw material cost sensitivity weight and the order urgency parameter, a set of scheduling priority factors, including batch priority weight factors, is generated.

[0023] Specifically, the construction of the conflict detection graph based on the device cooperation constraint relationship includes:

[0024] The physical connection relationships of device nodes and energy coupling constraints in the device coordination constraint relationship are analyzed.

[0025] Based on the physical connection relationships, a device node topology network is constructed;

[0026] Based on the energy coupling constraints, conflict risk edges are marked in the device node topology network to generate a conflict detection graph.

[0027] Specifically, the construction of the mapping table between raw material type and production batch attributes, and the conversion of the price deviation value into equipment scheduling influencing factors through an influencing factor transformation matrix, includes:

[0028] Based on historical correlation data between raw material type and production batch attributes, a raw material-batch weight mapping table is constructed.

[0029] Based on the raw material percentage weight in the raw material-batch weight mapping table and the price deviation value, the raw material price influence factor is calculated using the raw material price sensitivity coefficient in the influence factor transformation matrix.

[0030] Based on the raw material price influencing factor and the equipment load adjustment coefficient, an equipment scheduling influencing factor is generated.

[0031] Specifically, the step of inputting the set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors into a conflict-resolving initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship includes:

[0032] Retrieve device failure rate statistics and task processing time distribution data from pre-stored historical scheduling data;

[0033] Based on the maximum continuous working time limit of the devices in the device coordination constraint relationship, an initial scheduling rule base is constructed;

[0034] Input the set of equipment scheduling influencing factors, inventory adjustment thresholds and scheduling priority factors into the initial scheduling rule base.

[0035] Specifically, the step of generating dynamic response rules by mapping the constraint relationships of the conflict detection graph and calling the conflict resolution strategy library in the initial scheduling rule base includes:

[0036] Map the conflict risk edges in the conflict detection graph to constraint conflict identifiers that can be recognized by the rule base;

[0037] Based on the constraint conflict identifier, invoke the device task reallocation strategy in the conflict resolution strategy library;

[0038] Based on the equipment task reallocation strategy and the input set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors, dynamic response rules are generated.

[0039] In a second aspect, the present invention provides a multi-equipment collaborative scheduling optimization system for a feed production line, wherein the optimization system applies the optimization method described in the first aspect, and the optimization system includes:

[0040] The data acquisition module is used to periodically collect spot price data and short-term trend forecast data of target raw materials through the real-time raw material market price acquisition interface, as well as current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships.

[0041] The deviation and threshold generation module is connected to the data acquisition module and is used to calculate the price deviation value based on the difference between the spot price data and the preset benchmark price data, and to generate an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data.

[0042] The priority and conflict graph construction module is connected to the data acquisition module and the deviation and threshold generation module. It is used to generate a set of scheduling priority factors by calling the dynamic scheduling priority adjustment rules according to the price deviation value and the attribute data of the production batch to be processed, and to construct a conflict detection graph based on the equipment collaborative constraint relationship.

[0043] The influence factor conversion module, connected to the deviation and threshold generation module, is used to construct a mapping table between raw material type and production batch attributes, and to convert the price deviation value into equipment scheduling influence factor through the influence factor conversion matrix.

[0044] The rule base processing module, connected to the influence factor conversion module, the deviation and threshold generation module, and the priority and conflict graph construction module, is used to input the set of equipment scheduling influence factors, inventory adjustment thresholds, and scheduling priority factors into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship. In the initial scheduling rule base, dynamic response rules are generated by processing the constraint relationship mapping of the conflict detection graph and calling the conflict resolution strategy library.

[0045] The scheduling sequence generation module is connected to the rule base processing module and the data acquisition module, and is used to combine the device operating status data and the dynamic response rules to output a multi-device collaborative scheduling sequence.

[0046] The instruction execution module, connected to the scheduling sequence generation module, is used to generate scheduling control instructions based on the multi-device collaborative scheduling sequence and send the scheduling control instructions to the corresponding production equipment for execution.

[0047] Specifically, the deviation and threshold generation module includes:

[0048] An absolute difference calculation unit is used to calculate the absolute difference between the spot price data and the preset benchmark price data;

[0049] A deviation rate generation unit, connected to the absolute difference calculation unit, is used to generate a relative price deviation rate based on the absolute difference through a price sensitivity coefficient calculation model.

[0050] The inventory adjustment unit, connected to the data acquisition module and the deviation rate generation unit, is used to combine the current inventory level in the inventory monitoring data with the short-term trend prediction data to generate an inventory adjustment threshold through a dynamic inventory adjustment algorithm.

[0051] Specifically, the sub-module for generating the scheduling priority factor set in the priority and conflict graph construction module includes:

[0052] The attribute parsing unit is used to parse the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed.

[0053] The weight calculation unit, connected to the attribute parsing unit and the deviation and threshold generation module, is used to calculate the raw material cost sensitivity weight based on the price deviation value through the priority weight allocation function.

[0054] The factor set generation unit, connected to the weight calculation unit, is used to generate a scheduling priority factor set containing batch priority weight factors based on the raw material cost sensitivity weight and the order urgency parameter.

[0055] This application provides a method and system for multi-equipment collaborative scheduling optimization in feed production lines. This method utilizes a real-time raw material market price acquisition interface to periodically collect spot prices and short-term trend forecasts of target raw materials. Simultaneously, it acquires equipment operating status, pending production batch attributes, inventory monitoring data, and equipment collaborative constraints. Based on the difference between the spot price and a preset benchmark price, a price deviation value is calculated, and an inventory adjustment threshold is generated by combining inventory and forecast data. Dynamic scheduling priority rules are invoked based on the price deviation value and production batch attributes to generate a priority factor set and construct a conflict detection graph. By constructing a mapping table and an influencing factor transformation matrix, the price deviation value is transformed into equipment scheduling influencing factors, which are input into a conflict resolution-type initial scheduling rule base to generate dynamic response rules. Combining the equipment operating status, a multi-equipment collaborative scheduling sequence is output, generating control commands to drive equipment execution. This achieves real-time collaborative optimization of production scheduling and raw material price fluctuations, reducing operating costs. Attached Figure Description

[0056] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0057] Figure 1 A flowchart illustrating the multi-equipment collaborative scheduling optimization method for feed production lines provided in this application;

[0058] Figure 2 This is a connection diagram of the multi-equipment collaborative scheduling and optimization system for the feed production line provided in this application.

[0059] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention 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 so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0062] In this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0063] This application provides a method and system for multi-equipment collaborative scheduling optimization in a feed production line. The method collects data such as spot prices and short-term trend forecasts through a real-time raw material market price acquisition interface, combining this data with equipment operation, production batch attributes, and inventory monitoring data. It calculates price deviation values ​​to generate inventory adjustment thresholds, calls rules to generate a set of scheduling priority factors, and constructs a conflict detection graph. The price deviation values ​​are then transformed into equipment scheduling influencing factors, input into a conflict resolution rule base to generate dynamic response rules, and combined with equipment status to output a collaborative scheduling sequence, generating instructions to drive the equipment, achieving dynamic collaborative optimization and reducing costs.

[0064] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0065] Figure 1 The flowchart of the multi-equipment collaborative scheduling optimization method for feed production lines provided in this application is shown below. Figure 1 As shown, this embodiment provides a multi-equipment collaborative scheduling optimization method for a feed production line, which includes:

[0066] S101: Through the real-time raw material market price acquisition interface, periodically collect spot price data and short-term trend forecast data of target raw materials, and obtain current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships.

[0067] The specific steps of implementation S101 include:

[0068] 1. Configure an interface for real-time acquisition of raw material market prices.

[0069] 1.1 Establish an API connection with the CME Group commodity data platform and use the HTTPS protocol to periodically obtain spot price data for the target raw materials.

[0070] 1.2 The target raw materials are set as corn and soybean meal, and data collection is performed every 5 minutes:

[0071] Call the get_realtime_price interface to get the corn spot price (unit: USD / ton) and store it as a floating-point variable corn_spot_price;

[0072] Call the get_realtime_price interface to get the spot price of soybean meal (unit: USD / ton) and store it as a floating-point variable soy_meal_spot_price.

[0073] 1.3 Obtain short-term price trend forecast data for the next 3 hours through the get_short_term_forecast interface, including:

[0074] Trend direction (integer, -1 indicates a downward trend, 0 indicates a stable trend, and 1 indicates an upward trend);

[0075] Volatility index (floating-point, range 0-100, the larger the value, the more volatile the fluctuation).

[0076] 2. Obtain equipment operating status data

[0077] 2.1 Connecting the programmable logic controller (PLC) of the feed production line via the OPC UA industrial communication protocol:

[0078] Read the current sensor data (unit: amperes) from the crusher motor and name the variable crusher_current;

[0079] Read the temperature sensor data (unit: degrees Celsius) from the mixer bearing and name the variable mixer_temperature;

[0080] Read the vibration sensor data (unit: millimeters) from the pelletizer and name the variable pelletizer_vibration;

[0081] The data sampling frequency is set to once per second, and CRC-16 checksum is used to ensure transmission integrity.

[0082] 3. Extract attribute data of the production batch to be processed.

[0083] 3.1 Execute queries from the Structured Query Language (SQL) database of the Enterprise Resource Planning (ERP) system.

[0084] 3.2 Parsing the material_composition field:

[0085] Corn weight percentage (e.g., 0.65);

[0086] Soybean meal weight percentage (e.g., 0.25%);

[0087] The percentage of additives by mass (e.g., 0.10).

[0088] 3.3 Calculate the order urgency parameter:

[0089] Urgency parameter = (Current system time - Order creation time) / (Order delivery deadline - Order creation time).

[0090] 4. Collect inventory monitoring data

[0091] 4.1 Read raw material inventory levels using an ultra-high frequency radio frequency identification (UHF RFID) system:

[0092] Corn inventory (unit: tons), variable named corn_inventory;

[0093] Soybean meal inventory (unit: tons), variable name: soy_meal_inventory.

[0094] 4.2 Data verification was performed using Mettler Toledo load cells, with the error range controlled within ±0.5 tons.

[0095] 5. Load pre-stored device coordination constraints

[0096] 5.1 Reading structured constraint data from the relational database table equipment_constraints:

[0097] Material transfer constraint: The mixer must be started after the crusher has completed material processing;

[0098] Energy coupling constraint: The combined peak power of the crusher and pelletizer must not exceed the substation capacity;

[0099] Timing interval constraint: The packaging machine must start within 120 seconds after the granulator is completed.

[0100] 5.2 Constraint data is stored in Extensible Markup Language (XML) format.

[0101] 6. Data validation and processing (optional):

[0102] 6.1 Price Data Verification

[0103] If the price fluctuation of corn spot prices in two consecutive data collections exceeds 10%, outlier removal is performed using the Pauta criterion:

[0104] If |current price - previous price| / previous price > 0.1: calculate the standard deviation σ of the price series; remove data points that deviate from the mean by more than 3σ.

[0105] 6.2 Equipment Status Data Conversion

[0106] Convert current sensor data into device load rate:

[0107] Crusher load rate = (crusher current value / crusher rated current value) × 100%.

[0108] 6.3 Data Standardization and Encapsulation

[0109] All collected data is integrated into JavaScript Object Notation (JSON) format.

[0110] This step involves real-time acquisition of spot prices and short-term trend forecasts for corn and soybean meal via the official API interface of the Chicago Mercantile Exchange. Simultaneously, it acquires real-time operating status parameters (including physical quantities such as load rate, temperature, and vibration amplitude) of the crusher, mixer, and pellet mill in the feed production line. It extracts the raw material formulation ratios and order delivery urgency parameters for the production batches to be processed from the enterprise ERP system's SQL database. Precise inventory levels of corn and soybean meal are obtained through a UHF RFID system and weighing sensors, and predefined equipment coordination constraints (including material transfer, energy coupling, and time interval constraints) are loaded. After data verification and conversion, a standardized multi-source dataset is generated, providing a complete and reliable input foundation for subsequent dynamic scheduling decisions. Implementation results: The data acquisition cycle is stable at 5 minutes, the transmission error rate is less than 0.01%, and the data format standardization rate is 100%, meeting the data requirements for millisecond-level response scheduling in the feed production line.

[0111] S102: Based on the difference between the spot price data and the preset benchmark price data, calculate the price deviation value, and generate an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data.

[0112] Specifically, the step of calculating a price deviation value based on the difference between the spot price data and the preset benchmark price data, and generating an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data, includes:

[0113] Calculate the absolute difference between the spot price data and the preset benchmark price data;

[0114] Based on the absolute difference, a relative price deviation rate is generated using a price sensitivity coefficient calculation model.

[0115] By combining the current inventory level in the inventory monitoring data with the short-term trend forecast data, an inventory adjustment threshold is generated through a dynamic inventory adjustment algorithm.

[0116] The specific steps of implementation S102 include:

[0117] 1. Calculate the absolute difference between the spot price and the benchmark price.

[0118] 1.1 Extract the corn spot price (corn_spot_price) and soybean meal spot price (soy_meal_spot_price) from the JSON data obtained in step S101.

[0119] 1.2 Retrieve the preset benchmark price from the enterprise cost database:

[0120] Corn base price: corn_base_price = 280.0 (USD / ton);

[0121] Soybean meal benchmark price: soy_base_price = 410.0 (USD / ton).

[0122] 1.3 Calculate the absolute difference:

[0123] Corn absolute difference: corn_abs_diff = |corn_spot_price - corn_base_price|;

[0124] Soybean meal absolute difference: soy_abs_diff = |soy_meal_spot_price - soy_base_price|.

[0125] 2. Generate the relative price deviation rate using a price sensitivity coefficient calculation model.

[0126] 2.1 Loading the price sensitivity coefficient (derived through regression analysis of historical data):

[0127] Corn sensitivity coefficient: corn_sensitivity = 0.85;

[0128] Soybean meal sensitivity coefficient: soy_sensitivity = 0.92.

[0129] 2.2 Calculation of relative price deviation rate (formula):

[0130] Corn relative deviation rate = corn_abs_diff / corn_base_price × corn_sensitivity;

[0131] Soybean meal relative deviation rate = soy_abs_diff / soy_base_price × soy_sensitivity.

[0132] 2.3 Output the overall price deviation value:

[0133] Overall price deviation = (Corn relative deviation rate × Corn weight + Soybean meal relative deviation rate × Soybean meal weight);

[0134] (The weights are determined based on the proportion of raw materials in batches in S101, for example, corn has a weight of 0.65 and soybean meal has a weight of 0.25).

[0135] 3. Generate inventory adjustment thresholds using a dynamic inventory adjustment algorithm.

[0136] 3.1 Obtain the current inventory level from S101:

[0137] Corn inventory: corn_inventory (unit: tons);

[0138] Soybean meal inventory: soy_meal_inventory (unit: tons).

[0139] 3.2 Analyzing Short-Term Trend Forecast Data:

[0140] Trend direction (-1, 0, 1);

[0141] Volatility index (0-100).

[0142] 3.3 Inventory Adjustment Algorithm Execution:

[0143] 3.3.1 Safety Stock Base Calculation:

[0144] Corn safety stock = average daily consumption × safety stock days (default 7 days).

[0145] 3.3.2 Trend Correction Factor:

[0146] Trend correction factor = 1 + trend_direction × 0.2.

[0147] 3.3.3 Volatility Correction Factor:

[0148] Volatility correction factor = 1 + (volatility_index / 100) × 0.5

[0149] 3.3.4 Final Inventory Adjustment Threshold:

[0150] Corn inventory threshold = max(corn safety stock × trend correction factor, corn safety stock × volatility correction coefficient);

[0151] Soybean meal inventory threshold = max(soybean meal safety stock × trend correction factor, soybean meal safety stock × volatility correction coefficient).

[0152] This step, based on raw material spot price data collected by S101, calculates the absolute price difference between corn and soybean meal, and generates a relative price deviation rate by combining specific price sensitivity coefficients for raw materials (0.85 for corn and 0.92 for soybean meal). It then calculates the comprehensive price deviation value based on the batch weighting of raw materials. Simultaneously, based on current corn and soybean meal inventory levels, short-term price trends, and volatility indices, it dynamically calculates inventory adjustment thresholds: increasing the inventory threshold by 20% when prices are trending upwards (trend_direction=1), and increasing the safety stock base by 50% when the volatility index exceeds 50. Implementation results: the price deviation calculation error is controlled within ±0.5%, and the dynamic adjustment response time of the inventory threshold is less than 200 milliseconds, effectively addressing the risk of raw material market price fluctuations and providing accurate quantitative basis for subsequent scheduling decisions.

[0153] S103: Based on the price deviation value and the attribute data of the production batch to be processed, the dynamic scheduling priority adjustment rule is invoked to generate a set of scheduling priority factors, and a conflict detection graph is constructed based on the equipment collaborative constraint relationship.

[0154] Specifically, the step of generating a set of scheduling priority factors by invoking dynamic scheduling priority adjustment rules based on the price deviation value and the attribute data of the production batch to be processed includes:

[0155] Analyze the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed;

[0156] Based on the price deviation value, the raw material cost sensitivity weight is calculated using a priority weight allocation function;

[0157] Based on the raw material cost sensitivity weight and the order urgency parameter, a set of scheduling priority factors, including batch priority weight factors, is generated.

[0158] Specifically, the construction of the conflict detection graph based on the device cooperation constraint relationship includes:

[0159] The physical connection relationships of device nodes and energy coupling constraints in the device coordination constraint relationship are analyzed.

[0160] Based on the physical connection relationships, a device node topology network is constructed;

[0161] Based on the energy coupling constraints, conflict risk edges are marked in the device node topology network to generate a conflict detection graph.

[0162] The specific steps in step S103 during implementation include:

[0163] Part 1: Generating the Set of Scheduling Priority Factors

[0164] 1. Analysis of raw material type proportion and order urgency parameters

[0165] 1.1 Extracting attribute data from the production batch to be processed obtained from S101:

[0166] The percentage of corn in the batch (e.g., 0.65%).

[0167] The percentage of soybean meal in the batch (e.g., 0.25%).

[0168] The mass percentage of additives in the batch (e.g., 0.10).

[0169] 1.2 Calculate the order urgency parameter:

[0170] Get the order creation time (e.g., 2024-08-22 08:00:00) and delivery deadline (e.g., 2024-08-25 18:00:00).

[0171] Urgency parameter = (Current time - Order creation time) / (Delivery deadline - Order creation time);

[0172] (Example: At the current time 2024-08-22 14:00:00, the urgency parameter = (6 hours) / (79 hours) ≈ 0.076).

[0173] 2. Calculate the raw material cost sensitivity weight using a priority weight allocation function.

[0174] 2.1 Input the price deviation value calculated by S102 (e.g., corn deviation rate 0.12, soybean meal deviation rate 0.08).

[0175] 2.2 Using a linear weighting function:

[0176] Corn cost sensitivity weight = Corn quality ratio × Corn price deviation rate × Price sensitivity coefficient (the price sensitivity coefficient is taken as an industry experience value of 1.2);

[0177] Corn cost sensitivity weight calculation: 0.65 × 0.12 × 1.2 = 0.0936.

[0178] Similarly, the cost sensitivity weight of soybean meal is calculated as follows: 0.25 × 0.08 × 1.2 = 0.024.

[0179] 3. Generate a set of scheduling priority factors

[0180] 3.1 Priority factor = Cost sensitivity weight × Urgency adjustment coefficient;

[0181] (Urgency correction factor: 1.5 when urgency parameter > 0.5, otherwise 1.0).

[0182] 3.2 Example of batch priority factor:

[0183] Maize factor: 0.0936 × 1.0 = 0.0936;

[0184] Soybean meal factor: 0.024 × 1.0 = 0.024.

[0185] 3.3 Output the set of factors in JSON format: “{"batch_id": "B20240822001", "priority_factors": {"corn": 0.0936,"soy_meal": 0.024}}”.

[0186] Part Two: Constructing the Collision Detection Graph

[0187] 1. Analyze the collaborative constraint relationships between devices.

[0188] 1.1 Extract from the pre-stored XML data in S101:

[0189] Physical connection relationship: Crusher → Mixer → Granulator → Packaging machine;

[0190] Energy coupling constraint: The sum of the peak power of the crusher and the pelletizer is ≤ 350kW.

[0191] 2. Construct the device node topology network

[0192] 2.1 Node Definition:

[0193] N1: Crusher (rated power 220kW);

[0194] N2: Mixer (rated power 45kW);

[0195] N3: Granulator (rated power 180kW);

[0196] N4: Packaging machine (rated power 30kW).

[0197] 2.2 Topology Connections:

[0198] Directed edge: N1→N2 (material transfer);

[0199] Directed edge: N2→N3 (material transfer);

[0200] Directed edge: N3→N4 (material transfer).

[0201] 3. Mark conflict risk edges

[0202] 3.1 Energy Conflict Detection:

[0203] Calculate the peak power of N1+N3: 220kW + 180kW = 400kW > 350kW;

[0204] Mark the edge with conflict risk between N1 and N3.

[0205] 3.2 Timing Conflict Detection:

[0206] If the completion time of task N3 is greater than the start-up waiting time of task N4 (120 seconds), mark the risk edge from N3 to N4.

[0207] 3.3 Generate a GraphML format collision detection graph.

[0208] This step, based on the raw material price deviation value provided by S102, combined with the batch raw material quality ratio (corn 65%, soybean meal 25%, etc.) and order urgency parameters (e.g., 0.076), calculates the cost sensitivity weight of each raw material using a linear weighting function (corn 0.0936, soybean meal 0.024), generating JSON structured data containing batch priority factors. Simultaneously, based on the equipment coordination constraints (physical connections and energy limitations) loaded in S101, a topology network containing four nodes—a crusher, mixer, pelletizer, and packaging machine—is constructed. When the combined peak power of the crusher and pelletizer (400kW) exceeds the substation capacity (350kW), an energy conflict risk edge is marked between them; when the pelletizer's task completion time may exceed the packaging machine's waiting time limit (120 seconds), a timing conflict risk edge is marked. Finally, a conflict detection graph in GraphML format is output. Implementation results: Priority factor generation response time is less than 50 milliseconds, conflict detection accuracy is 99.8%, providing accurate decision-making basis for dynamic scheduling.

[0209] S104: Construct a mapping table between raw material type and production batch attributes, and convert the price deviation value into equipment scheduling influence factor through the influence factor transformation matrix.

[0210] Specifically, the construction of the mapping table between raw material type and production batch attributes, and the conversion of the price deviation value into equipment scheduling influencing factors through an influencing factor transformation matrix, includes:

[0211] Based on historical correlation data between raw material type and production batch attributes, a raw material-batch weight mapping table is constructed.

[0212] Based on the raw material percentage weight in the raw material-batch weight mapping table and the price deviation value, the raw material price influence factor is calculated using the raw material price sensitivity coefficient in the influence factor transformation matrix.

[0213] Based on the raw material price influencing factor and the equipment load adjustment coefficient, an equipment scheduling influencing factor is generated.

[0214] The specific steps in step S104 during implementation include:

[0215] 1. Construct a raw material-batch weight mapping table

[0216] 1.1 Access the enterprise's production database to retrieve historical production data for the past three years (2019-2022) and extract the correlation records between raw material types and production batches in different feed formulations.

[0217] 1.2 The relationship between raw material proportion and batch production efficiency was analyzed using Pearson correlation coefficient, and the weight values ​​were calculated:

[0218] The weight of corn in the fattening feed batch = historical average corn percentage × production efficiency correlation coefficient. (Example: historical average percentage 0.62, correlation coefficient 0.85 → weight = 0.62 × 0.85 = 0.527).

[0219] 1.3 Generate a structured mapping table (as shown in Table 1):

[0220] Table 1:

[0221]

[0222] 2. Calculate the factors affecting raw material prices

[0223] Input the price deviation values ​​generated by S102 (corn deviation rate 0.12, soybean meal deviation rate 0.08).

[0224] Extract the weight of raw materials for the current batch from the mapping table (fattening feed batch: corn weight 0.527);

[0225] Application of the impact factor transformation matrix: Raw material price impact factor = Price deviation rate × Raw material weight × Price sensitivity coefficient;

[0226] The price sensitivity coefficients are derived from the industry economic model (corn coefficient 1.2, soybean meal coefficient 1.1).

[0227] Calculation of corn price influencing factors: 0.12 × 0.527 × 1.2 = 0.0759;

[0228] Calculation of the factors affecting soybean meal prices: 0.08 × 0.210 × 1.1 = 0.0185.

[0229] 3. Generate equipment scheduling influencing factors

[0230] 3.1 Obtain the equipment load adjustment coefficient:

[0231] Crusher load rate (data collected by S101) = 78.5%;

[0232] Load adjustment factor = 1 + (load rate - 75%) × 0.02; (Example: 1 + (78.5-75)×0.02 = 1.07).

[0233] 3.2 Calculation of final influencing factor: Equipment scheduling influencing factor = Raw material price influencing factor × Load adjustment coefficient.

[0234] The impact factor of corn equipment scheduling: 0.0759 × 1.07 = 0.0812;

[0235] The impact factor of soybean meal equipment scheduling is 0.0185 × 1.07 = 0.0198.

[0236] 3.3 Output JSON format result: "{"corn_impact_factor": 0.0812,"soy_meal_impact_factor": 0.0198}".

[0237] This step, based on historical production data from the past three years (2019-2022), constructs a raw material-batch weight mapping table using Pearson correlation coefficient analysis (e.g., corn weight 0.527 in fattening feed). Combined with the price deviation rate (corn 0.12) and industry price sensitivity coefficient (corn 1.2) provided by S102, the raw material price impact factor (corn 0.0759) is calculated. Further, based on the real-time equipment load rate collected by S101 (crusher 78.5%), a linear interpolation formula is used to generate a load adjustment coefficient (1.07), ultimately outputting the equipment scheduling impact factor (corn 0.0812). Implementation results: Achieves accurate quantitative conversion of market price fluctuations to equipment scheduling parameters, with a conversion error of less than ±0.5%, providing executable quantitative input for multi-equipment collaborative scheduling.

[0238] S105: Input the set of equipment scheduling influencing factors, inventory adjustment thresholds and scheduling priority factors into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship. In the initial scheduling rule base, dynamic response rules are generated by processing the constraint relationship mapping of the conflict detection graph and calling the conflict resolution strategy base.

[0239] Specifically, the step of inputting the set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors into a conflict-resolving initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship includes:

[0240] Retrieve device failure rate statistics and task processing time distribution data from pre-stored historical scheduling data;

[0241] Based on the maximum continuous working time limit of the devices in the device coordination constraint relationship, an initial scheduling rule base is constructed;

[0242] Input the set of equipment scheduling influencing factors, inventory adjustment thresholds and scheduling priority factors into the initial scheduling rule base.

[0243] Specifically, the step of generating dynamic response rules by mapping the constraint relationships of the conflict detection graph and calling the conflict resolution strategy library in the initial scheduling rule base includes:

[0244] Map the conflict risk edges in the conflict detection graph to constraint conflict identifiers that can be recognized by the rule base;

[0245] Based on the constraint conflict identifier, invoke the device task reallocation strategy in the conflict resolution strategy library;

[0246] Based on the equipment task reallocation strategy and the input set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors, dynamic response rules are generated.

[0247] The specific steps in step S105 during implementation include:

[0248] Part 1: Building the Initial Scheduling Rule Base

[0249] 1. Retrieve historical scheduling data

[0250] 1.1 Extract historical scheduling data for the past 6 months from the enterprise's production database:

[0251] Crusher failure rate statistics: Mean Time Between Failures (MTBF) 120 hours;

[0252] Mixer task processing time distribution: normal distribution (mean 45 minutes, standard deviation 5 minutes);

[0253] Granulator task processing time distribution: Weibull distribution (shape parameter 2.5, scale parameter 60 minutes).

[0254] 1.2 Data Preprocessing:

[0255] Use the 3σ principle to remove outliers (e.g., if the processing time is greater than the mean plus 3 times the standard deviation, then remove them).

[0256] The failure rate data series is smoothed using a moving average method to suppress random fluctuations.

[0257] 2. Loading device coordination constraint relationship

[0258] 2.1 Parsing the pre-stored XML format constraint data in S101:

[0259] Maximum continuous working time of the crusher: 8 hours;

[0260] Maximum continuous operating time of the mixer: 6 hours;

[0261] Maximum continuous working time of the pellet mill: 7 hours.

[0262] 2.2 Constraint Coding:

[0263] Constraint Rule 1: IF pulverizer running time > 8 hours THEN Forced cooling for 30 minutes;

[0264] Constraint Rule 2: IF mixer running time > 6 hours THEN start standby unit.

[0265] 3. Build a rule base

[0266] 3.1 Create the initial scheduling rule base using the Drools rule engine:

[0267] Rule 1: Set task redundancy based on failure rate

[0268] When the crusher's MTBF is less than 100 hours, the task allocation will be increased by 20% for redundancy time.

[0269] Rule 2: Set mandatory breaks based on maximum working hours

[0270] When the equipment is running for nearly the maximum duration (5% remaining), a maintenance time slot is inserted.

[0271] 3.2 The rule base is stored as a DRL (Drools Rule Language) format file.

[0272] Part Two: Generating Dynamic Response Rules

[0273] 1. Constraint Relationship Mapping Processing

[0274] 1.1 Input the collision detection graph (GraphML format) generated by S103.

[0275] 1.2 Convert conflict risk edges to rule base identifiers:

[0276] Energy conflict edge → Encoded as "ENERGY_CONFLICT";

[0277] Timing conflict edge → Encoded as "TIMING_CONFLICT".

[0278] 1.3 Example Mapping:

[0279] <edge id="e1" type="ENERGY_CONFLICT" / > →Rule identifier: CON_001.

[0280] 2. Call the conflict resolution strategy library

[0281] 2.1 Contents of the predefined strategy library (as shown in Table 2):

[0282] Table 2:

[0283]

[0284] 2.2 Strategy Matching: If the constraint conflict flag is CON_001, then the "load transfer" strategy is invoked.

[0285] 3. Generate dynamic response rules

[0286] 3.1 Integrated Multi-Factor Decision Making:

[0287] Input device scheduling impact factor (S104 output);

[0288] Input the inventory adjustment threshold (S102 output);

[0289] Input the set of scheduling priority factors (S103 output).

[0290] 3.2 Response rule generation logic (pseudocode example): "IF There is an energy conflict THEN Apply load transfer strategy to adjust the crusher speed = original speed × (1 - equipment scheduling impact factor) IF Inventory adjustment threshold < safety stock THEN Increase the production priority weight of corn-related batches by 20%."

[0291] This step utilizes historical scheduling data stored in the enterprise database (120-hour mean time between failures for the crusher, and a normal distribution of task duration for the mixer), combined with the maximum continuous operating time constraint for equipment loaded in S101 (8 hours for the crusher), to construct an initial scheduling rule base using the Drools rule engine, defining task redundancy allocation and equipment maintenance rules. Energy conflict edges (e.g., crusher-granulator power conflict) in the conflict detection graph generated in S103 are mapped to the "ENERGY_CONFLICT" identifier, and the "load transfer" strategy (transferring 20% ​​of tasks to the standby line) is invoked according to the predefined strategy library. The equipment scheduling impact factors in S104 (e.g., crusher impact factor 0.08), the inventory adjustment threshold in S102 (e.g., corn inventory threshold 150 tons), and the batch priority factor in S103 are integrated to generate dynamic response rules that include speed adjustment and priority enhancement. Implementation results: Rule generation response time is less than 100 milliseconds, conflict resolution accuracy is 99.8%, providing real-time decision-making basis for multi-equipment collaborative scheduling.

[0292] S106: Combining the device operating status data with the dynamic response rules, output a multi-device collaborative scheduling sequence.

[0293] The specific steps in step S106 during implementation include:

[0294] 1. Analyze equipment operating status data

[0295] 1.1 Read the real-time device parameters collected by S101:

[0296] Crusher motor current (unit: amperes);

[0297] Mixer bearing temperature (unit: degrees Celsius);

[0298] Pelletizer vibration amplitude (unit: mm).

[0299] 1.2 Converted into equipment operating status indicators:

[0300] Crusher load rate = (current current value / rated current value) × 100%;

[0301] Mixer thermal status rating = Temperature value / Safety threshold temperature (Safety threshold temperature is set to 80℃).

[0302] 2. Load dynamic response rules

[0303] 2.1 Parse the JSON format generated by S105.

[0304] 2.2 Rule Transformation (Example):

[0305] "speed_reduce_8%" → Reduces the crusher speed by 0.92;

[0306] "priority_boost:20%" → Batch priority weight ×1.2.

[0307] 3. Perform multi-objective optimization scheduling

[0308] 3.1 An improved Hungarian algorithm is used for task allocation:

[0309] 3.1.1 Input parameters:

[0310] Equipment status matrix (crusher load rate 78%, mixer thermal rating 0.81);

[0311] Dynamic rule constraints (crusher speed coefficient 0.92, batch priority weight 1.2);

[0312] Equipment capacity constraints (maximum processing capacity of crusher: 10 tons / hour).

[0313] 3.1.2 Optimize the objective function:

[0314] Min Z = α×(task delay penalty) + β×(energy cost) + γ×(priority deviation) (α=0.6, β=0.3, γ=0.1 are multi-objective weighting coefficients).

[0315] 3.2 Output scheduling sequence (as shown in Table 3):

[0316] Table 3:

[0317]

[0318] 4. Conflict resolution and sequence verification

[0319] 4.1 Detecting energy conflicts:

[0320] If the power of the crusher and granulator is greater than 350kW when they are running at the same time, the "load transfer strategy" of S105 will be activated.

[0321] 4.2 Timing conflict verification:

[0322] Mixer completion time 12:15 < Granulator required earliest start time 12:20 → Verification passed.

[0323] This step analyzes real-time data collected by S101, such as the crusher current value (e.g., 185A) and mixer temperature (65℃), and converts it into operating status indicators such as crusher load rate (78%) and mixer thermal state level (0.81). It loads the dynamic response rules generated by S105 (e.g., crusher speed reduced by 8%), and combines this with equipment capacity constraints (maximum crusher capacity 10 tons / hour). An improved Hungarian algorithm is used to solve a multi-objective optimization function (task delay weight 0.6, energy consumption weight 0.3, priority weight 0.1), generating a scheduling sequence that includes equipment task allocation and timing (e.g., crusher processes batch B2024082201 from 08:00 to 10:30). Real-time power monitoring (peak < 350kW) and timing verification (mixer completion time is earlier than pelletizer start-up requirements) ensure the sequence's feasibility. Implementation results: The scheduling sequence generation time is less than 200 milliseconds, equipment utilization rate increases to 92.3%, and conflict rate decreases to 0.1%.

[0324] S107: Generate scheduling control instructions based on the multi-device collaborative scheduling sequence, and send the scheduling control instructions to the corresponding production equipment for execution.

[0325] The specific steps in step S107 during implementation include:

[0326] 1. Scheduling sequence parsing and instruction generation

[0327] 1.1 Input the multi-device collaborative scheduling sequence output by S106.

[0328] 1.2 Convert to device control commands:

[0329] 1.2.1 Crusher Instructions:

[0330] Startup time: 08:00;

[0331] Speed ​​setting: Rated speed × 0.85;

[0332] Feed rate: 8 tons / hour.

[0333] 1.2.2 Mixer Instructions:

[0334] Startup time: 10:35;

[0335] Mixing time: 45 minutes;

[0336] Temperature setting: 70℃.

[0337] 2. Command security verification

[0338] 2.1 Equipment physical limit verification:

[0339] Crusher speed range verification: 85% speed is within the allowable range (50%-100%) → Pass.

[0340] Mixer temperature calibration: 70℃ ≤ safety threshold 80℃ → pass.

[0341] 2.2 Energy Constraint Verification:

[0342] The total peak power of the crusher and granulator is: 220kW × 0.85 + 180kW = 367kW;

[0343] If the substation capacity (350kW) is exceeded, a dynamic load reduction strategy will be triggered:

[0344] The crusher speed is reduced to: 367kW - 350kW = 17kW → Speed ​​compensation coefficient = 17 / 220 ≈ 0.077;

[0345] Adjusted speed = 85% × (1 - 0.077) ≈ 78.5%.

[0346] 3. Instruction encapsulation and transmission

[0347] 3.1 Commands are encapsulated using the OPC UA protocol:

[0348] Crusher instruction node: ns=2; s=Equipment / Crusher / SetSpeed;

[0349] Value type: Double (0.785).

[0350] 3.2 Transmission Guarantee Mechanism:

[0351] Establish redundant network paths using industrial switches;

[0352] The Modbus-TCP protocol is used for instruction CRC verification;

[0353] Failure retransmission mechanism: maximum number of retries 3, timeout 500ms.

[0354] 4. Equipment execution feedback monitoring

[0355] 4.1 Real-time reading of device status register:

[0356] Crusher actual speed register: ns=2; s=Equipment / Crusher / ActualSpeed;

[0357] Mixer temperature register: ns=2; s=Equipment / Mixer / Temperature.

[0358] 4.2 Perform deviation handling:

[0359] If the actual rotation speed deviates from the set value by more than 5%, alarm code AL101 will be triggered.

[0360] If the temperature deviation persists for more than 3 minutes, the cooling compensation program will be activated.

[0361] This step parses the multi-device collaborative scheduling sequence generated in S106 (e.g., the crusher processes batch B2024082201 from 08:00 to 10:30, at 85% speed) into specific executable instructions. Equipment control commands are generated using parameters such as the crusher speed setpoint (rated speed × 0.85) and feed rate (8 tons / hour). Equipment physical limit checks (crusher speed allowable range 50%-100%) and energy constraint checks are performed (a dynamic load reduction strategy is triggered when peak power is 367kW > 350kW, lowering the speed to 78.5%). Instructions are encapsulated using the OPC UA protocol and transmitted to the PLC controller (e.g., crusher node ns=2; s=Equipment / Crusher / SetSpeed). By real-time monitoring of the equipment status register (actual speed register ns=2; s=Equipment / Crusher / ActualSpeed), an alarm (code AL101) is triggered for execution deviations > 5%, ensuring accurate instruction execution. Implementation results: Command transmission success rate of 99.99%, device response latency of <100ms, and execution deviation rate controlled within 0.3%.

[0362] This embodiment provides a method for optimizing the collaborative scheduling of multiple devices in a feed production line. This method utilizes a real-time raw material market price acquisition interface to periodically collect spot prices and short-term trend forecasts of target raw materials. Simultaneously, it acquires equipment operating status, pending production batch attributes, inventory monitoring data, and pre-stored equipment collaborative constraints. A price deviation value is calculated based on the difference between the spot price and a preset benchmark price. An inventory adjustment threshold is generated by combining inventory monitoring and short-term trend forecast data. Dynamic scheduling priority adjustment rules are invoked based on the price deviation value and production batch attributes to generate a scheduling priority factor set and construct a conflict detection graph. A mapping table between raw material types and production batch attributes is constructed, and the price deviation value is converted into equipment scheduling influence factors using an influence factor transformation matrix. These factors, along with the inventory adjustment threshold and the scheduling priority factor set, are input into a conflict resolution-type initial scheduling rule base to generate dynamic response rules. A multi-device collaborative scheduling sequence is output based on the equipment operating status, generating control commands to drive equipment execution, achieving dynamic collaborative optimization and reducing costs.

[0363] Figure 2 This is a connection diagram of the multi-equipment collaborative scheduling optimization system for the feed production line provided in this application, as shown below. Figure 2 As shown, this is the multi-equipment collaborative scheduling and optimization system for a feed production line provided in this embodiment. This system applies... Figure 1 The feed production line multi-equipment collaborative scheduling optimization method described in the embodiment includes an optimization system comprising:

[0364] The data acquisition module is used to periodically collect spot price data and short-term trend forecast data of target raw materials through the real-time raw material market price acquisition interface, as well as current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships.

[0365] The deviation and threshold generation module is connected to the data acquisition module and is used to calculate the price deviation value based on the difference between the spot price data and the preset benchmark price data, and to generate an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data.

[0366] The priority and conflict graph construction module is connected to the data acquisition module and the deviation and threshold generation module. It is used to generate a set of scheduling priority factors by calling the dynamic scheduling priority adjustment rules according to the price deviation value and the attribute data of the production batch to be processed, and to construct a conflict detection graph based on the equipment collaborative constraint relationship.

[0367] The influence factor conversion module, connected to the deviation and threshold generation module, is used to construct a mapping table between raw material type and production batch attributes, and to convert the price deviation value into equipment scheduling influence factor through the influence factor conversion matrix.

[0368] The rule base processing module, connected to the influence factor conversion module, the deviation and threshold generation module, and the priority and conflict graph construction module, is used to input the set of equipment scheduling influence factors, inventory adjustment thresholds, and scheduling priority factors into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship. In the initial scheduling rule base, dynamic response rules are generated by processing the constraint relationship mapping of the conflict detection graph and calling the conflict resolution strategy library.

[0369] The scheduling sequence generation module is connected to the rule base processing module and the data acquisition module, and is used to combine the device operating status data and the dynamic response rules to output a multi-device collaborative scheduling sequence.

[0370] The instruction execution module, connected to the scheduling sequence generation module, is used to generate scheduling control instructions based on the multi-device collaborative scheduling sequence and send the scheduling control instructions to the corresponding production equipment for execution.

[0371] Specifically, the deviation and threshold generation module includes:

[0372] An absolute difference calculation unit is used to calculate the absolute difference between the spot price data and the preset benchmark price data;

[0373] A deviation rate generation unit, connected to the absolute difference calculation unit, is used to generate a relative price deviation rate based on the absolute difference through a price sensitivity coefficient calculation model.

[0374] The inventory adjustment unit, connected to the data acquisition module and the deviation rate generation unit, is used to combine the current inventory level in the inventory monitoring data with the short-term trend prediction data to generate an inventory adjustment threshold through a dynamic inventory adjustment algorithm.

[0375] Specifically, the sub-module for generating the scheduling priority factor set in the priority and conflict graph construction module includes:

[0376] The attribute parsing unit is used to parse the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed.

[0377] The weight calculation unit, connected to the attribute parsing unit and the deviation and threshold generation module, is used to calculate the raw material cost sensitivity weight based on the price deviation value through the priority weight allocation function.

[0378] The factor set generation unit, connected to the weight calculation unit, is used to generate a scheduling priority factor set containing batch priority weight factors based on the raw material cost sensitivity weight and the order urgency parameter. Example

[0379] 1. Data Acquisition Module

[0380] 1.1 Real-time raw material market price acquisition interface

[0381] It uses the official API connection of the Chicago Mercantile Exchange to periodically collect corn spot price data and soybean meal spot price data via HTTPS protocol, while also obtaining short-term price trend forecast data, including trend direction parameters and volatility index parameters.

[0382] 1.2 Acquisition of Equipment Operating Status Data

[0383] The system connects to a Siemens programmable logic controller via the OPC UA industrial communication protocol to read the motor current value of the crusher, the bearing temperature value of the mixer, and the vibration amplitude value of the pellet mill in real time, with a sampling frequency of once per second.

[0384] 1.3 Extraction of Attribute Data for Production Batches to be Processed

[0385] Execute SQL query commands from the Enterprise Resource Planning (ERP) system's MySQL database to parse the raw material composition field, obtain the corn quality percentage parameter and soybean meal quality percentage parameter, and calculate the order urgency parameter.

[0386] 1.4 Inventory Monitoring Data Collection

[0387] The inventory levels of corn and soybean meal were read using an ultra-high frequency radio frequency identification system, and the data was verified using Mettler Toledo weighing sensors.

[0388] 1.5 Loading of Equipment Coordination Constraints

[0389] Read XML-formatted constraint data from relational database tables, including material transfer constraints and energy coupling constraints.

[0390] 1.6 Connection method:

[0391] The raw material market price interface and data acquisition module are directly connected via an RJ45 network port.

[0392] The programmable logic controller and the data acquisition module are connected via industrial Ethernet.

[0393] Technical results: Achieves millisecond-level synchronous acquisition of multi-source data with a transmission error rate of less than 0.01%.

[0394] 2. Deviation and Threshold Generation Module

[0395] 2.1 Absolute Difference Calculation Unit

[0396] The system retrieves the preset benchmark price data from the enterprise cost database, calculates the absolute difference between the spot price of corn and the benchmark price, and calculates the absolute difference between the spot price of soybean meal and the benchmark price.

[0397] 2.2 Deviation Rate Generation Unit

[0398] The price sensitivity coefficients generated by the linear regression analysis model were used to calculate the relative price deviation rates of corn and soybean meal.

[0399] 2.3 Inventory Adjustment Unit

[0400] By combining current inventory levels and short-term trend forecasts, and through calculation of the safety stock base and adjustment of the trend correction factor, corn inventory adjustment thresholds and soybean meal inventory adjustment thresholds are generated.

[0401] 2.4 Connection method:

[0402] The absolute difference calculation unit and the deviation rate generation unit transfer data through a shared memory area;

[0403] The inventory adjustment unit and the data acquisition module communicate using the Modbus-TCP protocol.

[0404] Technical benefits: Price deviation calculation response time is 200 milliseconds, and inventory thresholds dynamically adapt to market fluctuations.

[0405] 3. Priority and Conflict Diagram Construction Module

[0406] 3.1 Attribute Parsing Unit

[0407] Analyze the corn quality percentage parameter and order urgency parameter in the production batch attributes.

[0408] 3.2 Weight Calculation Unit

[0409] The cost sensitivity weights for corn and soybean meal are calculated using a linear weighting function.

[0410] 3.3 Factor Set Generation Unit

[0411] Generate a set of scheduling priority factors that includes batch priority weight factors, and output the data in JSON format.

[0412] 3.4 Conflict Detection Graph Construction Unit

[0413] Analyze the physical connection relationships and energy coupling constraints of the devices, construct the device node topology network, and label the energy conflict edges and temporal conflict edges.

[0414] 3.5 Connection method:

[0415] The attribute parsing unit and the weight calculation unit exchange data through a POSIX message queue;

[0416] The weight calculation unit and the deviation and threshold generation module are connected via industrial Ethernet.

[0417] Technical results: Priority factor generation and conflict detection are completed simultaneously, with an accuracy rate of 99.8%.

[0418] 4. Impact Factor Conversion Module

[0419] 4.1 Construction of Raw Material-Batch Weight Mapping Table

[0420] Based on Pearson correlation coefficient analysis of historical production data, the weight value of corn in fattening feed batches is generated.

[0421] 4.2 Calculation of Raw Material Price Influence Factors

[0422] The impact factors of corn price and soybean meal price were calculated using the impact factor transformation matrix.

[0423] 4.3 Generation of Equipment Scheduling Influencing Factors

[0424] Based on the real-time load rate parameters of the crusher, a linear interpolation formula is used to generate the equipment scheduling influence factor.

[0425] 4.4 Connection method:

[0426] The deviation and threshold generation module transmits the relative price deviation rate through a shared memory address 0x5000;

[0427] It communicates with the rule base processing module using the OPC UA protocol.

[0428] Technical effect: Achieve precise quantitative conversion of market price fluctuations into equipment parameters.

[0429] 5. Rule base processing module

[0430] 5.1 Initial Scheduling Rule Base Construction

[0431] By calling up equipment failure rate statistics and task processing time distribution data, and combining them with the equipment's maximum continuous working time limit, the Drools rule engine is used to define equipment maintenance rules.

[0432] 5.2 Dynamic Response Rule Generation

[0433] The conflict risk edge is mapped to a rule base identifier, the device task reallocation strategy is invoked, and a response rule containing the speed adjustment command is generated.

[0434] 5.3 Connection method:

[0435] The impact factor conversion module transmits JSON format data via fiber optic cable.

[0436] The priority and conflict graph construction module is connected via a PCIe bus.

[0437] Technical results: Rule generation response time is 100 milliseconds, and conflict resolution rate is 99.5%.

[0438] 6. Scheduling sequence generation module

[0439] 6.1 Multi-device collaborative scheduling sequence output

[0440] By combining real-time load rate parameters and dynamic response rules of the equipment, an improved Hungarian algorithm is used to solve the equipment task allocation sequence.

[0441] 6.2 Connection method:

[0442] Dynamic response rules are passed to the rule base processing module through a shared memory area;

[0443] It is connected to the data acquisition module via a time-sensitive network.

[0444] Technical results: The scheduling sequence generation time is 150 milliseconds, and the equipment utilization rate is increased to 92.3%.

[0445] 7. Instruction Execution Module

[0446] 7.1 Generation of Dispatch Control Instructions

[0447] The scheduling sequence is parsed to generate the crusher speed setting command and the feed rate command.

[0448] 7.2 Command Transmission and Execution

[0449] The address of the programmable logic controller register is written via the PROFINET industrial bus, and the CRC-16 check mechanism is used to ensure the integrity of the transmission.

[0450] 7.3 Connection method:

[0451] It is connected to the scheduling sequence generation module via an industrial Ethernet network;

[0452] It is directly connected to the production equipment via twisted-pair cable.

[0453] Technical results: Command transmission success rate of 99.99%, execution deviation rate of 0.3%.

[0454] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0455] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for optimizing the collaborative scheduling of multiple devices in a feed production line, characterized in that, The method includes: By using the real-time raw material market price acquisition interface, spot price data and short-term trend forecast data of target raw materials are periodically collected, and current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships are also obtained. Based on the difference between the spot price data and the preset benchmark price data, a price deviation value is calculated, and an inventory adjustment threshold is generated by combining the inventory monitoring data and the short-term trend forecast data. Specifically, calculating the price deviation value based on the difference between the spot price data and the preset benchmark price data, and generating the inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data, includes: calculating the absolute difference between the spot price data and the preset benchmark price data; generating a relative price deviation rate based on the absolute difference using a price sensitivity coefficient calculation model; and generating an inventory adjustment threshold by combining the current inventory level in the inventory monitoring data and the short-term trend forecast data using a dynamic inventory adjustment algorithm. Based on the price deviation value and the attribute data of the production batch to be processed, the dynamic scheduling priority adjustment rule is invoked to generate a scheduling priority factor set, and a conflict detection graph is constructed based on the equipment coordination constraint relationship. Specifically, constructing the conflict detection graph based on the equipment coordination constraint relationship includes: parsing the physical connection relationship and energy coupling constraint of the equipment nodes in the equipment coordination constraint relationship; constructing a topology network of equipment nodes based on the physical connection relationship; and marking conflict risk edges in the topology network of equipment nodes according to the energy coupling constraint to generate the conflict detection graph. A mapping table between raw material types and production batch attributes is constructed. The price deviation value is then transformed into an equipment scheduling impact factor using an influence factor transformation matrix. Specifically, constructing this mapping table and transforming the price deviation value into an equipment scheduling impact factor involves: constructing a raw material-batch weight mapping table based on historical correlation data between raw material types and production batch attributes; calculating the raw material price impact factor using the raw material price sensitivity coefficient in the influence factor transformation matrix, based on the raw material proportion weight in the raw material-batch weight mapping table and the price deviation value; and generating an equipment scheduling impact factor based on the raw material price impact factor and the equipment load adjustment coefficient. The set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors are input into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationships. Dynamic response rules are generated within this initial scheduling rule base through constraint relationship mapping of the conflict detection graph and invocation of the conflict resolution strategy library. Specifically, generating dynamic response rules within the initial scheduling rule base through constraint relationship mapping of the conflict detection graph and invocation of the conflict resolution strategy library includes: mapping conflict risk edges in the conflict detection graph to constraint conflict identifiers recognizable by the rule base; invoking equipment task reallocation strategies from the conflict resolution strategy library based on the constraint conflict identifiers; and generating dynamic response rules based on the equipment task reallocation strategies and the input set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors. By combining the device operating status data with the dynamic response rules, a multi-device collaborative scheduling sequence is output; Based on the multi-device collaborative scheduling sequence, a scheduling control instruction is generated and sent to the corresponding production equipment for execution.

2. The multi-equipment collaborative scheduling optimization method for a feed production line according to claim 1, characterized in that, The step of invoking dynamic scheduling priority adjustment rules based on the price deviation value and the attribute data of the production batch to be processed, and generating a set of scheduling priority factors, includes: Analyze the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed; Based on the price deviation value, the raw material cost sensitivity weight is calculated using a priority weight allocation function; Based on the raw material cost sensitivity weight and the order urgency parameter, a set of scheduling priority factors, including batch priority weight factors, is generated.

3. The multi-equipment collaborative scheduling optimization method for a feed production line according to claim 1, characterized in that, The step of inputting the set of equipment scheduling influencing factors, inventory adjustment thresholds, and scheduling priority factors into a conflict-resolving initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship includes: Retrieve device failure rate statistics and task processing time distribution data from pre-stored historical scheduling data; Based on the maximum continuous working time limit of the devices in the device coordination constraint relationship, an initial scheduling rule base is constructed; Input the set of equipment scheduling influencing factors, inventory adjustment thresholds and scheduling priority factors into the initial scheduling rule base.

4. A multi-equipment collaborative scheduling and optimization system for a feed production line, characterized in that, The optimization system applies the optimization method according to any one of claims 1-3, and the optimization system comprises: The data acquisition module is used to periodically collect spot price data and short-term trend forecast data of target raw materials through the real-time raw material market price acquisition interface, as well as current equipment operating status data, pending production batch attribute data, inventory monitoring data, and pre-stored equipment collaborative constraint relationships. The deviation and threshold generation module is connected to the data acquisition module and is used to calculate the price deviation value based on the difference between the spot price data and the preset benchmark price data, and to generate an inventory adjustment threshold by combining the inventory monitoring data and the short-term trend forecast data. The priority and conflict graph construction module is connected to the data acquisition module and the deviation and threshold generation module. It is used to generate a set of scheduling priority factors by calling the dynamic scheduling priority adjustment rules according to the price deviation value and the attribute data of the production batch to be processed, and to construct a conflict detection graph based on the equipment collaborative constraint relationship. The influence factor conversion module, connected to the deviation and threshold generation module, is used to construct a mapping table between raw material type and production batch attributes, and to convert the price deviation value into equipment scheduling influence factor through the influence factor conversion matrix. The rule base processing module, connected to the influence factor conversion module, the deviation and threshold generation module, and the priority and conflict graph construction module, is used to input the set of equipment scheduling influence factors, inventory adjustment thresholds, and scheduling priority factors into a conflict resolution-type initial scheduling rule base constructed based on pre-stored historical scheduling data and the equipment collaborative constraint relationship. In the initial scheduling rule base, dynamic response rules are generated by processing the constraint relationship mapping of the conflict detection graph and calling the conflict resolution strategy library. The scheduling sequence generation module is connected to the rule base processing module and the data acquisition module, and is used to combine the device operating status data and the dynamic response rules to output a multi-device collaborative scheduling sequence. The instruction execution module, connected to the scheduling sequence generation module, is used to generate scheduling control instructions based on the multi-device collaborative scheduling sequence and send the scheduling control instructions to the corresponding production equipment for execution.

5. The multi-equipment collaborative scheduling and optimization system for a feed production line according to claim 4, characterized in that, The deviation and threshold generation module includes: An absolute difference calculation unit is used to calculate the absolute difference between the spot price data and the preset benchmark price data; A deviation rate generation unit, connected to the absolute difference calculation unit, is used to generate a relative price deviation rate based on the absolute difference through a price sensitivity coefficient calculation model. The inventory adjustment unit, connected to the data acquisition module and the deviation rate generation unit, is used to combine the current inventory level in the inventory monitoring data with the short-term trend prediction data to generate an inventory adjustment threshold through a dynamic inventory adjustment algorithm.

6. The multi-equipment collaborative scheduling and optimization system for a feed production line according to claim 4, characterized in that, The priority and conflict graph construction module includes: The attribute parsing unit is used to parse the raw material type ratio and order urgency parameter in the attribute data of the production batch to be processed. The weight calculation unit, connected to the attribute parsing unit and the deviation and threshold generation module, is used to calculate the raw material cost sensitivity weight based on the price deviation value through the priority weight allocation function. The factor set generation unit, connected to the weight calculation unit, is used to generate a scheduling priority factor set containing batch priority weight factors based on the raw material cost sensitivity weight and the order urgency parameter.

Images