A microcomputer automatic control system for an internal mixer auxiliary machine

Abstract

The application provides a microcomputer automatic control system of an upper auxiliary machine of a mixer, and relates to the technical field of material pretreatment. The system adopts a two-stage hierarchical control architecture of an upper management computer and a lower PLC substation cabinet, the upper management computer is provided with management software and debugging software, the lower PLC substation cabinet is provided with bottom layer control software, and the upper and lower computers and each substation cabinet are connected through a field bus to realize bidirectional data interaction. The system comprises a formula dynamic management engine, a material pretreatment intelligent weighing control module, a device state real-time sensing module, an optimized feeding decision engine and a linkage control instruction issuing module. Through a visual formula configuration unit, a material parameter mapping unit and a self-adaptive process rule engine of the formula dynamic management engine, the system realizes standardized modeling of a material pretreatment formula, dynamic mapping of material properties and system labels, and automatic analysis and legality verification of process rules.

Description

Technical Field

[0001] This invention relates to the field of material pretreatment technology, specifically to a microcomputer automatic control system for auxiliary equipment on an internal mixer. Background Technology

[0002] In rubber mixing production, the development of automatic control systems for auxiliary equipment has evolved from early relay logic control and stand-alone PLC control to centralized configuration monitoring. Early systems relied primarily on operators manually setting weighing parameters based on paper formulas, controlling vacuum feeding and weighing valve operation via buttons or simple switches, and judging the timing and sequence of material feeding based on experience. Subsequent PLC control systems implemented basic sequential control and weighing interlocks, while the host computer configuration software displayed equipment operating status and simple reports, improving the level of production automation to some extent.

[0003] Although existing technologies can achieve automatic control of auxiliary equipment in internal mixing, the following problems still exist in actual use: Existing technologies mainly rely on manual input or paper document import for formula management, lacking visual dynamic configuration, automatic mapping of material attributes and system tags, and adaptive parsing and legality verification of process rules. This results in inconsistent formula parameter formats, conflicting process logic, and large fluctuations in the quality of different batches of production. Existing technologies often employ fixed modes or simple on / off control during material pretreatment weighing, failing to achieve intelligent segmented weighing that combines rapid feeding, slow replenishment, and micro-inching for different material characteristics. Furthermore, the lack of real-time fusion perception of the equipment's native signals and weighing status results in insufficient weighing accuracy and easy disruption of process sequence. Existing technologies mainly rely on preset fixed timing sequences or manual judgment in the material feeding decision and execution stages. They lack multi-constraint optimization matching based on real-time equipment state vectors, and the control process is mostly open-loop execution. It is impossible to perform closed-loop verification of feedback signals such as valve positioning and weighing deviation, resulting in improper material feeding timing and delayed anomaly detection.

[0004] To address the problems of the existing technologies, there is an urgent need for a microcomputer-based automatic control system for auxiliary equipment in internal mixers. This system should utilize a dynamic formula management engine to enable visualized configuration of formulas, dynamic mapping of material parameters, and adaptive process rule parsing. A material pretreatment intelligent weighing control module should enable intelligent segmented weighing and status generation of different materials. A real-time equipment status sensing module should achieve fusion sensing of native signals and weighing data. An optimized feeding decision engine should optimize feeding schemes under multiple constraints. A linkage control command issuance module should enable closed-loop execution of commands and anomaly feedback. A report management module should collect, statistically analyze, and provide historical feedback on operational data. This will improve formula compliance, weighing accuracy, feeding decision accuracy, and the overall closed-loop control capability of the system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a microcomputer-based automatic control system for auxiliary equipment on a mixer, which solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a microcomputer-based automatic control system for auxiliary equipment of a mixing mill. The system adopts a two-level hierarchical control architecture consisting of a host computer and lower-level PLC substations. The host computer carries management and debugging software, while the lower-level PLC substations carry underlying control software. The host computer, host computer, and each substation are connected via a fieldbus for bidirectional data interaction. The system includes a dynamic formula management engine, a material pretreatment intelligent weighing control module, a real-time equipment status sensing module, an optimized feeding decision engine, and a linkage control command issuance module, wherein: The dynamic formula management engine is deployed on the host computer and includes a visual formula configuration unit, a material parameter mapping unit, and an adaptive process rule engine. The visual formula configuration unit acquires the material pretreatment formula data uploaded by the user in real time and generates a standardized formula model. The material parameter mapping unit establishes a dynamic mapping relationship between material attributes and system tags. The adaptive process rule engine dynamically parses the standardized formula model, generates a standardized pretreatment instruction set, and transmits it to the intelligent weighing control module for material pretreatment. The intelligent weighing control module for material pretreatment is deployed in the lower-level PLC substation cabinet. It receives the standardized pretreatment instruction set to control the fast and slow jog weighing of carbon black, powder and oil, as well as the weighing process of rubber belt scale materials. It also generates weighing completion status data and transmits it to the real-time equipment status sensing module. The real-time equipment status sensing module is deployed in the lower-level PLC substation cabinet. It receives the weighing completion status data and collects the equipment operation status data by listening to the native signals. It then merges these data to generate a unified equipment status vector and transmits it to the optimized feeding decision engine. The optimized feeding decision engine operates collaboratively across upper and lower levels, receives the unified equipment state vector and reads the material pretreatment formula data in the standardized formula model, then generates an optimized feeding scheme and transmits it to the linkage control command issuing module. The linkage control command sending module is deployed in the lower-level PLC substation cabinet, receives the optimized feeding scheme, generates specific control commands and sends them to the actuator, and at the same time feeds back the execution results to the real-time equipment status sensing module. Each module works collaboratively through data interaction, relying on a two-level hierarchical architecture to achieve formula management, weighing control, real-time dynamic monitoring, feeding optimization, and closed-loop linkage control with the internal mixer for material pretreatment.

[0007] Preferably, the specific workflow of the recipe dynamic management engine is as follows: SPA1, the visual formula configuration unit first builds a menu interface, which is a Chinese drop-down menu interface. Then, it obtains the material pretreatment formula data uploaded by the user through the menu interface. The material pretreatment formula data includes raw material ratio data, error range and process steps, and then generates a standardized formula model. SPA2, the material parameter mapping unit establishes a dynamic mapping relationship between the physical properties of materials and the virtual labels of the system through the generated standardized formula model; SPA3, the adaptive process rule engine dynamically parses the material pretreatment formula data in the standardized formula model, and simultaneously performs automatic legality verification and logical matching of multiple process parameters; SPA4. The adaptive process rule engine generates a standardized preprocessing instruction set based on the parsing results and transmits it to the intelligent weighing control module for material preprocessing.

[0008] Preferably, the intelligent weighing control module for material pretreatment includes a vacuum adsorption weighing unit and a multi-segment feeding control unit. The two units work in tandem to perform material pretreatment and generate weighing completion status data. Wherein: The vacuum adsorption weighing unit is used to control the vacuum adsorption device to transport materials, and adopts three methods—fast feeding, slow replenishment, and micro-inching—combined with an automatic correction function to complete the dynamic weighing of carbon black powder and oil. The multi-segment feeding control unit receives the standardized preprocessing instruction set, controls the feeding device to feed the material into the internal mixer, and generates weighing completion status data, which is then transmitted to the equipment status real-time sensing module.

[0009] Preferably, the real-time device status sensing module includes a native signal monitoring unit and a status fusion processing unit, wherein: The native signal monitoring unit is used to monitor the PLC native communication signals, the internal mixer switch signals, and temperature and power analog signals, thereby collecting equipment operating status data. The state fusion processing unit fuses the operating state data and weighing completion state data to generate a unified equipment state vector, which is then transmitted to the optimized feeding decision engine.

[0010] Preferably, the optimized feeding decision engine includes a process parameter evaluation unit and a multi-constraint optimization matching unit, wherein: The process parameter evaluation unit is used to calculate the linkage logic relationship and the internal mixing feeding trigger condition based on the material pretreatment formula data in the standardized formula model. The multi-constraint optimization matching unit generates an optimized feeding scheme based on the logical relationship, the mixing feeding trigger condition, and the unified equipment state vector, and then transmits it to the linkage control command issuing module.

[0011] Preferably, the linkage control command issuing module includes a command generation unit and a closed-loop execution monitoring unit, wherein: The instruction generation unit is used to convert the optimized feeding scheme into a standardized control instruction sequence and issue it. The closed-loop execution monitoring unit is used to collect execution feedback signals in real time and compare them with the expected state.

[0012] Preferably, the specific steps for the closed-loop execution monitoring unit to implement closed-loop command control include: SPB1, a standardized control instruction sequence output by the instruction generation unit; SPB2: Sends control commands to the conveying device, weighing device and feeding device via fieldbus; SPB3: Real-time acquisition and verification of valve position signals, weighing values, and internal mixer feed gate status; SPB4 triggers an alarm signal and displays the alarm location when it detects weighing errors, timeouts, zero-point deviations, or valve misalignment, and feeds back the execution results to the real-time equipment status sensing module.

[0013] Preferably, the system also includes a report management module that interacts with the formula dynamic management engine. The report management module is used to collect the operating data of each module, generate shift reports, curve reports, alarm records and material statistics reports, and supports data query and printing. The data can be stored for three years.

[0014] Preferably, the report management module works in conjunction with the real-time equipment status sensing module to achieve remote monitoring through a network interface and feed historical data back to the formula dynamic management engine, supporting formula modification, production plan adjustment and hierarchical management of user passwords.

[0015] This invention provides a microcomputer-based automatic control system for auxiliary equipment on a Banbury mixer. It has the following beneficial effects: 1. This invention, through the visual formula configuration unit, material parameter mapping unit, and adaptive process rule engine of the formula dynamic management engine, realizes standardized modeling of material pretreatment formulas, dynamic mapping of material attributes and system tags, and automatic parsing and legality verification of process rules. It effectively solves the problems of inconsistent formula parameter formats and easy conflict of process logic in the prior art, and reduces the fluctuation of production quality in different batches.

[0016] 2. This invention achieves intelligent segmented weighing that combines rapid feeding, slow replenishment, and micro-inching for different material characteristics through the vacuum adsorption weighing unit and multi-segment feeding control unit of the intelligent weighing control module for material pretreatment, as well as the native signal monitoring unit and status fusion processing unit of the real-time equipment status sensing module. It also integrates the native signals of the equipment with the weighing status in real time, effectively solving the problems of insufficient weighing accuracy and easy disorder of process sequence in the prior art, and improving the accuracy and stability of the pretreatment process.

[0017] 3. This invention optimizes the process parameter evaluation unit and multi-constraint optimization matching unit of the feeding decision engine, as well as the instruction generation unit and closed-loop execution monitoring unit of the linkage control instruction issuance module. This enables the optimization of multi-constraint feeding schemes based on a unified equipment state vector and the closed-loop verification of instruction execution. It effectively solves the problems of improper feeding timing and delayed anomaly detection in the prior art, and improves the accuracy of feeding decisions and the reliability of overall system control. Attached Figure Description

[0018] Figure 1 This is a system framework diagram of the present invention; Figure 2 This is a flowchart of the recipe dynamic management engine of the present invention; Figure 3 This is a flowchart of the instruction closed-loop control in this invention; Figure 4 This is a schematic diagram of the front-end workbench interface of the system in this invention; Figure 5 This is a schematic diagram of the working monitoring interface of the vacuum adsorption weighing unit in the front end of the system of the present invention; Figure 6 This is a schematic diagram of the working monitoring interface of the multi-segment feeding control unit in the front end of the system of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation Example 1: A microcomputer-based automatic control system for auxiliary equipment of a mixing mill is disclosed. The system adopts a two-tiered hierarchical control architecture consisting of a host computer and lower-level PLC substations. The system is divided into an upper-level information management layer and a lower-level field equipment control layer, with clear division of labor and separation of control. The host computer, as the top-level management terminal, is equipped with industrially common configuration management software and equipment-specific debugging software. Specifically, mature industrial configuration software such as Siemens WinCC and KingSCADA can be used as the production management software, along with the PLC's STEP7 debugging software. It primarily handles upper-level tasks such as formula configuration, production plan editing, process parameter management, data statistics and traceability, remote monitoring, and system access control. It can complete production parameter input, formula verification and modification, shift plan preparation, and operational status monitoring. The lower-level PLC substation cabinet, as the underlying field execution terminal, is equipped with industrial-grade low-level control software. It possesses highly reliable and real-time field computing and logic execution capabilities, directly connecting to various field execution devices and sensing elements to complete signal acquisition, logic calculation, equipment start / stop control, and fault diagnosis in real time. The upper and lower level computers and each substation cabinet achieve bidirectional data interaction via a fieldbus, constructing a stable, high-speed, and interference-resistant distributed network data link. This enables the downward transmission of management commands from the upper level computer and the real-time uploading of field operating data, equipment status, and weighing data from the lower level computer, ensuring data synchronization, command interoperability, and collaborative operation between the upper and lower levels.

[0021] like Figure 1 As shown, a system framework diagram is provided. This system includes a dynamic formula management engine, a material pretreatment intelligent weighing control module, a real-time equipment status sensing module, an optimized feeding decision engine, and a linkage control command issuance module. Each module works collaboratively through data interaction, relying on a two-level hierarchical architecture to realize a progressive control process of "formula analysis and pretreatment → intelligent weighing and material preparation → equipment status sensing → feeding strategy optimization → closed-loop command execution".

[0022] The entire system consists of five core modules and one auxiliary module, such as Figure 4 As shown, a schematic diagram of the front-end workbench interface of the system in this invention is provided. Users can click on the corresponding workbench to view the progress of the six modules or make corresponding adjustments, as detailed below: (1) The first is the formula dynamic management engine. This engine is deployed in the configuration software environment of the host computer and relies on the dedicated hardware and software resources of the host computer to complete its operation. It relies on the large-capacity solid-state hard disk of the host computer to complete the formula data storage, relies on the CPU of the host computer to complete the process parameter calculation and analysis, relies on the configuration screen component of the host computer to complete the visual human-machine interaction, and relies on the bus communication port of the host computer to complete the data interaction with the lower PLC. This engine runs in the background of the host computer throughout the process and is only responsible for formula data management, process logic analysis, and preprocessing instruction generation. It does not participate in the hard logic control of the underlying equipment, separates the management calculation and the real-time control tasks on site, and avoids the problem of equipment lag and control delay caused by the occupation of the lower PLC's computing resources. The engine consists of three functional units that work together: a visual formula configuration unit, a material parameter mapping unit, and an adaptive process rule engine. These three units work in a progressive manner: the visual formula configuration unit first completes the manual formula input and standardized modeling, then the material parameter mapping unit completes the data standardization and adaptation, and finally the adaptive process rule engine completes the verification, parsing, and instruction generation, thus completing the entire set of upper-level formula preprocessing work step by step.

[0023] The visual formula configuration unit acquires the material preprocessing formula data uploaded by the user in real time and generates a standardized formula model. It should be noted that the visual formula configuration unit is the core of the system's human-computer interaction configuration. It relies on the interface editing component of the host computer configuration software to provide a visual operation carrier and supports users to independently input, modify, and retrieve production formula data. The material parameter mapping unit establishes a dynamic mapping relationship between material attributes and system tags; The adaptive process rule engine dynamically analyzes the standardized formula model, generates a standardized preprocessing instruction set, and transmits it to the intelligent weighing control module for material preprocessing.

[0024] like Figure 2 As shown, a flowchart of the recipe dynamic management engine is provided, and the specific workflow of the recipe dynamic management engine is as follows: First, a menu interface is built using the visual formula configuration unit. The menu interface is a Chinese drop-down menu interface. Then, the user-uploaded material pretreatment formula data is obtained through the menu interface. The material pretreatment formula data includes raw material ratio data, error range and process steps, and then a standardized formula model is generated. The logic and implementation of the menu interface in this step are as follows: The system is based on a configuration software development platform installed on a host computer. It calls the platform's built-in static text controls, numerical input controls, drop-down selection controls, data saving controls, and formula retrieval controls. Based on the fixed process requirements of the pre-treatment of the mixing materials, a custom partitioned functional interface is built. The interface strictly distinguishes between material parameter input areas, error threshold setting areas, process step selection areas, and formula saving and export areas. After the interface is built, the configuration software's internal variable binding function binds each input and selection control on the interface to the system's internal formula data cache variables. This ensures that interface operations are written to the system cache in real time, guaranteeing that every manually entered parameter is received and stored by the system in real time.

[0025] Users upload and enter formula data through this Chinese drop-down menu interface. The material pretreatment formula data collected by the system serves as the basis for all pre-processing steps in this internal mixing pretreatment production, specifically including three types of implementable production parameters: 1. Raw material proportioning data: This specifically refers to the target weighing weights of carbon black, powder, oil, and rubber, as well as the proportions between these materials. It serves as the core quantitative operating basis for weighing equipment. For example, in actual production, a certain batch of rubber product formulation might have a fixed weight ratio for carbon black (25.0 kg), processing oil (3.5 kg), and virgin rubber (50.0 kg). This fixed weight ratio directly determines the basic formulation proportions of the compound rubber. 2. Error Range Data: This specifically refers to the maximum positive and negative deviations allowed in the weighing process for each material, while also clearly defining the criteria and processing logic for materials exceeding these tolerances. For example, the system presets an allowable weighing error of ±0.2 kg for carbon black. Weighing results within the range of 24.8 kg to 25.2 kg are considered acceptable materials. Weighing results below 24.8 kg are considered insufficient, and the system automatically triggers a replenishment logic. Weighing results above 25.2 kg are considered out-of-tolerance waste materials, which are prohibited from entering the pretreatment process. 3. Process Step Data: Specifically refers to the order of material weighing, the segmented method of rapid feeding / slow feeding / jogging fine adjustment for a single material, the temporary storage sequence after pretreatment, and the pre-feeding constraints. For example, the actual process stipulates that carbon black powder must be weighed first, followed by liquid oil. The powder adopts a segmented feeding method combining fast, slow, and jogging, while the oil adopts only a micro-jogging feeding method. Only after all materials have been weighed and the equipment is stable can the material be temporarily stored and await the mixing and feeding instruction.

[0026] The purpose of obtaining the above complete data in this step is to replace the traditional manual paper formula and manual experience-based process control mode, quantify all pretreatment process requirements into numerical parameters that the system can recognize, and ensure that the entire process of subsequent weighing, analysis, and feeding is based on data, avoiding batch production quality fluctuations caused by differences in manual operation.

[0027] After collecting complete original formula data, the visual formula configuration unit generates a standardized formula model according to fixed steps: ① Data collection and classification: The system comprehensively scans all manually entered parameters on the interface, intelligently identifies and removes blank, invalid, and duplicate parameter entries, and automatically classifies and collects the remaining valid production parameters into three categories: proportion parameters, error parameters, and process parameters, forming three independent arrays of raw parameters, thus completing the initial sorting of messy manual data; ② Data unification and standardization: Standardize and unify all parameters after collection, unify the measurement unit, decimal precision, and data storage format of all parameters. For example, unify the weight unit of all materials to kilograms, unify the retention of one decimal place for values, and unify the coding format of process steps, so as to completely solve the system recognition abnormality problems caused by inconsistent units, messy decimal places, and non-standard formats during manual data entry. ③ Structured encapsulation modeling: According to the standard formula data structure template preset by the system, the standardized proportioning parameters, error threshold parameters, and process step parameters are accurately filled into the fixed field positions of the template to complete the structured binding and encapsulation of all parameters. Finally, a standardized formula model with complete fields, uniform format, fixed structure, and can be freely retrieved, recognized, and calculated by all functional modules of the system is generated and automatically temporarily stored in the host computer data cache area, providing a unique standard data source for subsequent parameter mapping and process analysis.

[0028] In actual rubber mixing production, operators input the following process requirements into the host computer interface for a specific rubber compound: carbon black weighing 25kg, oil weighing 3.5kg, a weighing error tolerance of ±0.2kg, and weighing carbon black first, followed by oil. The system automatically organizes and categorizes the manually entered values, processes, and error requirements, standardizes the units of measurement and data accuracy, and encapsulates them into a standardized formula model with a fixed structure. This avoids the problem of inconsistent manual parameter formats that prevent the system from recognizing the data, and provides a unified standard data template for subsequent automated production.

[0029] This step completes the transformation from manual process input to a structured data model of the system, solving the problems of scattered, unstandardized, and unrecognizable traditional formula parameters. The generated standardized formula model provides a unique and standardized data carrier for subsequent parameter mapping and process analysis, and is the data source for the entire formula engine.

[0030] Next, the material parameter mapping unit establishes a dynamic mapping relationship between the physical properties of materials and the virtual labels of the system through the generated standardized formula model; the specific execution steps are as follows: ① Parameter Extraction: The material parameter mapping unit reads the standardized formula model and accurately extracts the physical properties corresponding to each type of material in the model, including four core attributes: material type, material form, weighing characteristics, and feeding control characteristics, all with clearly defined meanings. Material types: Accurately distinguishes specific material categories such as carbon black, inorganic powders, processing oils, virgin rubber, and auxiliary materials for basic material classification and identification; Material form: Distinguishing between powder, liquid, and solid material forms is the basis for matching the feeding method to the system; Weighing characteristics: Based on the physical defects of different materials, powdery materials are characterized by easy dust generation, easy damage, and easy material accumulation on the weighing body; liquid oils are characterized by easy adhesion to pipe walls, large residue, and weighing lag; and solid rubber materials are characterized by stable weighing and no loss. Feeding control characteristics: Customized control characteristics are matched according to the material form. Powder is adapted to segmented fast and slow feeding characteristics, oil is adapted to micro-inching compensation feeding characteristics, and solid materials are adapted to uniform and stable feeding characteristics. ② Retrieve the system's virtual tag library: The system pre-stores a complete set of uniquely corresponding virtual data tags. Each tag corresponds to a material, a type of equipment, and a parameter, and serves as the unique identifier for data interaction, identification, and calculation among various modules within the system. ③ Dynamic binding mapping: Following the principle of "one item, one tag", the extracted physical properties of materials are automatically bound to the corresponding virtual tags of the system to establish a fixed mapping relationship; for example, the powder form, dust loss characteristics, and segmented feeding characteristics of carbon black are all bound to the exclusive tag Tag_Carbon; the liquid form, pipe wall residue characteristics, and jogging compensation characteristics of oil are all bound to the exclusive tag Tag_Oil. ④ Adaptive Update: If the formula material types, proportion parameters, or process steps are subsequently manually modified, the system will automatically unbind the original mapping relationship, re-extract the new material attributes, match the corresponding virtual tags, and complete the new binding, realizing the dynamic adaptive update of the mapping relationship without the need for manual modification of parameter configuration.

[0031] In actual production scenarios, carbon black is a powdery material that easily generates dust, oil is a liquid material that easily adheres to the walls, and rubber is a solid material that needs to be transported. The weighing and feeding characteristics of these different materials are completely different. The system uses dynamic mapping to bind the "carbon black material attributes" to a built-in virtual tag specific to carbon black, and the "oil material attributes" to a virtual tag specific to oil. When the system subsequently recognizes this tag, it can automatically match the feeding logic for powdery materials and the correction logic for liquid materials, preventing the system from confusing the control logic for different materials and achieving targeted adaptive control for different materials.

[0032] This step establishes a precise correspondence between real-world physical materials and virtual data in the system, resolving issues such as the system's inability to autonomously identify the characteristics of physical materials, data interaction errors between different modules, and parameter mismatches. This enables the system to accurately identify the actual materials and processes corresponding to each formula parameter, laying a data matching foundation for the subsequent accurate analysis of process rules.

[0033] Subsequently, the adaptive process rule engine dynamically parses the material pretreatment formula data in the standardized formula model and simultaneously performs automatic validity verification and logical matching of multiple process parameters.

[0034] The specific steps for dynamic parsing are as follows: ① Parameter splitting and classification: Traverse all fields of the standardized formula model and automatically split them into three independent parameter groups, specifically including the weighing parameter group, error constraint group, and process sequence group; Weighing parameter group: includes quantitative control parameters such as target weighing rated weight for various materials, rapid feeding switching weight threshold, slow feeding weight range, and end-point jog fine-tuning weight range; Error constraint group: includes precision control parameters such as the maximum allowable positive error, maximum negative error, out-of-tolerance judgment conditions, material shortage replenishment trigger conditions, and out-of-tolerance alarm trigger conditions for each material; Process sequence group: includes timing control parameters such as the order of weighing multiple materials, the switching conditions for single material feeding mode, the start and stop constraints of the process, the judgment conditions for switching between preceding and following processes, and the admission conditions for weighing completion.

[0035] ② Call the built-in process rule library: The system has built-in solidified general process rules for internal mixing pretreatment. These rules are publicly and universally used pretreatment equipment control standards in the rubber internal mixing industry. They are solidified in the system software program in advance and do not require manual writing or modification. Specifically, they include: material weighing priority rules, segmented feeding control rules, weighing error judgment rules, process sequence constraint rules, and parameter upper and lower limit legality rules. Material weighing priority rules: Industry-standard pretreatment guidelines, powders and auxiliary materials are weighed first, and liquid oils are weighed later to avoid oil adhering to powders and causing uneven mixing; Segmented feeding control rules: The feeding mode is switched according to the material weight margin. When the weight margin is sufficient, the material is fed quickly, and when it is close to the rated value, the material is fed slowly. The end micro-adjustment is made to ensure weighing accuracy. Weighing error judgment rules: Based on the preset error threshold, the weighing result is judged into three states: qualified, insufficient material, and out of tolerance, and corresponding processing logics of replenishing material, alarm, and scrapping are matched accordingly. Process sequence constraints: If the previous material weighing process is not completed or the equipment is not stably zeroed, the next process shall not be started to prevent process disorder and material mixing. Legal rules for upper and lower limits of parameters: All weighing weights and error values ​​must be within the rated working range of the equipment and the industry's allowable accuracy range. Operation beyond the limit parameters is strictly prohibited.

[0036] ③ Line-by-line logic matching and parsing: The separated weighing parameter group, error constraint group, and process sequence group are traversed, compared, and precisely matched with the five built-in process rules one by one. Combined with the current formula parameters, a complete set of exclusive and implementable process logic is parsed out. The specific matching and parsing process and output results are detailed as follows: The system first matches the weighing parameters with the segmented feeding control rules. Based on the rated weight of 25kg carbon black and 3.5kg oil, it determines that the carbon black is suitable for the "fast feeding + slow replenishment" mode, and the oil is suitable for the "micro-volume inching feeding" mode throughout the process. Then match the error constraint group parameters with the error judgment rules and lock the qualified judgment, short material replenishment, and out-of-tolerance alarm logic corresponding to the ±0.2kg error threshold. Finally, the process timing group parameters are matched with the process priority and timing constraint rules to lock the fixed process sequence of "carbon black first, oilseed" and the process switching conditions. Finally, the precise analysis outputs the process logic that can be directly executed by four types of equipment: Material feeding logic: Powdered carbon black is fed at a high flow rate and then fed at a slow speed when it approaches the rated value; liquid oil is fed at a low flow rate throughout the process, with a matching pipe wall residue compensation logic. Weighing switching logic: When the carbon black weighing weight reaches 90% of the rated value, slow feeding is started; when it reaches 98%, the end-point jogging fine adjustment is started to prevent overshoot and exceeding tolerance. Process execution sequence logic: The oil weighing process can only be unlocked after the carbon black weighing, the scale weight zeroing, and the equipment status stabilization are completed, to prevent the process from being interleaved or disordered. Error correction logic: If the weighing deviation is within the ±0.2kg threshold, the system will automatically calculate the deviation and complete the micro-compensation. If the deviation exceeds the threshold, the feeding will stop immediately and an audible and visual alarm will be triggered.

[0037] In actual production, after the system reads the mapped formula model, it first extracts the weighing parameters for 25kg of carbon black and 3.5kg of oil, the error parameter of ±0.2kg, and the timing parameters for feeding carbon black first and oil last. Then, it calls the built-in process rules to match the "fast feeding + slow replenishment" rule for powdered carbon black and the "uniform speed inching feeding + wall-hanging compensation correction" rule for liquid oil. At the same time, it verifies that the process of feeding carbon black first and oil last conforms to the pretreatment timing rules. Finally, it parses out the complete equipment control logic specific to this formula, transforming the abstract formula parameters into executable process action logic for the equipment.

[0038] While dynamically parsing, two core verification tasks are completed simultaneously in parallel: 1. Automatic legality verification: Check whether the entered weighing weight exceeds the equipment's range, whether the error threshold exceeds the industry's allowable standard, whether there are any missing process parameters, or whether the values ​​are out of bounds, etc., against the equipment's rated operating parameter range; mark abnormal parameters and keep logs. 2. Logic matching of multiple process parameters: Verify whether there are process conflicts between the weighing sequence of multiple materials, the feeding time, and the error constraints, such as time conflicts between preceding and following processes, and contradictory parameters, to ensure that the logic of the entire pretreatment process is self-consistent and can be executed normally.

[0039] During actual verification, if the operator mistakenly enters a carbon black weighing of 200kg, exceeding the rated capacity of the equipment's maximum weighing capacity of 80kg, the system will automatically determine that the parameter is illegal and report an error. If the entered process is oil first and then carbon black, which conflicts with the conventional process sequence of internal mixing pretreatment, the system will automatically identify the process logic contradiction and issue a warning to prevent incorrect formulas from being issued to production, thus avoiding problems such as equipment overload, incorrect material mixing, and product quality defects.

[0040] This step completes the standardized analysis and compliance screening of formula parameters, filtering out erroneous, invalid, and conflicting parameters, and transforming the original formula data into standardized logical relationships that meet the equipment process requirements, providing a compliant and accurate process logic basis for the subsequent generation of executable control instructions.

[0041] Finally, the adaptive process rule engine generates a standardized preprocessing instruction set based on the parsing results and transmits it to the intelligent weighing control module for material preprocessing.

[0042] ① Effective parameter extraction: Accurately extract the target weighing value, rapid feeding threshold, slow feeding range, jog fine-tuning parameters, error correction parameters, and process sequence parameters of various materials from the analysis results; ② Parameter instruction conversion: Convert process parameters into digital control instructions and analog numerical instructions that can be recognized by the lower-level PLC, thus completing the conversion from process parameters to electrical control instructions; ③ Instruction timing sorting: Sort, integrate, and encapsulate all control instructions according to the order of preprocessing processes; ④ Generate a complete instruction set: Output a standardized preprocessing instruction set in a unified format that can directly drive the operation of lower-level weighing equipment, and send it to the intelligent weighing control module for material preprocessing via fieldbus.

[0043] The standardized preprocessing instruction set specifically includes: various material target weighing setpoint instructions, fast / slow / jog feeding switching instructions, weighing error automatic correction instructions, material weighing start / stop instructions, preprocessing process timing control instructions, and weighing completion judgment feedback instructions. All of these are electrical control instructions that can be directly recognized and executed by the lower-level PLC.

[0044] Based on the aforementioned carbon black and oil production formulas, the system converts the analyzed process logic into specific equipment instructions: outputting instructions for rapid carbon black feeding start, instructions for switching to slow replenishment when carbon black approaches its rated weight, instructions for jogging oil feeding control, instructions for automatic correction of weighing errors, instructions for stopping after feeding completion, and instructions for timing switching. The entire instruction set is arranged according to the production sequence. After receiving the instructions, the lower-level PLC can directly and automatically control the weighing equipment to complete the entire set of material pretreatment and weighing operations without the need for manual operation of the equipment step by step.

[0045] This step completes the final transformation from upper-level process logic to lower-level equipment control commands, breaking down the data barriers between the formula management level and the on-site execution level, and providing accurate, orderly, and implementable control commands for the intelligent weighing preprocessing of lower-level modules.

[0046] This engine is the core data processing and process decision-making module of the entire internal mixer auxiliary automatic control system. It undertakes the entire process of formula input, standardized modeling, data adaptation, process verification, logic parsing, and instruction generation, realizing the upgrade of internal mixer pretreatment formula from manual experience-based control to standardized, intelligent, and automatic system control. The standardized instructions output by this engine form the foundation for all backend weighing, status sensing, feeding optimization, and closed-loop control, ensuring process uniformity, parameter compliance, and precise control throughout the system. It is the core source support for the system to achieve fully automatic closed-loop control.

[0047] (2) The intelligent weighing control module for material pretreatment is deployed in the lower-level PLC substation cabinet as the core pretreatment execution unit of the field execution layer of the whole system. The lower-level PLC substation cabinet itself has high reliability and high real-time field computing and logic execution capabilities. This module relies on the high real-time computing resources inside the PLC substation cabinet to complete all logic judgments and timing control. At the same time, in order to realize direct electrical connection with the field actuators, this module further calls the high-speed switch output module (used to quickly output on / off control signals to various valves and pumps) and analog acquisition module (used to collect the continuous weight signal of the weighing scale in real time) configured in the PLC substation cabinet, thereby completing the direct control of the field actuators and sensing elements such as vacuum adsorption conveying device, weighing scale, fast and slow jog valve, belt scale, and temporary storage bin.

[0048] This module runs in real time on the lower-level computer front end. It is responsible for receiving the standardized preprocessing instruction set issued by the upper-level formula dynamic management engine, and automatically controlling the complete weighing preprocessing process of carbon black, powder, oil and rubber. Finally, it generates weighing completion status data and transmits it to the equipment status real-time sensing module.

[0049] The material pretreatment intelligent weighing control module consists of two functional units that work together: a vacuum adsorption weighing unit and a multi-segment feeding control unit. The two units work in a hierarchical linkage relationship: first, the vacuum adsorption weighing unit completes the vacuum conveying and dynamic high-precision weighing of the material and generates intermediate weighing results; then, the multi-segment feeding control unit completes the temporary storage of the weighed qualified material, locking the feeding sequence, and encapsulating the final status data packet, and finally generates and uploads the weighing completion status data. like Figure 5 The diagram shows the working monitoring interface of the vacuum adsorption weighing unit in the system front end. This interface displays the operating status of the vacuum adsorption weighing unit, vacuum level, adsorption pressure, and adsorption time. The vacuum adsorption weighing unit controls the material conveying of the vacuum adsorption device and uses three methods—rapid feeding, slow replenishment, and micro-inching—combined with an automatic calibration function to dynamically weigh carbon black powder and oil. The specific execution steps are as follows: The first step is the reception and parsing of standardized preprocessing instruction sets: The vacuum adsorption weighing unit receives standardized preprocessing instruction sets issued by the formula dynamic management engine in real time via fieldbus. The instruction set clearly includes the material list to be weighed in the current batch (carbon black, inorganic powder, operating oil), the target weighing value for each material, the rapid feeding switching weight threshold, the slow replenishment weight range, the end-point jogging fine-tuning weight range, the maximum allowable positive deviation, the maximum allowable negative deviation, the out-of-tolerance handling logic, and the material shortage replenishment trigger conditions, among other quantitative parameters. The system performs a completeness check on the instruction set, confirming that there are no missing or invalid values, and then writes the parameters to the local weighing buffer, entering the equipment initialization preparation state.

[0050] The second step is the initialization and zero-point calibration of the weighing equipment: A comprehensive initialization is performed on the vacuum pump, main delivery valve, fast / slow jog valve, weighing platform, and pipeline pressure sensor for the actual scenario. First, the automatic tare and zero-point calibration program for the weighing platform is started, continuously acquiring weighing platform signals for multiple sampling periods (obtained through the aforementioned analog signal acquisition module). Once the zero-point drift is confirmed to be less than the set threshold and stable, the zero point is locked. Simultaneously, the pressure of the vacuum system is checked to ensure it reaches the set working range, and the presence of residual pressure or blockage signals in the delivery pipeline is detected. If any step of the initialization process malfunctions, a local audible and visual alarm is immediately triggered, and a fault code is reported, prohibiting the weighing process from proceeding.

[0051] The third step is vacuum adsorption conveying path matching and startup: A dedicated vacuum adsorption path is automatically matched based on the material type specified in the instruction. For powdered carbon black or inorganic powders, a high-flow-rate vacuum pump and main conveying butterfly valve are activated via a high-speed switching output module, adsorbing the material from the high-level silo to the buffer chamber above the weighing scale. During conveying, vacuum pressure and material flow rate are monitored in real time. If the pressure drops suddenly or the flow rate becomes abnormal, a blockage is detected, and automatic backflushing is initiated to clear the blockage. For liquid oils, a dedicated oil vacuum pump and micro-regulating valve are activated to control the oil to enter the oil scale at a stable low flow rate. Simultaneously, the pipe wall pre-wetting logic is activated to reduce residue during the initial conveying. After the material is conveyed to its destination, the main valve is closed, and the system enters the weighing waiting state.

[0052] The fourth step is the precise execution of three-stage dynamic weighing: The feeding mode is automatically switched strictly according to the segmented thresholds in the instructions to achieve high-precision dynamic weighing. Taking a target carbon black weighing of 25.0 kg with an allowable error of ±0.2 kg in actual production as an example: When the current real-time weight of the scale (continuously acquired by the analog quantity acquisition module) is less than 90% of the target value (i.e., less than 22.5 kg), the system opens the large-diameter fast feeding valve through the high-speed switch output module to achieve a large flow rate of rapid material feeding, shortening the overall weighing time; when the weight enters the 22.5 kg to 24.5 kg range, the fast valve is automatically closed and the small-diameter slow feeding valve is opened, significantly reducing the feeding speed and preventing material inertia overshoot; when the weight enters the 24.5 kg to 25.0 kg range, it switches to a micro-inching feeding mode, where each inching action controls only a very small flow rate of material to fall onto the scale, and after each inching action, it waits for the scale signal to stabilize before deciding whether to continue inching. The oil is operated in a forced micro-inching mode throughout the process, and the pipe wall residue compensation logic is activated simultaneously (automatically adding a small amount of material to the target value based on the historical average residue of the oil). The entire segment switching process is automatically completed by the high-speed comparison logic inside the PLC, without the need for manual intervention.

[0053] The fifth step is real-time automatic calibration and qualification judgment: During the weighing process, the weight signal of the scale is continuously collected at fixed short intervals and compared with the target value and error threshold in real time. If the final stable weight falls within the qualified range of 24.8kg to 25.2kg, the system determines that the material is weighed qualified, locks the current actual weight, and records the deviation value; if there is insufficient material (less than 24.8kg), the micro-incremental feeding program is automatically triggered until the weight enters the qualified range or the maximum feeding limit is reached; if the deviation is exceeded (greater than 25.2kg), all feeding valves are immediately closed, an audible and visual alarm is triggered, the "Carbon Black Out of Tolerance" alarm position is displayed on the local operation screen, and the batch of material is marked as out-of-tolerance waste, prohibiting it from entering the subsequent temporary storage and feeding process. At the same time, the scale is subjected to a second zero-point calibration to eliminate interference factors such as material accumulation, vibration, and temperature drift.

[0054] The sixth step is the generation and transmission of intermediate weighing results: After all powders and oils in the instruction list have been weighed to the required standard, the vacuum adsorption weighing unit generates an intermediate weighing result data packet containing "material type, target value, actual weighing value, deviation value, weighing time, qualification mark, and number of replenishments". This data packet is immediately transmitted to the multi-segment feeding control unit through the internal data link as the input basis for subsequent temporary storage and final state encapsulation.

[0055] In actual production scenarios, after the system receives a standardized instruction for a batch of "25.0 kg of carbon black, 3.5 kg of oil, and an error of ±0.2 kg", the vacuum adsorption weighing unit first completes the three-stage weighing of carbon black in the form of "fast-slow-jog" (actual weight is 24.95 kg, deviation is -0.05 kg, and it is deemed qualified), and then completes the full-process jog weighing of oil (actual weight is 3.52 kg, deviation is +0.02 kg, and it is deemed qualified). Throughout the process, no one intervenes in the valve opening and closing or manually observes the scale reading.

[0056] like Figure 6 The diagram shows the monitoring interface of the multi-segment feeding control unit in the system front-end. The interface displays the operating status, current feeding segment, amount of material already fed, and remaining feeding amount. The multi-segment feeding control unit receives standardized preprocessing instruction sets and intermediate weighing results uploaded by the vacuum adsorption weighing unit. It controls the feeding device to send qualified material to the internal mixer's temporary storage position or directly prepare for feeding, and generates final weighing completion status data to the equipment status real-time sensing module. The specific execution steps are as follows: The first step is to receive and verify the integrity of intermediate results: This involves receiving the intermediate weighing result data packets uploaded by the vacuum adsorption weighing unit in real time, verifying that all materials have been marked as "qualified," that the actual weight is within the error range, and that the timestamp of the data packet is within the valid window. If any material is unqualified or the data is abnormal, the subsequent process is immediately stopped and a report is submitted.

[0057] The second step is the control and confirmation of material arrival in the temporary storage bins: A high-speed switch output module controls the opening of the corresponding temporary storage bin's feed valve, allowing qualified carbon black, powder, and oil to be fed into their respective independent temporary storage hoppers. Simultaneously, a rubber belt scale is activated to continuously and dynamically weigh the solid rubber compound and synchronously feed it into the rubber compound temporary storage position. Both level sensors and weight signals confirm that the material in each temporary storage bin is in place and stable.

[0058] The third step is to lock the feeding sequence and check the preconditions: Based on the process sequence requirements in the standardized pretreatment instruction set, a strict feeding sequence is implemented for the temporarily stored materials (carbon black / powder feeding preparation must be completed first, followed by oil, and finally rubber). Simultaneously, the signal indicating the internal mixer's feeding door is closed, the rotor's operating status, and the mixing chamber temperature are monitored in real time to ensure they meet the preconditions for feeding. If any precondition is not met, the temporary storage is locked, prohibiting any feeding actions to prevent accidental feeding.

[0059] The fourth step is the final weighing completion status data packet encapsulation and uploading: All information, including vacuum adsorption weighing results, actual weight of the rubber belt scale, material arrival signals from each temporary storage bin, equipment readiness flags, total weighing time, and pass rate statistics, is integrated and encapsulated according to a pre-set fixed data structure to generate a unified weighing completion status data packet. This packet is transmitted in real-time via the fieldbus to the real-time equipment status sensing module with the highest priority, serving as the core input for subsequent equipment status fusion and material feeding decisions.

[0060] This module completes the implementation from standardized instructions at the higher level to high-precision weighing execution and status feedback on site, providing accurate, real-time, and structured material preparation status data support for subsequent equipment status perception and optimized material feeding.

[0061] (3) The real-time equipment status sensing module is deployed in the lower-level PLC substation cabinet, serving as the core for status acquisition and fusion at the field execution layer of the entire system. The lower-level PLC substation cabinet already has signal acquisition capabilities. Based on this, this module further utilizes the native communication port (used for data interaction with other substation cabinets and the host computer), the high-speed switch input module (used for quickly acquiring various switch signals and operation signals), and the high-precision analog signal acquisition module (used for acquiring continuously changing analog signals such as temperature, power, and pressure) within the PLC substation cabinet to achieve comprehensive monitoring of various native signals on site. This module runs in real-time on the lower-level computer front end, directly monitoring various native signals on site and the weighing completion status data uploaded by the intelligent weighing control module for material preprocessing. It is responsible for fusing discrete, multi-source, and heterogeneous equipment operation information into a unified, structured, and time-aligned equipment status vector, and transmitting it to the optimized feeding decision engine in real time.

[0062] The real-time equipment status sensing module includes a native signal monitoring unit and a status fusion processing unit. The two units work in a hierarchical linkage relationship: the native signal monitoring unit first completes the real-time acquisition and preprocessing of all field signals, and then the status fusion processing unit completes the time alignment, normalization fusion and vector generation of multi-source data. The native signal monitoring unit is used to monitor the PLC's native communication signals, the mixer's switching signals, and temperature and power analog signals, thereby collecting equipment operating status data. The specific steps are as follows: The first step is signal channel initialization and communication link establishment: After the system is powered on, it automatically configures the PLC communication port parameters, the address mapping of the high-speed digital input module, and the range and filtering parameters of the analog acquisition module, establishing a stable high-speed data link with the internal mixer controller, various weighing devices, conveying devices, valve actuators, temperature sensors, and current transformers. Simultaneously, bus heartbeat monitoring is activated to ensure the communication link is online in real time.

[0063] The second step involves continuous real-time monitoring of three core signal categories: The following three types of signals are continuously monitored at a fixed sampling period: PLC native communication signals, including internal communication status between lower-level substations, bus heartbeat packets, equipment fault codes, and communication interruption flags; Internal mixer switch signals (acquired via a high-speed switch input module), including the opening / closing positions of the feeding door and unloading door, rotor forward / reverse rotation, emergency stop button status, safety light curtain triggering, and compressed air pressure switch status; and temperature and power analog signals (acquired via a high-precision analog acquisition module), including continuous analog values ​​such as the front rotor temperature, rear rotor temperature, mixing chamber temperature, real-time motor current, real-time motor power, vacuum system pressure, and cooling water temperature. All signals are accurately timestamped before being entered into the raw data buffer.

[0064] The third step is raw data preprocessing and interference removal: the acquired switch signals undergo both hardware and software debouncing, and rising / falling edge events are detected; analog signals are subjected to moving average filtering and abnormal transition removal, and the raw voltage / current signals are converted into engineering units (°C, A, kW, MPa). The preprocessed clean data stream is written to the local real-time database for use by the state fusion processing unit.

[0065] The status fusion processing unit fuses the operational status data and weighing completion status data to generate a unified equipment status vector, which is then transmitted to the optimized feeding decision engine. The specific steps are as follows: The first step is multi-source data synchronous reception and time alignment: Real-time reception of equipment operation status data streams uploaded by the native signal monitoring unit, and weighing completion status data packets uploaded by the intelligent weighing control module for material pretreatment. A unified timestamp mechanism is used to align the two types of data to the same time window (usually based on the latest weighing completion time), ensuring no time misalignment during fusion.

[0066] The second step is data normalization and key feature extraction: Switching quantities are encoded into standard 0 / 1 binary features (e.g., feed door closed = 1, rotor running = 1, emergency stop not triggered = 1); analog quantities are normalized according to a preset range, and key status features are extracted (e.g., whether the rotor temperature is within the allowable process range, whether the motor power is stable, and whether the vacuum pressure is normal). Simultaneously, features such as "all material weighing qualification marks, actual weight of each material, and temporary storage bin arrival mark" are extracted from the weighing completion status data package.

[0067] The third step is the unified equipment status vector structure generation: Following a pre-set fixed vector structure template, all extracted features are filled in according to a fixed field order to generate a unified equipment status vector with fixed dimensions, a uniform format, and direct parsing by subsequent modules. Typical vector fields include: weighing completion flag, actual carbon black weight, actual oil weight, actual rubber weight, feed door closed flag, rotor running flag, real-time rotor temperature value, real-time motor power value, overall fault alarm flag, overall equipment ready flag, and timestamp.

[0068] The fourth step involves real-time uploading and historical caching: the generated unified equipment status vector is transmitted to the optimized feeding decision engine in real time via the fieldbus with the highest priority. Simultaneously, historical vectors from the most recent periods are cached in a local circular buffer on the lower-level machine for subsequent fault tracing and report management modules to access.

[0069] In actual production scenarios, when carbon black, oil, and rubber have all been weighed and qualified, the internal mixer's feeding door is closed, the rotor temperature is within the normal process range, and the motor power is stable without any fault alarms, the state fusion processing unit quickly generates a complete state vector and uploads it, providing millisecond-level accurate panoramic equipment status for subsequent feeding decisions.

[0070] (4) The optimized feeding decision engine operates collaboratively between the upper and lower levels, receives the unified equipment status vector and reads the material pretreatment formula data in the standardized formula model, then generates an optimized feeding scheme and transmits it to the linkage control command issuing module; it serves as the core strategy decision unit of the entire system. This engine relies on the process calculation capabilities of the upper management computer (used for complex logic operations and multi-constraint solutions) and the real-time status input uploaded by the lower PLC substation cabinet to complete all operations. After receiving the unified equipment status vector uploaded by the equipment status real-time perception module, the engine synchronously reads the standardized formula model cached in the formula dynamic management engine, is responsible for calculating the multi-material linkage logic relationship and the internal mixing feeding trigger conditions, and generates the current optimal feeding scheme under multiple constraints, and finally transmits it to the linkage control command issuing module.

[0071] The optimized feeding decision engine comprises a process parameter evaluation unit and a multi-constraint optimization matching unit. These two units work in a hierarchical, interconnected manner: first, the process parameter evaluation unit accurately calculates the formula process logic and triggering conditions; then, the multi-constraint optimization matching unit comprehensively matches the current state with multiple constraints and generates the final executable solution. Specifically: The process parameter evaluation unit is used to calculate the linkage logic and the internal mixing feeding trigger conditions based on the material pretreatment formula data in the standardized formula model; the specific steps are as follows: The first step is to synchronously read the standardized formula model: read the standardized formula model of the current batch in the formula dynamic management engine in real time through the upper and lower level data interaction interface, and extract the raw material ratio data, error range, and all parameters related to feeding in the process steps, including the feeding sequence requirements of each material, the allowed feeding time window, the allowed feeding temperature range and power range of the internal mixer, and the maximum time limit for material temporary storage.

[0072] The second step involves precise calculation of the multi-material linkage logic: based on the formulation and process requirements, strict linkage constraints are constructed between multiple materials. The system employs a linkage logic algorithm based on time priority, and the specific calculation formula is as follows: ; in: : No. Planned material feeding start time (unit: seconds); : No. Planned material feeding start time (unit: seconds); : No. The estimated feeding duration (in seconds) for each material is directly read from the process step data in the standardized formulation model. No. The mandatory waiting interval (in seconds) after the material is fed is given by the process timing constraints in the standardized formulation model to ensure that the preceding materials are fully dispersed or reacted.

[0073] This formula mandates that the start time of the next material must be later than the completion of the previous material's feeding and the necessary waiting period, thus ensuring a strict sequence and minimum time interval, and avoiding material mixing or timing conflicts. The system calculates by substituting the materials in order of priority (e.g., carbon black → oil → rubber) to obtain a complete linkage timing chain.

[0074] The third step is to generate the set of trigger conditions for internal mixer feeding: combining the real-time process window of the internal mixer and the formula requirements, a complete set of trigger conditions for allowing feeding of the current batch is generated. The system adopts a trigger determination algorithm that combines multi-condition logic and weighted scoring. The specific calculation formula is as follows: ; in: : Final decision result for feeding materials (Boolean value, true indicates feeding is allowed, false indicates feeding is prohibited); : No. The values ​​of the trigger sub-conditions are (0 or 1). Typical sub-conditions include: all materials are weighed and qualified, the internal mixer feed door is closed, the rotor is running, the rotor temperature is within the allowable range of the formula, the motor power fluctuation is less than the set threshold, there are no fault alarms, and the temporarily stored materials are in place. The total number of triggering sub-conditions; Weighted composite score; : No. The weighting coefficient of each triggering sub-condition (the value ranges from 0 to 1 and is preset according to the importance of the process, for example, temperature conditions have a higher weight). : The minimum score threshold for allowing material input (given by process parameters in the standardized formulation model).

[0075] In this formula, the first part uses a logical AND operation to ensure that all hard safety and process conditions are met; the second part uses a weighted score to quantitatively evaluate the soft conditions, and only when the comprehensive score reaches or exceeds the threshold is feeding finally allowed. The combination of the two ensures both safety and process flexibility.

[0076] The multi-constraint optimization matching unit generates an optimized feeding scheme based on logical relationships, mixing material feeding trigger conditions, and a unified equipment state vector, and then transmits it to the linkage control command issuance module. The specific steps are as follows: The first step is a real-time comparison of the state vector and trigger conditions: the latest unified equipment state vector received is compared precisely with the set of trigger conditions generated by the process parameter evaluation unit. Only when all trigger conditions are met will the subsequent optimization and solution stage begin; if any condition is not met, the process remains in a waiting state, and the specific condition that is not met is recorded for alarm and tracing purposes.

[0077] The second step is optimization under multiple constraints: Under the premise of satisfying all triggering conditions, and comprehensively considering multiple objectives such as hard constraints on material feeding sequence, time window constraints, material storage time constraints, current equipment load constraints, and historical deviation patterns, a built-in priority sorting and time window matching algorithm is used to generate the optimal material feeding scheme for the current moment. The scheme explicitly includes: the precise feeding sequence of each material, the start time of feeding each material, the expected feeding duration, and rollback and alarm strategies in case of abnormal interruptions; specifically: The built-in priority sorting and time window matching algorithm is divided into two consecutive execution processes. The overall calculation is completed based on the system's stored formula standard process parameters and real-time status data of on-site equipment. The step-by-step execution logic is as follows: The first stage involves performing material priority ranking and sorting calculations. The system first retrieves the preset basic material process priority levels within the standardized formula model. Core fillers such as carbon black and oils, which directly affect the mixing quality, are set to the highest priority, auxiliary filler powders are set to medium priority, and basic rubber raw materials are set to regular priority. Simultaneously, a dynamic correction coefficient is added to complete a secondary priority adjustment. The dynamic correction coefficient is determined by real-time data. If a certain type of material has been weighed and has been in a waiting-to-feed state for a long time, the current temporary priority of that material is increased. If the corresponding feeding screw conveyor is overloaded and there is a risk of overload, the temporary priority of that material is decreased. After all materials have completed priority scoring and sorting, a preliminary material feeding sorting queue is output.

[0078] The second phase involves time window matching verification and scheme correction calculations. The system retrieves the running sequence of the internal mixer and the available working time periods of each feeding and conveying device to form independent time windows. The material sorting queue output from the first phase is matched and compared with the idle time windows of each device one by one. First, the available time period of the dedicated conveying device corresponding to each material is matched to determine whether the duration required for material feeding can be completely embedded in the idle time interval of the device. If the feeding time of a single material exceeds the current idle window of the corresponding device, the matching of the next idle time period is postponed, and the time difference generated by the postponement is recorded simultaneously. Then, the maximum material storage time limit constraint is checked. If the total waiting time of the material after the postponement exceeds the allowable storage threshold specified in the formula, the current matching path is determined to be invalid, and the system automatically returns to the previous level material sorting node, lowers the sorting order of the material corresponding to the high-load device, and restarts the time window matching.

[0079] Repeatedly iterate and adjust the sorting and time window matching and verification operations until all materials are matched to a feasible feeding time interval that does not violate all constraints such as temporary storage time limit, equipment load, and equipment occupancy period. After the iteration is completed, integrate all material sorting results, feeding start time matched by each material, feeding duration, and synchronously bind the preset equipment interruption handling logic, and finally output a unique optimal feeding scheme that meets all process and equipment constraints.

[0080] The third step is to optimize the feeding scheme encapsulation and real-time transmission: the optimal feeding scheme obtained by the solution is encapsulated according to the fixed data structure preset by the system, and transmitted in real time to the linkage control command transmission module with the highest priority through the fieldbus, as the direct basis for the generation of the final command.

[0081] In actual production scenarios, when the state vector shows that carbon black, oil, and rubber have all been weighed and qualified, and the internal mixer is in the allowable feeding window, the multi-constraint optimization matching unit quickly generates an optimization scheme of "first feed carbon black → feed oil at a set interval → feed rubber at a set interval," and clarifies the abnormal handling strategy for each stage. This avoids the mixing quality fluctuations and equipment impact problems caused by the disordered feeding sequence, improper timing, and experience-based intervals in traditional manual feeding.

[0082] (5) The linkage control command issuance module is deployed in the lower-level PLC substation cabinet. It receives the optimized feeding scheme, generates specific control commands and issues them to the actuators. At the same time, it feeds back the execution results to the real-time equipment status sensing module. It serves as the core of the final command execution and closed-loop monitoring of the entire system's field execution layer. The lower-level PLC substation cabinet already has the ability to control the start and stop of the equipment. This module further calls the high-speed switch output module (used to quickly output on / off control signals) and analog output module (used to output continuous adjustment signals to equipment such as frequency converter pumps) in the PLC substation cabinet to directly drive the actuators such as conveying devices, weighing devices, feeding valves, and belt scales. This module runs in real time in the lower-level machine front end. It is responsible for receiving the optimized feeding scheme issued by the optimized feeding decision engine, converting it into a standardized control command sequence that can directly drive the field equipment and issuing it in a precise timing sequence. At the same time, it collects the execution feedback signal in real time through closed-loop monitoring and compares it with the expected status to ensure that each command is executed accurately. Finally, it feeds back the complete execution results to the real-time equipment status sensing module. The linkage control command issuance module includes a command generation unit and a closed-loop execution monitoring unit. These two units work in a hierarchical linkage relationship: first, the command generation unit converts the optimized scheme into a standardized control command sequence and issues it; then, the closed-loop execution monitoring unit performs real-time monitoring, verification, and result feedback of the execution process. The instruction generation unit is used to convert the optimized feeding scheme into a standardized sequence of control instructions and issue it; the specific steps are as follows: The first step is to optimize the real-time reception and parsing of the feeding scheme: The optimized feeding scheme data packets issued by the optimized feeding decision engine are received in real time through the fieldbus, and all parameters such as the feeding sequence of each material, the feeding start time of each material, the expected duration, the target execution mechanism address, and the exception handling strategy are accurately parsed and written into the local instruction cache.

[0083] The second step is the standardized control instruction sequence conversion: converting the process-level parameters in the scheme into digital control instructions and analog numerical instructions that can be directly recognized and executed by the lower-level PLC. For example, "add carbon black" is converted into "open the carbon black feeding valve (set the corresponding output point to 1 through the high-speed digital output module) + start the carbon black vibrator"; "add oil" is converted into "turn on the oil pump (digital output) + set the pump frequency (write the corresponding value through the analog output module)", etc. All instructions are arranged in chronological order to form a complete standardized control instruction sequence.

[0084] The third step is to issue precise timing information via the fieldbus: according to the precise start-up time specified in the plan, each instruction in the instruction sequence is issued to the target actuator (conveyor, weighing device, feeding device) via the fieldbus at the corresponding time, ensuring that the timing error is controlled within an acceptable range.

[0085] like Figure 3 As shown, a flowchart of instruction closed-loop control is provided. The closed-loop execution monitoring unit is used to collect execution feedback signals in real time and compare them with the expected state. The specific steps to implement instruction closed-loop control include: First, the standardized control instruction sequence output by the instruction generation unit is received and bound to the expected execution status table to form an "instruction-expected feedback" lookup table. Then, control commands are sent to the conveying device, weighing device and feeding device via fieldbus; at the same time, the corresponding feedback signal acquisition channels are activated, wherein the digital feedback is acquired through the high-speed digital input module, and the analog feedback such as weight is acquired through the analog acquisition module. Subsequently, the valve position signal, weighing value, and internal mixer feed gate status are collected and verified in real time. Specifically, all feedback signals, including valve position signals (open / closed), real-time weight change of the weighing scale, internal mixer feed gate status, and belt scale operation signals, are collected in real time at fixed short cycles and compared with the expected execution status table item by item. For example, after issuing the command "Open carbon black feed valve", the command is considered successfully executed only if "carbon black valve open = 1" is received within the set time and the weight of the weighing scale begins to decrease steadily. Finally, when weighing errors, timeouts, zero-point deviations, or valve misalignment are detected, an audible and visual alarm is immediately triggered. The specific alarm location and alarm type are simultaneously displayed on the local operation screen and the host computer interface. At the same time, the complete execution results (success / failure, actual time consumed, and exception code) are fed back to the real-time equipment status sensing module in real time for status vector updates and subsequent decision-making.

[0086] (6) The system also includes an auxiliary report management module. This module interacts with the dynamic formula management engine, collecting operational data from each module to generate shift reports, curve reports, alarm records, and material statistics reports. It also supports data querying, printing, and at least three years of historical data retention. Specifically, this module interacts with the dynamic formula management engine and is deployed on the host computer. The host computer has large-capacity storage and network communication capabilities. This module relies on the host computer's large-capacity solid-state drive (for long-term historical data storage), high-speed network communication interface (for remote data interaction), and the report components built into the configuration software to complete all operations. The report management module collects all operational data from each core module of the system, automatically generating shift reports, curve reports, alarm records, and material statistics reports. It also supports data querying, printing, and at least three years of historical data retention.

[0087] The report management module works in conjunction with the real-time equipment status sensing module to achieve remote monitoring via a network interface, and feeds historical data back to the dynamic formula management engine to support formula modification, production plan adjustment, and hierarchical management of user passwords; the specific steps are as follows: The first step is the real-time acquisition and categorized storage of all operational data: Through the upper and lower level data interaction interface, all operational data is collected periodically, including formula change records and version information from the formula dynamic management engine, weighing results and deviation data from the material pretreatment intelligent weighing control module, the status vector history of the equipment status real-time sensing module, the feeding scheme and decision-making time of the optimized feeding decision engine, and the command execution results and alarm records from the linkage control command issuance module. The collected data is categorized into four main types: "shift, material, equipment, and alarm," and written to the local real-time database on the upper computer.

[0088] The second step involves the automatic generation and visualization of various report types: Automatically generating shift reports by shift (including key KPIs such as shift output, consumption of each material, weighing pass rate, average weighing time, and feeding success rate); automatically generating curve reports by time axis (temperature curve, power curve, weighing weight trend curve, feeding sequence Gantt chart, etc.); real-time scrolling recording of alarm records (alarm occurrence time, alarm type, alarm location, alarm duration, and processing result); and generating material statistics reports by material dimension (total consumption of each material, number of deviations, number of replenishments, average deviation, etc.). All reports support one-click export or direct printing.

[0089] The third step is long-term storage and remote monitoring of historical data: All raw data and generated reports are cyclically stored on the host computer's local solid-state drive for no less than three years, supporting quick retrieval and query by time, shift, material, and alarm type. Simultaneously, through industrial Ethernet or wireless network interfaces, remote clients (workshop offices, quality inspection departments, management) can monitor the current production status in real time, view real-time curves and historical reports, achieving truly transparent remote management.

[0090] The fourth step involves closed-loop feedback and access control of historical data: Key data such as historical pass rate trends, out-of-tolerance patterns, equipment failure frequency, and average production cycle time are periodically fed back to the formula dynamic management engine. Process engineers can use this data to optimize error thresholds in the formula, adjust feeding intervals, or modify production plans. Simultaneously, hierarchical access control for user passwords is implemented: operators can only view real-time status and shift reports; process engineers can modify formula parameters and view historical data; administrators have full access and are responsible for managing user accounts and passwords.

[0091] This auxiliary module completes the entire lifecycle of system operation data collection, statistics, storage, display, and closed-loop optimization.

[0092] In summary, this system is based on a two-level hierarchical control architecture. Through the progressive collaboration of five core modules and one auxiliary module, it forms a complete control chain: "Formula analysis and preprocessing → intelligent weighing and material preparation → equipment status perception → feeding strategy optimization → closed-loop command execution → data feedback optimization". The formula dynamic management engine is deployed on the host computer, generating a standardized preprocessing instruction set and sending it to the intelligent weighing control module for material preprocessing. This module is deployed on the lower-level PLC, generating weighing completion status data after weighing and temporary storage, and transmitting it to the real-time equipment status perception module. The real-time equipment status perception module merges the weighing completion status data with the field operation signals into a unified equipment status vector and uploads it to the optimized feeding decision engine. The optimized feeding decision engine combines the status vector and formula data to generate an optimized feeding scheme and sends it to the linkage control command issuing module. The linkage control command issuing module converts the scheme into control commands and issues them for execution, while simultaneously feeding the execution results back to the real-time equipment status perception module. The report management module collects data from the entire system to generate reports, supports remote monitoring, and feeds historical data back to the formula dynamic management engine for formula modification and production plan adjustment. Each module achieves bidirectional data synchronization through a fieldbus, ensuring stable closed-loop operation of the system.

[0093] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A microcomputer-based automatic control system for auxiliary equipment on a mixer, characterized in that, The system adopts a two-tiered control architecture consisting of a host management computer and lower-level PLC substations. It includes a dynamic formula management engine, an intelligent material pretreatment weighing control module, a real-time equipment status sensing module, an optimized feeding decision engine, and a linkage control command issuance module. The dynamic formula management engine includes a visual formula configuration unit, a material parameter mapping unit, and an adaptive process rule engine. The visual formula configuration unit acquires user-uploaded material pretreatment formula data in real time and generates a standardized formula model. The material parameter mapping unit establishes a dynamic mapping relationship between material attributes and system tags. The adaptive process rule engine dynamically parses the standardized formula model, generates a standardized pretreatment instruction set, and transmits it to the intelligent weighing control module for material pretreatment. The intelligent weighing control module for material pretreatment receives the standardized pretreatment instruction set to control the weighing process of the material and generates weighing completion status data, which is then transmitted to the real-time equipment status sensing module. The real-time equipment status sensing module receives the weighing completion status data and simultaneously collects equipment operation status data by listening to native signals. It then merges these data to generate a unified equipment status vector, which is transmitted to the optimized feeding decision engine. The optimized feeding decision engine receives the unified equipment state vector and reads the material pretreatment formula data in the standardized formula model, then generates an optimized feeding scheme and transmits it to the linkage control command issuing module. The linkage control command issuing module receives the optimized feeding scheme, then generates specific control commands based on it and issues them to the actuator, while simultaneously feeding back the execution results to the real-time equipment status sensing module.

2. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 1, characterized in that, The specific workflow of the recipe dynamic management engine is as follows: SPA1, the visual formula configuration unit first builds a menu interface, then obtains the material preprocessing formula data uploaded by the user through the menu interface, and then generates a standardized formula model; SPA2, the material parameter mapping unit establishes a dynamic mapping relationship between the physical properties of materials and the virtual labels of the system through the generated standardized formula model; SPA3, the adaptive process rule engine dynamically parses the material pretreatment formula data in the standardized formula model, and simultaneously performs automatic legality verification and logical matching of multiple process parameters; SPA4. The adaptive process rule engine generates a standardized preprocessing instruction set based on the parsing results and transmits it to the intelligent weighing control module for material preprocessing.

3. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 2, characterized in that, The intelligent weighing control module for material pretreatment includes a vacuum adsorption weighing unit and a multi-segment feeding control unit. The two units work together in a step-by-step manner to perform material pretreatment and generate weighing completion status data.

4. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 3, characterized in that, The real-time device status sensing module includes a native signal monitoring unit and a status fusion processing unit, wherein: The native signal monitoring unit is used to monitor the PLC native communication signals, the internal mixer switch signals, and temperature and power analog signals, thereby collecting equipment operating status data. The state fusion processing unit fuses the operating state data and weighing completion state data to generate a unified equipment state vector, which is then transmitted to the optimized feeding decision engine.

5. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 4, characterized in that, The optimized feeding decision engine includes a process parameter evaluation unit and a multi-constraint optimization matching unit, wherein: The process parameter evaluation unit is used to calculate the linkage logic relationship and the internal mixing feeding trigger condition based on the material pretreatment formula data in the standardized formula model. The multi-constraint optimization matching unit generates an optimized feeding scheme based on the logical relationship, the mixing feeding trigger condition, and the unified equipment state vector, and then transmits it to the linkage control command issuing module.

6. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 1, characterized in that, The linkage control command issuance module includes a command generation unit and a closed-loop execution monitoring unit, wherein: The instruction generation unit is used to convert the optimized feeding scheme into a standardized control instruction sequence and issue it. The closed-loop execution monitoring unit is used to collect execution feedback signals in real time and compare them with the expected state.

7. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 6, characterized in that, The specific steps for the closed-loop execution monitoring unit to implement closed-loop command control include: SPB1, a standardized control instruction sequence output by the instruction generation unit; SPB2: Sends control commands to the conveying device, weighing device and feeding device via fieldbus; SPB3: Real-time acquisition and verification of valve position signals, weighing values, and internal mixer feed gate status; SPB4 triggers an alarm signal and displays the alarm location when it detects weighing errors, timeouts, zero-point deviations, or valve misalignment, and feeds back the execution results to the real-time equipment status sensing module.

8. The microcomputer-based automatic control system for auxiliary equipment on a mixer according to claim 1, characterized in that, The system also includes a report management module that interacts with the recipe dynamic management engine, and the report management module is used to collect the operating data of each module.

9. The microcomputer automatic control system for auxiliary equipment on a mixer according to claim 8, characterized in that, The report management module works in conjunction with the real-time equipment status sensing module to achieve remote monitoring through a network interface and feeds back historical data to the formula dynamic management engine, supporting formula modification, production plan adjustment and hierarchical management of user passwords.

Images