Beam-slab prefabrication production line scheduling control system based on industrial internet platform

Abstract

The application relates to the technical field of industrial automation, and discloses a beam slab prefabrication production line scheduling control system based on an industrial internet platform, which comprises a data acquisition module, a state mapping module, a resource evaluation module, a scheduling decision module, a constraint optimization module and an execution feedback module. The modules work cooperatively to complete the whole-process management and control of production line data acquisition, state mapping, resource evaluation, scheduling decision, constraint optimization and execution feedback, realize production line mold table scheduling, thermal compensation control and equipment collaborative integrated management. The application integrates production line multi-source data and converts the data into quantitative control variables, combines feedforward intervention and closed-loop feedback mechanism to optimize scheduling, realizes whole-process standardization and intelligentization of production line scheduling, helps to solve the problems of poor data cooperation, response lag and unreasonable resource matching of traditional production line scheduling, guarantees stable operation of the production line, reduces energy consumption and resource waste, and improves production quality of beam slab prefabricated components and operation efficiency of the production line.

Description

Technical Field

[0001] This invention relates to the field of industrial automation technology, specifically to a scheduling and control system for a beam and slab prefabrication production line based on an industrial internet platform. Background Technology

[0002] The precast beam and slab production line is a specialized production line for the industrialization of buildings and the production of components for transportation infrastructure. It takes precast beam and slab components as the production object and relies on the coordinated processes of mold table transfer, pouring operation, temperature control and curing, kiln entry and exit, and material transfer to complete the production process. It is a key industrial facility for realizing the large-scale and standardized manufacturing of precast components.

[0003] In the current precast beam and slab production line, information such as temperature data, workstation status, and material inventory is collected independently by different equipment. Production scheduling relies on manual experience to determine the order of mold entry and exit from the kiln. Heat curing parameters are set with fixed values, and equipment operating parameters are issued by manual instructions. There is a lack of a unified data interaction mechanism between the various links of the production line, and a complete scheduling and control system has not been formed.

[0004] Because the temperature acquisition, workstation scheduling, thermal compensation control, and equipment operation of the beam and slab precast production line are independent of each other, discrete logistics data and continuous thermodynamic data cannot be processed in a unified manner. There is a lack of quantitative calculation methods for equivalent age and workstation blockage. Maintenance resources and production status cannot be matched in real time. Scheduling decisions lack optimization basis. The execution process lacks closed-loop feedback and dynamic correction, resulting in a disconnect between production line scheduling and actual production status, poor production rhythm, high energy consumption and insulation resource consumption, and a direct impact on the quality stability of precast components and the overall operating efficiency of the production line. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a scheduling and control system for beam and slab prefabrication production lines based on an industrial internet platform. This system aims to solve the problems of scattered data, lack of quantitative basis for scheduling decisions, unreasonable resource matching, and lack of closed-loop feedback in the execution process of beam and slab prefabrication production lines.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a scheduling and control system for a beam and slab prefabrication production line based on an industrial internet platform, comprising: The data acquisition module, state mapping module, resource assessment module, scheduling decision module, constraint optimization module, and execution feedback module work together to achieve full-process scheduling and closed-loop control of the production line. The specific innovative principles are explained below: The data acquisition module collaboratively acquires various types of data through its included physical state acquisition unit, production line topology synchronization unit, and environment and resource perception unit. The physical state acquisition unit acquires the internal temperature sequences collected by temperature measurement nodes within each flow mold station of the production line. The production line topology synchronization unit establishes a data interface with the manufacturing execution system server through an industrial internet platform, reads the execution parameters of the physical compensation instructions at the production line pouring station and the consumption status of the downstream insulation resources, and converts them into a unified time-dimensional matrix. The environment and resource perception unit acquires environmental data from the production line workshop and resource parameters of insulation blankets and heating blankets, and encapsulates them into a one-dimensional heat compensation resource vector. The data collected by each unit is timestamped and standardized to generate the heterogeneous data set of the production line. After protocol encapsulation, it is synchronously uploaded to the manufacturing execution system server, providing data support for subsequent scheduling decisions and solving the data collaboration difficulties caused by the dispersed and inconsistent formats of traditional data acquisition.

[0007] The state mapping module collaborates with the equivalent age calculation unit and the congestion risk quantification unit. The equivalent age calculation unit introduces a thermodynamic maturity model and uses a discrete numerical integration method to accumulate the internal temperature sequence acquired by the physical state acquisition unit along the time axis cycle by cycle to calculate the current mature equivalent age of the precast beam component. The preset standard target age is then subtracted from the mature equivalent age to obtain the remaining equivalent age threshold. The congestion risk quantification unit extracts the production line station parameters from the time dimension matrix generated by the production line topology synchronization unit. After calculation and removal of negative numbers using a maximum value function, it outputs the upstream station congestion margin. Subsequently, the remaining equivalent age threshold and the upstream station congestion margin are converted into digital control variables that the system can process for subsequent modules to call. This realizes the transformation of physical data into control variables, provides a quantitative basis for scheduling decisions, and solves the problem of traditional scheduling relying on experience and lacking quantitative indicators.

[0008] The resource assessment module includes a thermodynamic demand prediction unit and a supply-demand matching verification unit. The thermodynamic demand prediction unit quantifies the additional thermal resistance required to maintain the safe temperature difference between the inside and outside of the component based on Newton's law of cooling and Fourier's law of thermal conduction. The supply-demand matching verification unit compares the thermal compensation resource parameters with the additional thermal resistance and the active compensation power to determine the resource matching status. It combines the heterogeneous data set with the remaining equivalent age threshold of the production line mold to calculate the maximum equivalent age compensation limit of the production line under the current environmental conditions, thereby achieving the matching of thermal demand and resource supply. This provides support for early demolding intervention and avoids component quality problems caused by resource waste or insufficient compensation.

[0009] The scheduling decision module includes a feedforward intervention judgment unit, which verifies the blocking margin, the remaining threshold of the equivalent age, and the resource matching status in real time. When the preset conditions are met, it outputs an intervention activation signal. The action sequence generation unit responds to the intervention activation signal, calculates the compensation heat curing time and the mold scheduling priority score, and generates the equipment scheduling sequence according to the score from high to low. That is, the forced kiln exit command and priority sequence for the target mold, realizes feedforward intervention scheduling based on real-time parameters, solves the problems of lag in traditional scheduling response and unreasonable mold priority allocation, and improves the production line operating efficiency.

[0010] The constraint optimization module includes an objective function construction unit and a heuristic search unit. The objective function construction unit establishes a comprehensive evaluation cost function of the production line's global energy consumption and production cycle, and injects hard constraints such as the maximum power supply capacity of the workshop, the physical collision avoidance safety distance of the crane, and the upper limit of the total amount of available heating materials. The heuristic search unit uses the particle swarm optimization algorithm to iteratively solve the problem, extract the optimal equipment scheduling sequence, parse it into the underlying mechanical motion speed control command and output it, so as to achieve a balance between energy consumption and production efficiency, taking into account both economy and practicality.

[0011] The execution feedback module includes an execution status tracking unit and a dynamic compensation and correction unit. The execution status tracking unit drives the crane and transfer equipment to perform mold hoisting and transfer actions according to the motion speed control commands output by the constraint optimization module. At the same time, it collects equipment operation feedback messages and calculates the action hysteresis delay. When the action hysteresis delay exceeds the preset safety grace period, it triggers a hardware-level audible and visual alarm and suspends subsequent transfer commands. The dynamic compensation and correction unit continuously reads the actual temperature sequence of the components returned by the physical state acquisition unit, calculates the maturity compensation gap, and reports the physical compensation command execution parameters and the downstream insulation resource consumption status in real time. The manufacturing execution system server updates the resource library, forming a global control closed loop for the production line. This achieves full-process control of scheduling, execution, feedback, and correction, and improves the stability of scheduling control.

[0012] This invention integrates the data collaboration capabilities of an industrial internet platform with production line scheduling requirements through the collaborative design of various modules. This achieves standardized data collection, precise state mapping, rational resource assessment, intelligent scheduling decisions, and closed-loop execution feedback. It solves the core problems in the scheduling process of existing beam and slab prefabrication production lines, improves the overall operational efficiency of the production line and the quality of component production, and reduces energy consumption and resource waste.

[0013] This invention provides a scheduling and control system for a precast beam and slab production line based on an industrial internet platform. It offers the following advantages: 1. This invention acquires multi-source production line data through the collaborative acquisition of the physical state acquisition unit, production line topology synchronization unit, and environment and resource perception unit of the data acquisition module. The acquired data is then timestamped and format standardized. In conjunction with the state mapping module, the discrete numerical integration method is used to calculate the equivalent maturity age of components and extract workstation parameters to output blocking margins. This transforms physical data into digital control variables that the system can call. By integrating the dispersed production line data and providing quantitative basis, this invention helps to solve the problems of poor coordination and lack of quantitative indicators in traditional scheduling data, thus ensuring the reliability of scheduling decisions.

[0014] 2. This invention verifies the congestion margin, the remaining threshold of the equivalent age, and the resource matching status in real time through the scheduling decision module. It outputs an intervention activation signal and calculates the compensation heat curing time and the priority score of the module scheduling. Combined with the constraint optimization module, it establishes a comprehensive evaluation cost function of the global energy consumption and production cycle of the production line. The particle swarm optimization algorithm is used to solve the optimal equipment scheduling sequence. This invention realizes feedforward intervention and scheduling optimization based on real-time parameters, balances the energy consumption and production efficiency of the production line, and improves the situation of traditional scheduling response lag and unreasonable allocation of module priority.

[0015] 3. This invention drives the crane and transfer equipment to perform mold hoisting and transfer actions through the execution status tracking unit of the execution feedback module, monitors the equipment operation feedback and calculates the action hysteresis delay, and works with the dynamic compensation correction unit to continuously read the actual temperature sequence of the components, calculate the maturity compensation gap, report the execution parameters and update the resource library, forming a global control closed loop for the production line; realizing full-process control of scheduling, execution, feedback and correction, reducing resource waste and component quality problems, and ensuring stable operation of the production line. Attached Figure Description

[0016] Figure 1 This is a system framework diagram of an embodiment of the present invention; Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 3 This is a timing diagram of equipment cross-control and physical quantity changes in a specific application embodiment of the present invention; Figure 4 This is a comparison chart showing the trend of blocking margin changes under different scheduling strategies in a specific application embodiment of the present invention. Detailed Implementation

[0017] The technical solutions in 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.

[0018] See attached document Figure 1 This invention provides a scheduling and control system for a beam and slab prefabrication production line based on an industrial internet platform, comprising: The system includes a data acquisition module, a resource assessment module, a scheduling decision module, a constraint optimization module, and an execution feedback module.

[0019] In this embodiment, the scheduling and control system operates based on a multi-layered physical hardware base distributed throughout the workshop. The hardware base includes edge computing gateways deployed at the front end of the production line and at each workstation, temperature-measuring thermocouples embedded within each transfer mold, an environmental sensor matrix distributed at fixed locations throughout the workshop, a centrally managed manufacturing execution system server, and multiple programmable logic controllers (PLCs) directly connected to the underlying devices. The edge computing gateways are physically connected to the temperature-measuring thermocouples and environmental sensor matrix via an industrial communication bus, forming a sensing network for the underlying physical state. A closed-loop data transmission network is constructed between the edge computing gateways, the manufacturing execution system server, and the PLCs via an industrial Ethernet network, ensuring the flow of multi-source heterogeneous data between the physical and information layers.

[0020] The data acquisition module is deployed inside the edge computing gateway. It continuously acquires the internal temperature sequence of the mold platform, the current topology parameters of the production line, and the three-dimensional wind speed and ambient temperature data of the workshop according to a fixed sampling period. The internal temperature sequence directly reflects the hydration reaction process of the component materials. The data acquisition module encapsulates the collected heterogeneous data sets according to a protocol and synchronously uploads them to the Manufacturing Execution System (MES) server.

[0021] The data acquisition module operates on the computing resources of the manufacturing execution system server, responsible for converting received discrete material flow status data and continuous thermodynamic and physical data into digital control variables that the system can process. Based on the internal temperature sequence, the data acquisition module calculates the equivalent age remaining threshold for each mold in real time. This threshold characterizes the microscopic thermodynamic gap required for the component to reach the standard demolding state. Simultaneously, the data acquisition module calculates the congestion margin of upstream stations based on production line topology parameters, thereby quantifying the time boundary at which upstream casting stations face forced shutdown due to a lack of available molds.

[0022] The resource assessment module establishes communication with the data acquisition module at its data input end. The resource assessment module extracts wind speed and temperature data from the workshop and, combined with the physical resource availability data of the downstream ambient temperature buffer zone, calculates the maximum equivalent age compensation limit under the current environmental conditions. This calculation process establishes a functional mapping relationship between the intensity of the external cold source and the physical properties of available insulation coverings, limiting subsequent system scheduling actions to not exceed the actual physical compensation capacity.

[0023] The scheduling decision module, acting as the control center, maintains data interaction with both the data acquisition module and the resource assessment module. It continuously monitors the congestion margin and uses this as a trigger to switch the system's scheduling state machine. When the congestion margin falls below a preset safety time margin, the system enters a scheduling interruption state. In this state, the scheduling decision module compares the remaining equivalent age threshold of each module with the current maximum equivalent age compensation limit. When the tolerance constraints are met, the scheduling decision module generates a forced kiln exit command and priority sequence for the target module, truncating the original fixed-time-window queuing logic.

[0024] The constraint optimization module, deployed on the programmable logic controller (PLC) side, is responsible for receiving and parsing the forced kiln exit command issued by the scheduling decision module. It reads workshop wind speed distribution data and combines it with spatial topological coordinates to generate a crane servo movement trajectory that avoids areas of strong convection, thus suppressing uncontrolled heat loss during the aerial transfer phase. Furthermore, the constraint optimization module extracts the dynamic anti-sway safety envelope parameters of the servo drive, applies hard constraints to the crane's operating speed, and outputs motion speed control commands for the underlying mechanical structure.

[0025] The execution feedback module's control terminal is directly connected to the constraint optimization module. The execution feedback module drives the overhead crane and rail-guided vehicle to perform the hoisting and transfer actions of the mold platform according to the generated trajectory and speed curve. After the physical transfer is completed, the execution feedback module reports the execution parameters of the physical compensation command and the consumption status of downstream insulation resources to the manufacturing execution system server. Based on this, the manufacturing execution system server updates the resource library, forming a global control closed loop for the system across the dimensions of perception, decision-making, execution, and write-off.

[0026] See attached document Figure 2 This invention provides a scheduling and control method for a beam and slab prefabrication production line based on an industrial internet platform, comprising the following steps: S1, the data acquisition module acquires the internal temperature sequence of the mold, the current topology parameters of the production line, and the three-dimensional wind speed and ambient temperature data of the workshop according to a fixed sampling period, and encapsulates the acquired heterogeneous data set according to the protocol and uploads it synchronously to the data acquisition module. S2, the data acquisition module calculates the equivalent age remaining threshold of each mold in real time based on the internal temperature sequence, and calculates the blockage margin of the upstream station based on the production line topology parameters, quantifying the time boundary of the upstream pouring station facing forced shutdown due to lack of idle molds. S3, the resource assessment module extracts wind speed and temperature data from the workshop, combines it with physical resource availability data from the downstream ambient temperature buffer zone to calculate the maximum equivalent age compensation limit under the current environmental conditions, and establishes a functional mapping relationship between the intensity of the external cold source and the physical properties of available insulation coverings. S4, the scheduling decision module continuously monitors the blocking margin. When the blocking margin is lower than the preset safety time margin, it switches to the scheduling interruption state. In this state, it compares the remaining equivalent age threshold of each mold with the current maximum equivalent age compensation limit. When the tolerance constraint is met, it generates a forced kiln exit command and priority sequence for the target mold. S5, the constraint optimization module receives and parses the forced kiln discharge command, reads the workshop wind speed distribution data and combines it with spatial topological coordinates to generate the crane servo movement trajectory that avoids strong convection areas, extracts the dynamic anti-sway safety envelope parameters of the servo driver to apply hard constraints to the crane's running speed, and outputs the motion speed control command of the underlying mechanical structure. S6, the execution feedback module drives the overhead crane and the rail-guided vehicle to perform the hoisting and transfer actions of the mold platform according to the generated overhead crane servo movement trajectory and motion speed control commands. After the physical transfer is completed, it reports the execution parameters of the physical compensation command and the consumption status of downstream insulation resources to the manufacturing execution system server, forming a global control closed loop.

[0027] The data acquisition module is deployed within the edge computing gateway to obtain real-time status data of the underlying hardware facilities and production environment. The data acquisition module includes a physical status acquisition unit 110, a production line topology synchronization unit 120, and an environment and resource perception unit 130.

[0028] The physical state acquisition unit 110 establishes a communication connection with the temperature measurement nodes deployed inside each precast mold platform. In this embodiment, the temperature measurement nodes use K-type or J-type thermocouples, which are directly in contact with the core area of ​​the concrete component through pre-embedding to collect the real-time temperature of the hydration reaction process. The temperature measurement nodes transmit the collected internal temperature sequence to the edge computing gateway through an RF module based on the LoRa or Zigbee wireless communication protocol. Regarding the specific network deployment and node networking protocol of the wireless sensors in the industrial field, those skilled in the art can choose a star or mesh topology based on the workshop space span. The specific hardware networking implementation of the wireless sensors is a well-known technology and will not be elaborated here.

[0029] Considering the high-temperature and humid steam medium inside the curing kiln during the steam curing process, radio frequency signals are prone to physical attenuation and data packet loss when penetrating air containing suspended water droplets. Therefore, the physical state acquisition unit 110 incorporates a disconnection estimation and polynomial fitting compensation program. In precast concrete processes, components have high specific heat capacity and thermal inertia, and their core temperature changes smoothly and continuously over a very short time, without abrupt jumps. Based on this general physical principle, the polynomial fitting compensation program uses the rate of change of historical valid data to extrapolate missing temperature nodes.

[0030] Specifically, when the physical state acquisition unit 110 detects that the number of consecutive lost temperature data cycles for a specific module is less than a preset safety disconnection threshold, it triggers a polynomial fitting compensation program. The safety disconnection threshold is typically set to a range of 3 to 5 sampling cycles. The physical state acquisition unit 110 extracts the slope of the continuous temperature data within the historical time window before the disconnection, performs linear interpolation, and outputs a fitted temperature estimation sequence to maintain the continuity of the control system state machine. The calculation logic of the polynomial fitting compensation program is implemented by the following formula: ; In the formula, This is a fitted estimate of the temperature corresponding to the time of the current lost message. This is the last successfully received actual temperature observation value before the communication connection was lost. The actual temperature observation value at the starting point of the historical effective time window. The data sampling timestamp at the current moment. This is the end timestamp of the historical valid temperature data. This is the timestamp of the starting point for historical valid temperature data. This is the preset depth parameter for the historical observation window.

[0031] The range of values ​​for the historical observation window depth parameter is set to 5≤ ≤10, this parameter is determined by extracting a valid sample that is sufficient to smooth short-term signal jitter and does not deviate from the current thermodynamic trend. To ensure the integrity of the underlying logic of the computing system, the physical state acquisition unit 110 verifies the absolute value of the time denominator in real time before performing the interpolation operation. If detected... If the value approaches 0, meaning there is overlap in sampling time points within the historical valid time window or a system clock anomaly, the physical state acquisition unit 110 directly intercepts the slope calculation command and outputs the slope instead. This serves as the current fitted value to prevent calculation overflow caused by division by zero anomalies.

[0032] The above fitting process utilizes the continuous heat release characteristics of the physical object to maintain uninterrupted system operation during short-term communication disruptions.

[0033] After acquiring the underlying physical state of the controlled object, to further assess the overall line's operational load, the production line topology synchronization unit 120 establishes a low-level data interface connection with the database of the manufacturing execution system server via industrial Ethernet. The industrial Ethernet uses the OPCUA or Profinet industrial standard communication protocol. The production line topology synchronization unit 120 reads in real-time from the manufacturing execution system server the current number of queued molds at the casting station, the remaining number of buffer positions, the average processing time per station, and the expected timestamp of the most recent idle mold arriving at the casting station. The physical causal basis for extracting the above specific topology parameters is that the number of queued molds and the average processing time together determine the current production line's material throughput limit, while the remaining number of buffer positions directly defines the critical point of mechanical congestion. The production line topology synchronization unit 120 converts the discrete material scheduling data into a unified time-dimensional matrix, characterizing the current physical production line's resilience to process stagnation. In this embodiment, the time-dimensional matrix is ​​constructed as a one-dimensional column vector, specifically expressed as: ; In the formula, For time dimension matrix, This represents the current number of queuing stations. This represents the number of remaining buffer bits. This represents the average processing time per workstation. This is the expected arrival timestamp.

[0034] When constructing the time dimension matrix, the production line topology synchronization unit 120 adopts multi-dimensional weighted fusion logic, which comprehensively evaluates the dynamic logistics pressure of the system by introducing the average processing cycle time and the expected arrival time into the matrix operator.

[0035] Besides the state of the controlled object itself and the production line topology, the external environment and available resources are also key boundaries constraining feedforward scheduling. Therefore, the environment and resource sensing unit 130 establishes underlying communication links with the environmental sensor matrix deployed at fixed locations in the workshop and the warehouse management system, respectively. The environmental sensor matrix consists of an array of wind speed sensors and temperature sensors distributed at the intersections of a two-dimensional planar grid in the workshop. The wind speed and temperature sensors continuously transmit the workshop's three-dimensional wind speed flow field data and ambient temperature data to the edge computing gateway via an RS485 communication bus and using the Modbus RTU protocol. The reason for synchronously collecting the correlation between flow field and temperature data is that the combined intensity of the workshop's cold source directly determines the convective heat transfer coefficient on the surface of the kiln components, thus constituting the boundary condition for the allowable exposure time.

[0036] The environment and resource sensing unit 130 synchronously extracts the current available physical resource status of the downstream ambient temperature buffer zone from the interface of the warehouse management system. The physical resource status specifically includes the real-time physical inventory quantity of available thermal blankets and heating pads, as well as the corresponding model-calibrated thermal impedance parameters. The environment and resource sensing unit 130 encapsulates the available resource quantity and thermal impedance parameters into a one-dimensional thermal compensation resource vector. The one-dimensional thermal compensation resource vector is specifically expressed as: ; In the formula, For a one-dimensional heat compensation resource vector, For the inventory quantity of thermal blankets, This refers to the thermal resistance value of the insulation blanket. The number of heating blankets available. This refers to the rated heating power of the heating blanket.

[0037] This resource vector serves as the physical basis for implementing demolding heat compensation, ensuring that the compensation requirements of the feedforward scheduling instructions fall within the existing material carrying capacity of the workshop.

[0038] After completing the initial acquisition of multi-source independent data, to address the timing misalignment issue caused by the different sources of heterogeneous data, the data acquisition module performs unified timestamp alignment and data format standardization conversion on the acquired internal temperature sequence, time dimension matrix, three-dimensional wind speed and flow field data, ambient temperature data, and one-dimensional thermal compensation resource vector, generating a multi-source heterogeneous data set. In the control logic for multi-dimensional data alignment, since the sampling frequency of the edge-side physical temperature sensor is much higher than the production scheduling data update frequency of the manufacturing execution system server, the data acquisition module uses the highest sampling clock on the edge side as the primary reference timing.

[0039] For logistics scheduling and warehousing resource data with relatively slow update frequencies, the data acquisition module invokes a zero-order hold control strategy to smoothly extend and fill the time-domain data, ensuring that the feature matrices of each dimension achieve operational synchronization within the same computation cycle, thus eliminating scheduling decision errors caused by misalignment of the time dimension of heterogeneous data. The data acquisition module synchronously outputs multi-source heterogeneous data sets to the data acquisition module, providing multi-dimensional data support for subsequent equivalent age calculation and feedforward scheduling actions.

[0040] The data acquisition module runs on the manufacturing execution system server and is used to transform the heterogeneous data collected independently at the lower level into feature variables that the control system can directly perform logical operations and scheduling decisions. Specifically, the data acquisition module includes an equivalent age calculation unit 210 and a congestion risk quantification unit 220.

[0041] The equivalent age calculation unit 210 receives the internal temperature sequence synchronized by the data acquisition module to assess the curing maturity of precast components in each mold. In the physical process of concrete precasting, the increase in component strength is essentially the cumulative result of the exothermic reaction of cement hydration, and the rate of this reaction is highly dependent on the coupling effect between the curing environment temperature and the internal core temperature. To eliminate the assessment error caused by the variable temperature curing process, the equivalent age calculation unit 210 introduces a thermodynamic maturity model based on Arrhenius's law, converting the actual temperature change process into an equivalent isothermal time at a standard reference temperature. Based on the aforementioned principle, the equivalent age calculation unit 210 uses a discrete numerical integration method to accumulate the internal temperature sequence periodically along the time axis, calculating the current mature equivalent age of the component. The discrete calculation logic for the mature equivalent age is implemented by the following formula: ; In the formula, The mature equivalent age calculated at the current moment. The apparent hydration reaction activation energy of concrete materials. This is the universal gas constant. For the first The core temperature of the component measured in each sampling period. The preset standard reference temperature, This represents the discrete sampling step size.

[0042] For the characteristic parameters in the mature equivalent age formula, the universal gas constant The value is fixed at 8.314 J / (mol·K), and the preset standard reference temperature is... In engineering practice, the temperature is typically set at 20 degrees Celsius. Additionally, the apparent hydration reaction activation energy... The value range is set to 30000 J / mol to 45000 J / mol. Apparent hydration reaction activation energy. The specific values ​​are pre-calibrated using laboratory calorimetry based on the cement type and admixture formula of the precast components and entered into the system database. To ensure the stability of the system operation, before performing discrete integral calculations, the equivalent age calculation unit 210 calculates the input core temperature of the components. Perform physical rationality truncation and amplitude limitation.

[0043] If sensor noise interference is detected, causing the temperature at the measuring point to deviate from near absolute zero, resulting in a change in the denominator... If the temperature approaches 0, or the temperature at the measuring point exceeds the preset physical extreme value upper limit (the physical extreme value upper limit is set to 95℃ to 100℃ to correspond to the boiling temperature of the water inside the concrete under normal pressure), the equivalent age calculation unit 210 will forcibly set the contribution value of the abnormal sampling point to zero, thereby avoiding variable overflow caused by division by zero anomalies or divergence in exponential function calculations.

[0044] After obtaining the mature equivalent age, the equivalent age calculation unit 210 retrieves the standard target age required for demolding of the corresponding component. The equivalent age calculation unit 210 calculates the remaining equivalent age threshold by subtracting the mature equivalent age from the standard target age. The remaining equivalent age threshold serves as a quantitative indicator characterizing the physical component's distance from the safe demolding state. When the remaining equivalent age threshold approaches zero, it means that the internal hydration structure of the component has the mechanical strength to bear its own weight and external lifting loads, and can trigger subsequent transfer actions.

[0045] In addition to assessing the curing state of the precast components themselves, the control system also needs to assess the logistical load pressure of the entire production line to achieve overall production line management. The congestion risk quantification unit 220 extracts the time dimension matrix generated by the data acquisition module to calculate the congestion margin of the upstream workstations. The congestion margin reflects the remaining time boundary for upstream casting nodes to be forced to stop and wait due to a lack of available molds without proactive feedforward intervention. To accurately obtain the boundary value, the congestion risk quantification unit 220 extracts the current number of queued molds, the remaining number of buffer positions, and the average processing cycle time per workstation to quantify the physical capacity of the production line to handle the current process backlog. The logic for calculating the congestion margin by the congestion risk quantification unit 220 is implemented by the following formula: ; In the formula, For blocking margin, To find the maximum value function, This represents the number of remaining buffer bits. The queuing penalty weighting coefficient, This represents the current number of queuing stations. This represents the average processing time per workstation. For the expected arrival timestamp, This is the current system evaluation timestamp.

[0046] The queuing penalty weight coefficient ω is set to a value between 1.0 and 1.5. The basis for this determination is that there is a mechanical delay in the start-stop movement of the mold table in the physical production line, and a long queue will cause non-linear logistics losses. Therefore, by introducing a penalty weight greater than 1, the theoretically calculated safe time is artificially compressed, leaving sufficient fault tolerance space for the underlying mechanical actions.

[0047] To ensure the fault tolerance of the decision logic, the blocking risk quantification unit 220 nests a maximum value function around the formula. When the difference in the latter half of the parentheses is negative, it indicates that an idle mold platform will soon arrive at the pouring station, and its arrival time is earlier than the time when the buffer space is completely consumed, indicating that the system is in a state of ample supply with no blocking risk. In this ample supply state, the blocking risk quantification unit 220 directly truncates the blocking margin to 0 to prevent negative time overflow from interfering with the optimization space of the downstream scheduling algorithm.

[0048] Through the above calculations, the congestion margin, as a key pre-parameter triggering the switching of the system scheduling state machine, establishes a numerical mapping from the underlying logistics arrangement to the global downtime risk. Subsequently, the data acquisition module synchronously sends the equivalent age remaining threshold and the congestion margin to the scheduling decision module and the resource assessment module, driving the control system into the next level of control closed loop.

[0049] The resource assessment module runs on the manufacturing execution system server and is responsible for quantifying the physical impact risk of external cold sources on precast components before triggering early demolding intervention, and for verifying the feasibility of insulation materials required after demolding. Specifically, the resource assessment module includes a thermodynamic demand prediction unit 310 and a supply-demand matching verification unit 320.

[0050] The thermodynamic demand prediction unit 310 receives ambient temperature data and three-dimensional wind speed and flow field data synchronized by the data acquisition module. Combined with the internal temperature sequence output by the data acquisition module, it calculates the additional thermal resistance required after demolding of the precast component. In the steam curing process, when a high-temperature precast component is suddenly exposed to the ambient temperature environment of the workshop, a temperature gradient will be generated between the surface of the precast component and the ambient air. Heat exchange will cause the concrete surface to shrink. If the surface shrinkage stress exceeds the current tensile strength limit of the concrete, thermal shock cracks will occur on the surface of the precast component. To suppress the intensity of convective heat transfer, an insulating medium with specific thermal resistance properties must be covered on the surface of the precast component. Based on Newton's law of cooling and Fourier's law of thermal conductivity, the thermodynamic demand prediction unit 310 quantifies the additional thermal resistance required to maintain the temperature difference between the inside and outside of the precast component within a safe range. The calculation logic for the additional thermal resistance is implemented by the following formula: ; In the formula, To add thermal resistance, For the core temperature of the component, For ambient temperature, This is the maximum permissible heat flux density.

[0051] For the characteristic parameter in the additional thermal resistance formula, the maximum allowable heat flux density The value range is set to 150W / m 2 Up to 300W / m 2 Maximum permissible heat flux density The specific values ​​are determined offline by finite element analysis of structural mechanics based on the geometric thickness of the precast components, the thermal expansion coefficient of the concrete material, and the real-time tensile strength. The thermodynamic demand prediction unit 310 outputs the calculated additional thermal impedance in real time, transforming the macroscopic crack prevention requirements into specific physical material indicators.

[0052] After obtaining the physical material indicators, the system needs to further confirm whether the underlying workshop has the material basis to perform protective actions. The supply and demand matching verification unit 320 extracts the one-dimensional thermal compensation resource vector constructed by the data acquisition module and compares the material attributes within the one-dimensional thermal compensation resource vector with the additional thermal resistance. The supply and demand matching verification unit 320 extracts the thermal resistance value of the insulation blanket and the inventory quantity of the insulation blanket from the one-dimensional thermal compensation resource vector. If it is determined that the thermal resistance value of the insulation blanket is greater than or equal to the additional thermal resistance, and the inventory quantity of the insulation blanket is greater than zero, it indicates that the current workshop has the physical conditions to perform passive thermal insulation coverage. When the above judgment conditions are met, the supply and demand matching verification unit 320 sets the resource matching status to true and locks the corresponding inventory of one insulation blanket.

[0053] Considering that in low-temperature seasons or when the three-dimensional wind speed flow field in the workshop exhibits strong convection, the thermal resistance value of conventional thermal insulation blankets cannot meet the heat insulation requirements, and simple passive covering carries the risk of failure. To cope with high heat dissipation conditions, the supply-demand matching verification unit 320 extracts the available quantity of heating blankets and the rated heating power of the heating blankets from the one-dimensional heat compensation resource vector, and initiates the active heat compensation verification logic. The supply-demand matching verification unit 320 calculates the active compensation electrical power required to make up for the heat gap based on the difference between the maximum allowable heat flux density and the current actual heat transfer intensity.

[0054] If the rated heating power of the heating blanket is greater than the required active compensation power, and the available quantity of heating blankets is greater than zero, the supply-demand matching verification unit 320 will also set the resource matching status to true. When both passive insulation and active compensation verifications fail to meet the conditions, the supply-demand matching verification unit 320 will set the resource matching status to false. As a hard physical constraint, the resource matching status bit, together with the equivalent age remaining threshold output by the data acquisition module, is input to the subsequent scheduling decision module, which helps to prevent process accidents caused by improper scheduling that result in thermodynamic damage to precast components.

[0055] The scheduling decision module runs on the manufacturing execution system server, receives correlation parameters output by the data acquisition module and the resource assessment module, comprehensively evaluates the supply and demand status of the production line, and issues control commands. Specifically, the scheduling decision module includes a feedforward intervention determination unit 410 and an action sequence generation unit 420.

[0056] Based on the underlying physical and status data acquired by the aforementioned unit, the feedforward intervention judgment unit 410 extracts the congestion margin and equivalent age remaining threshold synchronized by the data acquisition module, as well as the resource matching status bit output by the resource assessment module. Conventional fixed-cycle production is prone to logistical stagnation during process fluctuations. To mitigate the risk of logistical congestion, the feedforward intervention judgment unit 410 incorporates multi-dimensional condition triggering logic. In the specific anti-congestion judgment process, the feedforward intervention judgment unit 410 continuously verifies whether the congestion margin is less than the preset congestion warning time threshold. The congestion warning time threshold is set to a range of 10 to 20 minutes, with the specific value determined based on the average reciprocating travel time of the workshop handling crane. When the congestion margin is less than the congestion warning time threshold, it indicates that the upstream pouring station is about to experience a production interruption due to mold table congestion, and the feedforward intervention judgment unit 410 immediately reads the equivalent age remaining threshold.

[0057] After assessing the congestion risk in the time dimension, the system needs to further verify the process feasibility at the thermodynamic level. If the remaining equivalent age threshold is less than the preset early demolding tolerance and the resource matching status is true, the feedforward intervention judgment unit 410 outputs an intervention activation signal. The early demolding tolerance is set to 5% to 10% of the standard target age, used to define the safe mechanical boundary for demolding precast components in a slightly incompletely cured state.

[0058] After outputting the intervention activation signal, the system needs to address how to compensate for the physical state loss caused by premature demolding through scheduling instructions. Based on the above process requirements, the action sequence generation unit 420 responds to the intervention activation signal and calculates the compensating heat curing time for the target precast component. Since the precast component is removed from the high-temperature curing kiln prematurely, the hydration reaction rate of the precast component under normal temperature or active heating conditions will change. To ensure the final product quality meets the standards, the action sequence generation unit 420, based on the principle of equivalent age conservation, converts the missing maturity into the actual physical time under the current compensating environment. The logic for calculating the compensating heat curing time by the action sequence generation unit 420 is implemented by the following formula: ; In the formula, To compensate for the heat curing time, This is the equivalent remaining age threshold. It is an exponential function with the natural constant as its base. The activation energy for the apparent hydration reaction. This is the universal gas constant. To compensate for the set temperature, This is the preset standard reference temperature.

[0059] Regarding the characteristic parameters in the formula for compensating for heat curing time, the set temperature of the compensating medium is... The value range is set to 40 degrees Celsius to 60 degrees Celsius. Compensation medium set temperature. The steady-state thermal equilibrium temperature corresponding to the rated heating power and coverage area of ​​the selected heating blanket is measured. The action sequence generation unit 420 ensures, through the above calculations, that the final accumulated mechanical properties of the precast components after receiving compensated heating in the ambient temperature buffer zone of the workshop remain equivalent to those of the standard curing process.

[0060] In real-world large-scale factory environments, control systems often face complex situations where multiple production lines simultaneously issue intervention requests. In such cases, the action sequence generation unit 420 needs to prioritize multiple candidate molds for intervention. Considering both logistics efficiency and the quality and safety of prefabricated components, the action sequence generation unit 420 constructs an evaluation operator based on the real-time state parameters of each candidate mold, calculating the mold scheduling priority score for each candidate mold. The calculation logic for the mold scheduling priority score is implemented using the following formula: ; In the formula, Scoring the scheduling priority of the model station. This is a weighting coefficient based on the urgency of time. The congestion warning time threshold, For blocking margin, This is the demolding safety weighting coefficient. To allow for early demolding tolerance, This is the remaining threshold for the equivalent age.

[0061] Time urgency weighting coefficient Weighting coefficient for demolding safety All parameters are dimensionless, with time urgency as the weighting factor. The value range is set to 0.6 to 0.8, which is the demolding safety weighting coefficient. The value range is set to 0.2 to 0.4, and the time urgency weighting coefficient is... Weighting coefficient for demolding safety The sum is 1. Time urgency weighting coefficient. Weighting coefficient for demolding safety The setting is based on giving higher scheduling response tilt to macroscopic anti-blocking requirements, while taking into account the degree of similarity of microscopic thermodynamic states.

[0062] After prioritizing, the action sequence generation unit 420 generates specific equipment scheduling sequences according to the mold scheduling priority scores from high to low. The equipment scheduling sequences include crane transfer instructions and heating blanket start / stop instructions. To achieve closed-loop linkage between the information layer and the physical execution layer, the action sequence generation unit 420 maps and distributes the equipment scheduling sequences to the corresponding register addresses of the underlying programmable logic controller and the warehouse execution mechanism via industry-standard communication protocols such as Profinet or Modbus TCP. This coordinates the physical equipment in the workshop to complete demolding and thermal compensation operations, helping to prevent production line mechanical lock-up and maintain the dynamic balance of production line logistics.

[0063] The constraint optimization module runs on the manufacturing execution system server. It receives the equipment scheduling sequence and compensation thermal curing time output by the scheduling decision module, and seeks the optimal balance between overall system energy consumption and logistics efficiency under multi-dimensional physical boundary conditions. Specifically, the constraint optimization module includes an objective function construction unit 510 and a heuristic search unit 520.

[0064] In actual factory environments, concurrent execution of multiple devices can easily lead to a surge in energy consumption. During the coordinated process of demolding and compensatory heating of precast components, the instantaneous power aggregation of multiple heating blankets and the frequent start-stop of heavy cranes pose a risk of exceeding the load limit of the workshop power grid, potentially causing grid overload or protective power outages. To seek a trade-off between energy consumption and efficiency, the objective function construction unit 510 establishes a comprehensive evaluation cost function for global bus energy consumption and production cycle, mapping time cost and energy consumption together into a unified cost. The logic for calculating the comprehensive evaluation cost by the objective function construction unit 510 is implemented by the following formula: ; In the formula, To comprehensively evaluate the costs, Energy consumption weighting coefficient For the self-discrete task number equal to 1 to The summation operator, This represents the total number of device tasks currently active. For the first The rated operating power of each device For the first The continuous working time of each device. The time penalty weighting coefficient, To find the maximum value function, For the first The final completion timestamp of each device.

[0065] In the formula parameters of the above comprehensive evaluation cost, the energy consumption weight coefficient is... With time penalty weighting coefficient All are dimensionless constants. Energy consumption weighting coefficient The value range is set to 0.4 to 0.6, and the time penalty weight coefficient is... The value range is set to 0.4 to 0.6, and the energy consumption weighting coefficient λ_1 and the time penalty weighting coefficient are... The sum is strictly equal to 1. Energy consumption weighting coefficient. With time penalty weighting coefficient The specific value is dynamically adjusted based on the current electricity billing cycle of the factory. For example, the energy consumption weighting coefficient may be appropriately increased during peak electricity billing periods. This forces the system to adopt a conservative strategy of lengthening the scheduling cycle to reduce concurrent power.

[0066] After establishing the optimization objective of minimizing the comprehensive evaluation cost, the objective function construction unit 510 simultaneously injects the maximum power supply capacity of the workshop, the physical collision avoidance safety distance of the cranes, and the upper limit of the total amount of available heating materials as hard constraints into the solution space. If a certain scheduling combination is predicted to cause the instantaneous total power of the workshop to exceed the maximum power supply capacity of the workshop, or to cause geometric interference between two cranes in the same three-dimensional coordinate system, the objective function construction unit 510 directly assigns an infinite value to the comprehensive evaluation cost of the corresponding scheme, thereby eliminating infeasible schemes that do not meet the physical boundary conditions in the subsequent iterative optimization process.

[0067] Faced with a constraint solution space containing numerous discrete variables and exhibiting high nonlinearity, traditional exhaustive search methods suffer from exponentially increasing computational complexity, making it difficult to find the optimal solution within a limited scheduling period. Therefore, the heuristic search unit 520 employs a particle swarm optimization algorithm based on multi-agent cooperation for iterative solution.

[0068] At the implementation level, the position vectors of particles in continuous space are sorted according to their numerical values, and then mapped to discrete device scheduling execution sequences. During algorithm evolution, each virtual particle represents a potential combination of device scheduling time arrangements. To avoid getting trapped in local minima and accelerate optimization convergence, the heuristic search unit 520 combines historical exploration experience with swarm optimization information to continuously update the search velocity of the entire particle swarm in the solution space, and synchronously calculates the new positions of the particles based on the updated velocity. The logic for calculating the particle velocity update by the heuristic search unit 520 is implemented by the following formula: ; In the formula, For the updated particle velocity, This is the inertia weighting coefficient. The current particle velocity, For individual learning factors, A random number between 0 and 1. This represents the optimal position in the history of an individual particle. This is the particle's current position. As a global learning factor, Another random number between 0 and 1 This represents the best historical position for the entire population.

[0069] Regarding the hyperparameters in the particle velocity update formula, the inertia weight coefficient... The value range is set to 0.5 to 0.9, for the individual learning factor. With global learning factor The value range is typically set between 1.5 and 2.0. Inertia weighting coefficient. In the early stages of algorithm iteration, larger values ​​are used to enhance the global space exploration capability, while in the later stages of iteration, the values ​​are linearly decreased to improve the local search accuracy.

[0070] After multiple rounds of dynamic iteration and reaching the maximum preset number of iterations, the search process converges. The heuristic search unit 520 extracts the optimal equipment scheduling sequence corresponding to the minimum comprehensive evaluation cost from the population. The constraint optimization module finally transforms the optimal equipment scheduling sequence into low-level machine code containing specific timestamps and equipment numbers, which is then transmitted to the registers of the underlying physical actuators via industrial Ethernet network nodes. This helps to achieve efficient flow of manufacturing resources and lean control of workshop energy consumption while ensuring the thermodynamic performance of prefabricated components.

[0071] The execution feedback module runs on the manufacturing execution system server, receives the execution results of the underlying machine code issued by the constraint optimization module, and performs closed-loop status monitoring of the prefabricated components in the ambient temperature buffer area. Specifically, the execution feedback module includes an execution status tracking unit 610 and a dynamic compensation correction unit 620.

[0072] To prevent operational deviations caused by mechanical failures or communication delays when executing the optimal equipment scheduling sequence, the execution status tracking unit 610 collects real-time operational feedback messages from the crane and heating blanket via industrial Ethernet. In the specific monitoring mechanism, the execution status tracking unit 610 compares the command issuance timestamp with the actual equipment response timestamp parsed from the status word in the feedback message to calculate the hysteresis delay of the action execution. Due to inherent network jitter and mechanical start-stop inertia in industrial settings, to avoid false alarms, the system directly triggers a hardware-level audible and visual alarm and suspends subsequent transfer commands during this period when the action hysteresis delay exceeds a preset safety grace period, thus preventing physical collisions. In this embodiment, the safety grace period is set to a range of 2 to 5 seconds, and its specific value is determined based on a comprehensive calibration of the maximum transmission jitter cycle of the underlying control bus and the engagement delay time of the AC contactor.

[0073] In addition to physical monitoring of the mechanical action execution status, verifying the actual thermodynamic curing effect is also a core aspect of controlling process quality. Due to dynamic fluctuations in workshop ambient temperature and physical differences in the adhesion of heating blankets, the actual temperature rise curve of precast components during the compensation stage often deviates from the theoretical expectation. To quantify this deviation, the dynamic compensation correction unit 620 continuously reads the actual temperature sequence returned by the data acquisition module and calculates the deviation between the actual thermodynamic integral and the theoretical compensation target. The logic for calculating the maturity compensation gap by the dynamic compensation correction unit 620 is implemented by the following formula: ; In the formula, To compensate for the maturity gap, The equivalent remaining age threshold calculated by the preceding module. The self-discrete compensation cycle number equal to 1 to The summation operator, Self-discrete compensation cycle number. This represents the total number of valid sampling cycles since the self-starting compensated heating began. The activation energy for the apparent hydration reaction. This is the universal gas constant. For the first The component compensation core temperature measured during each compensation cycle. The preset standard reference temperature, This represents the discrete sampling step size.

[0074] Based on the above deviation calculation results, the system needs to determine whether intervention is necessary. When a maturity compensation gap is detected to be greater than zero and the current physical time has reached the preset compensation heat curing time, it indicates that the actual environmental heat loss is too large, and the original plan has failed to meet the critical requirements for the demolding strength of the precast components. To fill this maturity gap, the dynamic compensation correction unit 620 activates the adaptive correction logic based on proportional control to calculate the additional compensation time that needs to be extended. The logic for calculating the additional compensation time by the dynamic compensation correction unit 620 is implemented by the following formula: ; In the formula, Additional compensation time, This is the temperature response gain coefficient. To compensate for the maturity gap.

[0075] Regarding the characteristic parameter in the additional compensation time formula, the temperature response gain coefficient This is a dimensionless conversion constant, with a value range of 1.2 to 1.5. Temperature response gain coefficient. The specific value is derived from the average cooling rate of the workshop in historical months. In the actual heat transfer process, by introducing a gain coefficient greater than 1, the system can effectively overcome the problem of heat transfer efficiency decay caused by the reduction of the temperature difference between the inside and outside of the component at the end of the heating process. After the calculation is completed, the dynamic compensation correction unit 620 generates updated heating blanket control instructions and reissues them to the underlying programmable logic controller.

[0076] Through the above-mentioned correction mechanism, the execution feedback module continuously performs rolling calculations until the maturity compensation gap approaches zero or is less than zero, and finally completes the closed-loop control of the entire thermal compensation process, ensuring the consistency of the mechanical properties of precast components under complex working conditions.

[0077] Specific application examples: See attached document Figure 3 The present invention provides a scheduling and control system for a beam and slab prefabrication production line based on an industrial internet platform, which may include the following specific application implementation steps: The data acquisition module assesses the initial logistics status of the upstream pouring station on the current production line as normal flow and sets the congestion margin of the upstream station, a target control parameter, to a safe range. The feedforward intervention judgment unit 410 within the scheduling decision module reads the preset congestion warning time threshold, with a standard setting of 15 minutes. The data acquisition module obtains a time-dimensional matrix composed of the remaining number of buffer positions, the number of queuing modules, and the average processing time per station, representing three-dimensional bottom-level state characteristic data with values ​​of 1, 4, and 15. The data acquisition module reads the system's built-in queuing penalty weight coefficient, set to 1.2. The data acquisition module calculates the product of the number of queuing modules, the average processing time, and the penalty weight, and, combined with the physical capacity of the remaining buffer positions, calculates the current upstream station's congestion margin value as 12 minutes based on the congestion margin calculation formula. The feedforward intervention judgment unit 410 compares and finds that the congestion margin value of 12 minutes is lower than the congestion warning time threshold of 15 minutes. The system then generates a state machine switching signal to put the production line into a scheduling interruption state.

[0078] After entering the scheduling interruption state, the data acquisition module continuously collects the internal temperature sequence of the target mold at a fixed sampling frequency of 50Hz. The data acquisition module extracts the microscopic thermodynamic data within the time window and calculates the current mature equivalent age value as 44h using the discrete numerical integration method. The pre-set standard target age within the system is 48h. The data acquisition module calculates the remaining equivalent age threshold as 4h. The system's preset early demolding tolerance value is 4.8h. Since the remaining equivalent age threshold of 4h falls within the allowable range of the early demolding tolerance of 4.8h, the action sequence generation unit 420 immediately generates a forced kiln discharge command and sends it to the constraint optimization module to execute the early demolding operation, recording the current system running time of 120min as the baseline intervention time.

[0079] After early demolding, the feedback module accesses the resource assessment module and reads the rated heating power of the currently associated heating blanket equipment as 800W. The system's preset compensation medium temperature for the target precast component is 50℃. Based on the exponential calculation formula of the equivalent age conservation principle, the action sequence generation unit 420 calculates the compensation heat curing time for compensating for early kiln exit losses to be 90 minutes.

[0080] While recording the baseline intervention time, the execution feedback module drives the heating blanket to perform active thermal compensation. During the actual thermal compensation period, the dynamic compensation correction unit 620 extracts the actual temperature feedback data within 60 minutes after the baseline intervention time and calculates that the maturity compensation gap caused by the intrusion of cold air in the workshop is 15 hours. The dynamic compensation correction unit 620 reads the temperature response gain coefficient value as 1.2 and calculates the additional compensation time that needs to be extended as 18 minutes based on the adaptive correction logic. The dynamic compensation correction unit 620 marks the heating blanket as an active compensation device and sets the target heating power of the device to rise to 1200W in the correction interval.

[0081] Combined with appendix Figure 3 The horizontal axis represents time, and the vertical axis represents physical quantity output. The feedback module performs cross-regulation of equipment status within a time interval of 175 minutes to 195 minutes. (Appendix) Figure 3 The horizontal axis represents the running time, in minutes. (See attached image) Figure 3 The vertical axis on the left represents the heating power of the heating blanket, in watts (W). (See attached image.) Figure 3 The vertical axis on the right represents the temperature of the component's compensation core, in °C. (See attached image.) Figure 3 The thick solid line represents the change in the heating power of the heating blanket, and the thick dashed line represents the change in the core temperature of the component compensation device. Within the interval from 120 minutes to the baseline control time of 180 minutes, the heating power of the active compensation device remains at 800W, and the core temperature of the component compensation device slowly decreases from 50℃ to 42℃ due to interference from the ambient cold source. When the current operating time variable enters the time overlap window period of 180 minutes to 190 minutes, the dynamic compensation correction unit 620 calculates and issues control commands based on the power compensation control algorithm, causing the attached... Figure 3 The heating power of the heating blanket, represented by the thick solid line, increases linearly from 800W to 1200W. This is simultaneously controlled by the injection of additional heat flux density. Figure 3The component compensation core temperature, represented by the thick dashed line, rises non-linearly exponentially from 42℃ to 50℃. Within the interval 190 minutes after the current operating time variable reaches the end of the time overlap window until the end of the compensation cycle, the heating blanket's heating power stabilizes at 1200W, and the component compensation core temperature stabilizes at 50℃. The system-driven underlying thermodynamic intervention equipment completes the gradient increase of physical output power and the thermal equilibrium reconstruction of the component's internal temperature within 10 minutes.

[0082] Experimental verification and effect comparison section: See attached document Figure 4 This invention provides a beam and slab precast production line scheduling and control system based on an industrial internet platform, which may include a step of extracting randomized controlled test data from 20 independent precast mold tables. The test randomly and evenly divides the 20 precast mold tables into a fixed-cycle demolding group and a dynamic feedforward scheduling group. (See attached document.) Figure 4 The horizontal axis represents the system's continuous operating time, in minutes. (See attached image) Figure 4 The vertical axis represents the upstream station congestion margin calculated by the system, in minutes. (See attached image) Figure 4 The solid line in the figure represents the trend of the blocking margin value of the fixed-cycle demolding group. (Appendix) Figure 4 The dashed line in the figure represents the trend of the congestion margin value of the dynamic feedforward scheduling group. The fixed-cycle demolding group system locks the standard target age as the sole demolding criterion and refuses to execute any early intervention commands during the complete 400-minute test cycle. The dynamic feedforward scheduling group system loads the linkage control program of each module, and executes early kiln exit and cross-regional resource scheduling actions based on equivalent age compensation when the congestion warning time threshold is triggered.

[0083] In the appendix Figure 4 During the 0-120 minute operating period, the solid and dashed lines completely overlap. The congestion margin values ​​for both the fixed-cycle demolding group and the dynamic feedforward scheduling group decreased linearly from the initial 45-minute value to the 15-minute value at 120 minutes as the pouring operation continued and the molds physically accumulated. When the timeline reached 120 minutes, the two groups exhibited different scheduling strategies and system state feedback patterns. The fixed-cycle demolding group continued its mechanized waiting logic; its congestion margin value crossed the warning line at 120 minutes and returned to zero at 165 minutes. Subsequently, in the 165-400 minute time interval, the solid line of the fixed-cycle demolding group showed extremely low periodic stagnation, with its congestion margin value fluctuating repeatedly within the extremely congested range of 0-5 minutes, indicating a substantial mechanical deadlock and production halt in the production line.

[0084] The dynamic feedforward scheduling group triggers a 15-minute congestion warning time threshold at 120 minutes. The system records this moment as the baseline intervention time and initiates forced kiln unloading and reconfiguration actions for the target mold. (Combined with appendix...) Figure 4 The dashed line shows the trajectory of the dynamic feedforward scheduling group's blocking margin value during the transition period of early demolding stage buffer space release. From 15 minutes, it rises sharply to 38 minutes at 140 minutes. At 140 minutes, all bottom-level transfer operations are completed, and with the intervention of new idle molds, the blocking margin value represented by the dashed line completely escapes the danger zone approaching zero. Subsequently, within the time interval of 140 minutes to 400 minutes, the dashed line presents a sawtooth-shaped healthy cycle curve. The blocking margin value of the dynamic feedforward scheduling group consistently oscillates dynamically within the safe range of 25 minutes to 42 minutes. Experimental data verifies that the multi-module cascaded feedforward control and equivalent aging compensation mechanism adopted in this system can effectively avoid the risk of physical production line deadlock and guide the system's material flow buffer capacity to achieve substantial recovery.

Claims

1. A scheduling and control system for a precast beam and slab production line based on an industrial internet platform, characterized in that, include: The data acquisition module acquires the heterogeneous data set of the production line associated with the industrial internet platform, encapsulates the heterogeneous data set of the production line according to the protocol, and then synchronously uploads it to the manufacturing execution system server. The state mapping module receives discrete logistics data and continuous thermodynamic data of the production line transmitted by the industrial internet platform, converts the discrete logistics data and continuous thermodynamic data of the production line into digital control variables, and calculates the remaining threshold of the equivalent age of the production line mold and the blockage margin of the upstream pouring station. The resource assessment module combines the heterogeneous data set with the remaining equivalent age threshold of the production line module to calculate the maximum equivalent age compensation limit of the production line under the current environment. The scheduling decision module monitors the congestion margin of the upstream pouring station in real time, controls the switching of the system scheduling state machine, and generates the forced kiln exit command and priority sequence of the target mold when the tolerance constraint is met. The constraint optimization module receives and parses the forced kiln discharge command, combines it with the production line equipment parameters issued by the industrial internet platform, and outputs the motion speed control command of the underlying machinery. The execution feedback module executes the production line mold table hoisting and transfer action according to the motion speed control command, and at the same time reports the physical compensation command execution parameters and the downstream insulation resource consumption status, forming a global control closed loop for the production line. The specific steps performed by the state mapping module are as follows: The equivalent age calculation unit and the blockage risk quantification unit included in the state mapping module work together to introduce the equivalent age calculation unit into the thermodynamic maturity model. Using the discrete numerical integration method, the internal temperature sequence is accumulated periodically along the time axis to calculate the current mature equivalent age of the beam and slab precast component. The preset standard target age is subtracted from the mature equivalent age to obtain the remaining threshold of the equivalent age of the production line mold. The congestion risk quantification unit extracts the current number of queuing modules, the remaining number of buffer positions, and the average processing cycle time per station. It quantifies the physical capacity of the production line to handle the current process accumulation. After calculation and removing negative numbers through the maximum value function, it outputs the congestion margin of the upstream station. The remaining threshold of the equivalent age of the production line mold and the blockage margin of the upstream workstation are converted into digital control variables for use by the scheduling decision module and the resource evaluation module. The resource assessment module includes a thermodynamic demand prediction unit and a supply and demand matching verification unit. The thermodynamic demand prediction unit quantifies the additional thermal resistance required to maintain the safety of the internal and external temperature difference of the component based on Newton's law of cooling and Fourier's law of thermal conductivity. The supply and demand matching verification unit compares the thermal compensation resource parameters with the additional thermal impedance and active compensation power to determine the resource matching status, and calculates the maximum equivalent age compensation limit of the production line based on the parameters.

2. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform as described in claim 1, characterized in that, The data acquisition module performs the following steps: The data acquisition module includes a physical state acquisition unit, a production line topology synchronization unit, and an environment and resource perception unit, which work together to acquire multiple types of data. The physical state acquisition unit acquires the internal temperature sequence collected by the internal temperature measurement nodes of each flow mold in the production line. The production line topology synchronization unit establishes a data interface with the manufacturing execution system server through the industrial internet platform, reads the current number of queued molds, the remaining number of buffer positions, the average processing time per workstation, and the expected arrival timestamp of the most recent idle mold to the pouring workstation, and converts them into a unified time dimension matrix. The environment and resource sensing unit acquires environmental data of the production line workshop and resource parameters of thermal insulation blankets and heating blankets, and encapsulates them into a one-dimensional thermal compensation resource vector. The internal temperature sequence, the time dimension matrix, the production line workshop environmental data, the resource parameters, and the one-dimensional thermal compensation resource vector are timestamped and standardized to generate the production line heterogeneous data set. After protocol encapsulation, the production line heterogeneous data set is synchronously uploaded to the manufacturing execution system server.

3. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform as described in claim 1, characterized in that, The specific steps executed by the scheduling decision module are as follows: The scheduling decision module includes a feedforward intervention judgment unit and an action sequence generation unit, which verifies the upstream workstation blocking margin, the equivalent age remaining threshold of the production line module, and the resource matching status, and outputs an intervention activation signal when the conditions are met. The action sequence generation unit responds to the intervention activation signal, calculates the compensation heat curing time and the mold scheduling priority score, and generates the forced kiln exit command and the priority sequence according to the score.

4. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform as described in claim 3, characterized in that, The action sequence generation unit calculates the priority score as follows: The score is obtained by multiplying the difference between the congestion warning time threshold and the upstream workstation congestion margin by the time urgency weighting coefficient, and multiplying the difference between the early demolding tolerance and the production line mold equivalent age remaining threshold by the demolding safety weighting coefficient.

5. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform according to claim 1, characterized in that, The constraint optimization module performs the following steps: The constraint optimization module includes an objective function construction unit and a heuristic search unit, which establishes a comprehensive evaluation cost function of global energy consumption and production cycle of the production line and injects hard constraints. The heuristic search unit uses a particle swarm optimization algorithm to iteratively solve the problem, extract the optimal equipment scheduling sequence, parse and convert it into the underlying mechanical motion speed control command, and output it.

6. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform according to claim 5, characterized in that, The iterative solution steps of the particle swarm optimization algorithm are as follows: The virtual particle position vector is mapped to the production line equipment scheduling execution order. The particle position and velocity update parameters are calculated, and the process converges after iterating to a preset number of times. The optimal scheduling sequence is then extracted and converted into underlying machine code.

7. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform according to claim 1, characterized in that, The execution feedback module performs the following steps: The execution feedback module includes an execution status tracking unit and a dynamic compensation correction unit, which drives the crane and transfer equipment to perform the platform action, collects equipment operation feedback and calculates action hysteresis delay, and triggers an alarm and suspends the instruction when the threshold is exceeded. The dynamic compensation correction unit calculates the maturity compensation gap, reports the execution parameters of the physical compensation command and the consumption status of the downstream insulation resources, and updates the resource library to form the global control closed loop.

8. The beam and slab prefabrication production line scheduling and control system based on an industrial internet platform according to claim 7, characterized in that, The correction steps of the dynamic compensation correction unit are as follows: When the maturity compensation gap is greater than zero and the preset compensation heat curing time is reached, the adaptive correction logic is activated to calculate the additional compensation time and issue the updated heating blanket control command. The calculation continues until the maturity compensation gap is no greater than zero.

Images