A kind of small minced meat frying production line parameter collaborative control system and method

Abstract

The application relates to the field of food oil frying control systems, and discloses a small fried meat oil frying production line parameter cooperative control system and method, which comprises a main control programmable logic controller which is combined with three-dimensional point clouds and physical quality data to generate a corrected quality sequence and store the corrected quality sequence into a first-in-first-out shift register queue; position pulses are converted into displacement integrals, a quality sequence is extracted to calculate theoretical heat load power and solve a feedforward limiting speed; target material blocks are determined through forward addressing based on the displacement integrals and heat transfer lag time, a causal prediction feedforward heat power is deduced, and a final heating valve opening degree instruction is generated by combining actual temperature data to adjust heat input; based on actual temperature, shell forming heat integrals are calculated, and ripeness compensation speed limit values are generated by combining finished product ripeness characteristic data, the feedforward limiting speed and the ripeness compensation speed limit values are arbitrated to generate a mesh belt target running speed instruction. The scheme solves the heat imbalance problem caused by load fluctuation and guarantees the consistency of batch output material ripeness.

Description

Technical Field

[0001] This invention relates to the field of food frying control systems, specifically to a collaborative control system and method for parameters of a small crispy pork frying production line. Background Technology

[0002] In the continuous, large-scale production of crispy fried pork, the equipment typically operates at a fixed conveyor belt speed. When faced with a sudden surge in feed volume at high loads, the large influx of room-temperature material absorbs a significant amount of heat, easily exceeding the physical limit of the heating device's transient heating capacity. This mismatch between the feed heat load and the system's maximum heating capacity often leads to a sharp drop in oil temperature at the front end of the production line, disrupting the stability of the basic heat supply.

[0003] Existing temperature control methods mostly rely on a single closed-loop feedback control mechanism. Due to the inherent physical hysteresis of the frying medium during heat transfer, by the time the temperature sensor detects a drop in oil temperature and issues an opening adjustment command, heat compensation has usually already become physically delayed. Post-hoc adjustments based on real-time temperature deviations are ill-suited to handling continuous and fluctuating heat load shocks. During compensation, this can easily lead to integral saturation of the control loop and temperature overshoot in the later stages of heating, thus reducing the accuracy of multi-zone temperature control.

[0004] Furthermore, there is often a conflict in parameter adjustments between the physical limits of the production line's heating capacity and the residence time required for the material to complete gelatinization and shaping. Existing continuous frying equipment lacks a collaborative feedback mechanism for the actual physical state of the finished product, and cannot dynamically adjust the conveyor belt speed based on the actual accumulated heat energy absorbed by the material during the heating process, ultimately resulting in poor consistency in the doneness of batch-produced materials. At the same time, in extreme heat deficit conditions, existing production lines struggle to proactively intervene in the feed load at the front end, and lack a mechanism for targeted physical isolation of substandard products. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a collaborative control system and method for parameters of a crispy pork frying production line, which solves the problem of heat imbalance in multiple temperature zones and inconsistent finished product cooking caused by fluctuations in material input load in existing continuous frying production lines.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The first aspect of the present invention provides a collaborative control system for parameters of a fried pork crisp production line, including a main programmable logic controller, a three-dimensional laser profilometer, a dynamic weighing device, a mesh belt drive module, a multi-temperature zone frying tank, and a quality detection module.

[0008] The main programmable logic controller divides the effective physical length of the continuously running material feeding conveyor belt into multiple discrete blocks according to a fixed size interval;

[0009] The actual position pulses fed back by the mesh belt drive module are converted into external trigger synchronization signals, which drive the three-dimensional laser profilometer set above the mesh belt to perform line laser scanning at fixed physical displacement intervals. Simultaneously, the dynamic weighing device set in the bottom support structure is used to obtain the surface spatial coordinate data and actual physical mass data of the material.

[0010] The main control unit reconstructs the discrete distribution volume of materials based on spatial coordinate data, and derives the dynamic equivalent density by combining physical mass data. It then calculates the specific mass values ​​of discrete blocks through multiplication operations, constructing a first-in-first-out shift register queue that maps to physical spatial displacement. This data queue synchronously updates the block coordinates as the absolute physical displacement integral of the conveyor belt accumulates, establishing a spatial distribution data model of the production line's feed load.

[0011] This system sets up a block sliding window at the front end of the frying tank inlet in the downstream area to extract the local material load, rated linear velocity and size spacing, and calculates the theoretical heat load power by combining the total unit mass heat energy absorbed by the material entering the oil.

[0012] The system reads the opening data of the proportional heating valves set in the heating circuits of each physical temperature zone in real time, and calculates the net available compensation power limit by combining the difference between the maximum rated thermal power of the temperature zone and the current output thermal power. When the input heat load demand exceeds the system's maximum heating capacity, the main control device uses the net available compensation power limit and the sum of the absolute heat demand to inversely calculate the feedforward limiting speed constrained by power. By reducing the conveyor belt feed speed, peak-shaving control of the feed heat load is implemented to prevent the oil temperature in the front-end area of ​​the system from deviating from the process set value due to heat overdraft.

[0013] To compensate for the physical hysteresis characteristics of heat transfer, this system uses a rotary encoder in the mesh belt drive module to obtain physical displacement and update data coordinates. It combines the heat transfer hysteresis time constant of each independent physical temperature zone, the current mesh belt running speed, and the block coordinates to perform forward addressing, thereby determining the material block that is about to enter the target temperature zone in advance. Based on the extracted mass value sequence and the staged unit mass heat energy, the system infers the causal prediction feedforward heat power corresponding to each temperature zone.

[0014] The causal predictive feedforward thermal power is converted into an opening increment command through an inverse mapping function, and then algebraically superimposed with the proportional-integral-derivative (PID) basic feedback control command calculated based on actual data acquired from temperature sensors. The system employs a saturation constraint function to impose numerical boundary constraints on the superposition result. When the superimposed command value exceeds the physically permissible maximum opening or falls below the minimum opening boundary, the system executes a numerical clamping assignment and triggers an integral freeze command to suspend the accumulation of the integral term, preventing control integral saturation.

[0015] In the material discharge detection stage, the system performs time integral calculations by combining the effective part of the temperature difference between the actual oil temperature and the physical gelatinization initiation threshold and the shaping reaction rate constant during the material residence period, and constructs the shell forming heat integral that reflects the degree of material maturation.

[0016] By combining the finished product characteristic data acquired by the discharge detection camera with the molded material, the system corrects the preset basic process target to obtain the standard maturity target value, and calculates the maturity compensation speed limit based on the deviation between the heat integral and the target value. The main control device compares the feedforward limit speed and the maturity compensation speed limit, and extracts the smaller value as the target running speed command.

[0017] When the speed command is lower than the set minimum process line speed, the system outputs the inverter running speed locked by the bottom line to the mesh belt drive module, and simultaneously sends a limit command to the front feeder to reduce the heat load of the feed. It also marks the corresponding material with a defect traceability tag in the data queue, and links the end sorting mechanism to perform directional interception and defective product diversion.

[0018] The second aspect of the present invention provides a method for coordinated control of parameters of a fried pork frying production line, which is applied to the coordinated control system provided in the first aspect of the present invention.

[0019] Based on the three-dimensional spatial distribution of materials and the results of mass feedback, this method generates a corrected mass sequence at discrete displacement nodes and constructs a feed load distribution model.

[0020] The theoretical heat load is calculated by extracting the mass sequence within the sliding window. When the heating power is limited, the feedforward limit rate is calculated by comparing it with the net available compensation power limit to limit the feed heat load.

[0021] The position pulse is converted into displacement integral, and the target material is determined by mapping and deduction based on the heat transfer hysteresis time and block coordinates. The staged feedforward heat power is deduced and cross-coupled with the feedback loop of actual temperature to generate heating valve opening command to regulate heat input.

[0022] The heat integral of shell forming is calculated based on the actual temperature, and the temperature equivalent closed-loop intervention of the maturity compensation speed limit is combined with the finished product inspection data. The feedforward limit speed and the maturity compensation speed limit are arbitrated and extracted, and the final mesh belt running speed command is output.

[0023] This invention provides a collaborative control system and method for parameters of a fried pork crisp production line. It has the following beneficial effects:

[0024] 1. This invention utilizes data collected by a three-dimensional laser profilometer and a dynamic weighing device to construct a first-in-first-out shift register queue for discrete material blocks, extracts the corrected mass sequence within the sliding window to calculate the theoretical heat load power, and combines the net available compensation power limit of the heating module to solve the feedforward limiting speed. This achieves active peak shaving control of the feed heat load, avoiding a significant drop in front-end oil temperature due to insufficient transient heating capacity of the heating device under sudden high-load feeding conditions, thus ensuring the stability of the basic heat supply of the production line.

[0025] 2. This invention combines the absolute physical displacement integral with the heat transfer lag time of each physical temperature zone to perform forward addressing, predetermines the target material block and extrapolates the staged causal prediction feedforward heat power, and then converts the feedforward heat power into an opening increment command and performs algebraic superposition and boundary constraints with the basic feedback control command generated based on the temperature deviation. This realizes the cross-coupling adjustment of the heating control loop, compensates for the physical lag characteristics of the heating medium in the heat transfer process, prevents the integral saturation and temperature overshoot problems that are easy to occur in traditional closed-loop control, and improves the accuracy of multi-temperature zone temperature control.

[0026] 3. This invention calculates the heat integral of shell forming during the material residence cycle, and combines it with finished product maturity characteristic data to correct the basic process target to calculate the maturity compensation speed limit. Logical arbitration is performed between the feedforward limit speed and the maturity compensation speed limit to output the final mesh belt target running speed command. When the speed triggers the minimum process limit, the feeding reduction and end-of-line defective product diversion are linked, establishing a system collaborative closed loop based on the physical state of the finished product. This solves the parameter adjustment conflict between the physical limits of the production line's heating and the required shaping time of the material, ensures the consistency of the maturity of the batch-produced material, and provides a means of physical isolation of defective products in abnormal operating conditions. Attached Figure Description

[0027] Figure 1 This is a hardware topology diagram of the collaborative control system for the parameters of the crispy pork frying production line of the present invention.

[0028] Figure 2 This is a flowchart of the collaborative control method for parameters of the crispy pork frying production line of the present invention;

[0029] Figure 3 This is a schematic diagram of the feedforward peak-shaving speed control principle of the present invention;

[0030] Figure 4 This is a schematic diagram of the multi-temperature zone causal feedforward mapping deduction logic of the present invention;

[0031] Figure 5 This is a schematic diagram of the anti-saturation feedforward feedback cross-coupling heating control principle of the present invention;

[0032] Figure 6 This is a data curve diagram showing the trend of feed heat load changing over time under sudden high load conditions according to the present invention;

[0033] Figure 7 This is a comparison curve of the actual oil temperature changes in the first temperature zone of the traditional PID control and the collaborative control under the corresponding sudden high load conditions of the present invention.

[0034] Figure 8 This is a comparative data curve of the linear velocity change of the conveyor belt under the corresponding sudden high load conditions of the present invention, showing the traditional constant speed mode and the adaptive feedforward speed limiting mode. Detailed Implementation

[0035] 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.

[0036] See attached document Figure 1 This invention provides a collaborative control system for parameters of a fried pork production line, comprising: a material feeding conveyor belt, a three-dimensional laser profilometer, a dynamic weighing device, a main control programmable logic controller, a conveyor belt drive module, a multi-temperature zone frying tank, a heating adjustment module, and a quality detection module.

[0037] A 3D laser profilometer is positioned above the material feeding conveyor belt to collect 3D point cloud data of the material. A dynamic weighing device is installed in the bottom support structure of the material feeding conveyor belt to obtain actual physical mass data. The conveyor belt drive module is used to drive the material feeding conveyor belt. The conveyor belt drive module includes a rotary encoder mounted on the physical drive shaft and a conveyor belt drive frequency converter that controls the conveyor belt speed. The rotary encoder transmits the collected actual position pulses to the main control programmable logic controller (PLC). The main control PLC issues frequency converter speed commands to the conveyor belt drive frequency converter according to its internal logic, forming a closed-loop control circuit.

[0038] The multi-temperature zone fryer is located downstream of the material feeding conveyor belt, and is divided into multiple independent physical temperature zones along the material conveying direction. Each physical temperature zone is equipped with a temperature sensor to monitor the actual temperature data. The heating regulation module includes proportional heating valves corresponding to the heating circuits of each physical temperature zone. These proportional heating valves adjust the heat input of their respective zones according to the final heating valve opening command from the main programmable logic controller (PLC). The quality inspection module includes a discharge detection camera located at the discharge end of the multi-temperature zone fryer. The discharge detection camera is aimed at the formed material leaving the multi-temperature zone fryer to collect finished product cooking characteristic data and feed it back to the main PLC.

[0039] The system is externally configured with a front-end feeder and an end-end sorting mechanism. The main control programmable logic controller (PLC) is interconnected bidirectionally with a 3D laser profilometer, dynamic weighing device, belt drive frequency converter, temperature sensor, proportional heating valve, and discharge detection camera via industrial Ethernet protocol. The input and output nodes of the main control PLC are respectively connected to the drive start / stop control terminal of the front-end feeder and the execution terminal of the end-end sorting mechanism, realizing coordinated control from the feeding end to the discharging end.

[0040] See attached document Figure 2 This invention provides a method for coordinated control of parameters in a fried pork frying production line, comprising the following steps:

[0041] S10. Feed Heat Load Array Reconstruction. The main control programmable logic controller (PLC) receives the actual position pulse from the rotary encoder, triggering a 3D laser profilometer to perform point cloud scanning on the material surface of the conveyor belt, acquiring 3D point cloud data of the material, and calculating the discrete distribution volume of the material for each discrete material block based on the 3D point cloud data. Simultaneously, it collects the actual physical mass data acquired by the dynamic weighing device and calculates and generates a dynamic equivalent density coefficient. Combining the discrete distribution volume of the material with the dynamic equivalent density coefficient, it generates a corrected mass sequence for each discrete material block. The main control PLC internally constructs a first-in-first-out (FIFO) shift register queue corresponding to the material feed conveyor belt position, and sequentially stores the generated corrected mass sequences into the FIFO shift register queue.

[0042] S20. Feedforward Peak Shaving Speed ​​Limit Calculation. To address the transient heat load impact caused by materials entering the FIFO shift register queue, the main programmable logic controller (PLC) defines a block sliding window corresponding to the inlet section of the multi-temperature zone fryer in the FIFO shift register queue. The corrected mass sequence within the block sliding window is extracted, and the theoretical heat load power corresponding to the corrected mass sequence entering the oil at the rated linear velocity is calculated. The main PLC synchronously reads the real-time opening data of the proportional heating valves within the physical temperature zone corresponding to the inlet section of the multi-temperature zone fryer, and calculates the net available compensation power limit at the current moment based on the design maximum rated heat power of the corresponding physical temperature zone. When the theoretical heat load power exceeds the net available compensation power limit, the feedforward speed limit that matches the heating load is calculated based on the energy conservation limit.

[0043] S30. Multi-temperature zone causal feedforward mapping derivation. As the conveyor belt drive frequency converter dynamically adjusts the conveyor belt's operating state, the main control programmable logic controller reads the actual position pulses fed back by the rotary encoder and converts them into absolute physical displacement integrals. The absolute physical displacement integrals are used to update the block coordinates of the first-in-first-out shift register queue. Based on the updated block coordinates, and considering the preset heat transfer lag time for each physical temperature zone in the multi-temperature zone frying tank, combined with the actual position pulses and the preset acceleration / deceleration slope parameters of the conveyor belt drive frequency converter, forward addressing is used to determine the target material block that will reach the corresponding physical temperature zone after the heat transfer lag time. The corrected mass sequence corresponding to the target material block is extracted, and the causal prediction feedforward heat power corresponding to each physical temperature zone is calculated based on the corrected mass sequence.

[0044] S40, Cross-coupled Heating Control Execution. Based on the causal predicted feedforward heat power obtained through deduction, the main control programmable logic controller (PLC) acquires the actual temperature data of each physical temperature zone fed back by temperature sensors, and generates basic feedback control commands through proportional-integral-differential (PID) operations. The causal predicted feedforward heat power is converted into an opening increment command through an inverse mapping function. The basic feedback control command and the opening increment command are algebraically superimposed, and a saturation constraint function is applied to constrain the superposition result within the physical working range, generating the final heating valve opening command. The main control PLC outputs the final heating valve opening command to the proportional heating valve of the corresponding physical temperature zone, driving the proportional heating valve to perform heat replenishment for the physical temperature zone;

[0045] S50, Time-Temperature Equivalent Integral Closed-Loop and Arbitration Boundary Intervention. As the proportional heating valve dynamically adjusts the actual temperature data of the oil tank, the material moves within the continuously heated multi-temperature zone frying tank. The main programmable logic controller (PLC) accumulates the actual temperature data of each discrete material block reaching the discharge end of the multi-temperature zone frying tank along the time dimension, combining historical data, to calculate the shell forming heat integral for each discrete material block throughout the entire process. Based on the finished product's maturity characteristic data, the preset basic maturity target is corrected to obtain the standard maturity target value. The integral difference between the standard maturity target value and the shell forming heat integral is calculated, and the residence delay time is calculated based on the integral difference. Finally, a maturity compensation speed limit is generated based on the residence delay time.

[0046] The main programmable logic controller (PLC) arbitrates the speed between the feedforward limiting speed and the ripeness compensation speed limit to generate a target operating speed command. When the target operating speed command falls below the preset minimum process line speed, it sends an inverter operating speed command locked to the minimum process line speed to the conveyor belt drive inverter. Simultaneously, it marks the corresponding block with a defect traceability tag in the first-in-first-out shift register queue and synchronously outputs control signals to the front-end feeder and the end-end sorting mechanism to perform coordinated control of the crispy pork frying production line parameters.

[0047] In conjunction with step S10, the specific implementation process of feed heat load array reconstruction based on multi-source information fusion is described in detail. The specific implementation steps of feed heat load array reconstruction include the following sub-steps:

[0048] S11. Discretization of the physical space of the conveyor belt and synchronous triggering with hardware. The main programmable logic controller (PLC) discretizes the effective physical length of the continuously running material feeding conveyor belt according to fixed dimensional intervals. Divided into multiple discrete material blocks. Size spacing. The speed of the material feed conveyor belt is determined based on the operation cycle of the control system and the reference operating speed of the material feed conveyor belt. It is usually set in the range of mm to cm to balance control accuracy and calculation load.

[0049] After physical spatial discretization, a rotary encoder installed on the physical drive shaft of the material feed conveyor outputs actual position pulses as the conveyor moves. The main programmable logic controller (PLC) receives these pulses and converts them into external trigger synchronization signals, which are then sent to the 3D laser profilometer. Based on the external trigger synchronization signal, the 3D laser profilometer performs line laser scanning at fixed physical displacement intervals, acquiring the 3D spatial coordinate data of the surface material and generating 3D point cloud data corresponding to the current discrete material block. This fixed physical displacement interval is matched to the resolution parameters of the rotary encoder and the pulse counting frequency multiplication setting of the main PLC. This hardware-level synchronization triggering mode based on mechanical displacement reduces the interference of material feed conveyor speed fluctuations on the scanning accuracy along the running direction.

[0050] S12. 3D Point Cloud Analysis and Distribution Volume Extraction. The 3D laser profilometer transmits the acquired 3D point cloud data of the material to the memory unit of the main control programmable logic controller (PLC). The PLC performs a background reference plane filtering operation, identifying and removing background data points flush with the surface of the material feed conveyor belt, retaining valid material point clouds that are above the reference plane. The determination of whether a point is above the reference plane depends on a preset reference plane height threshold, which is calibrated by scanning the material feed conveyor belt multiple times under no-load conditions and extracting the standard deviation of the Gaussian distribution of its surface undulations. The PLC uses a random sampling consensus algorithm or a height threshold segmentation method to perform background filtering and reference plane coordinate system fitting of the 3D point cloud data.

[0051] After background reference plane filtering is completed, the main programmable logic controller (PLC) divides the space into grids and accumulates the product of the projected area of ​​each unit grid corresponding to each data point in the effective material point cloud and the height relative to the reference plane to calculate the discrete distribution volume of the current discrete material block. The projected area per unit grid is calculated using the physical pixel resolution of the 3D laser profilometer and the lens field of view calibration parameters.

[0052] S13. Real-time closed-loop calculation of dynamic equivalent density coefficient. Due to the irregular shape and overlapping of the batter on the surface of the crispy pork during processing, the 3D laser profilometer can only acquire the geometric features of the visible areas on the top and sides of the material within discrete material blocks. To avoid quality measurement deviations caused by relying solely on 3D features, the system introduces a dynamic weighing device located at the bottom to establish a physical quality calibration closed loop. The measuring tray of the dynamic weighing device covers a fixed number of continuous discrete material blocks in physical space and continuously collects the actual physical quality data of the material within these blocks. This fixed number is calculated by dividing the physical length of the measuring tray of the dynamic weighing device by the dimensional spacing. The calculation yielded the results. The moving average filtering logic for the actual physical quality data is implemented based on a time window sliding average algorithm.

[0053] The main programmable logic controller (PLC) is based on moving average filtering logic and calculates the dynamic equivalent density coefficient in real time. The calculation formula is as follows:

[0054] ;

[0055] in, This represents the dynamic equivalent density coefficients output by the solution at the current moment; This represents the actual physical mass data collected by the dynamic weighing device at the current moment; This represents the total number of discrete material blocks covered by the physical measurement tray of the dynamic weighing device along the conveying direction; This indicates the current moment as the th weighing device reaches the end of its range. Sequence index of discrete material blocks at the weighing end; Indicates the first The discrete distribution volume of material corresponding to a discrete material block in a weighing interval; Indicates from the first The discrete material block of the weighing interval to the first The sum of the discrete distribution volumes of the material in each discrete material block at the weighing end.

[0056] To ensure the integrity of the algorithm logic, the main programmable logic controller (PLC) checks the cumulative sum of the discrete distribution volumes of the material before performing the division operation. When this cumulative sum is lower than a preset volume threshold, it indicates that there is no material or too little material in the corresponding section of the current material feeding conveyor belt. The system will then stop updating and maintain the dynamic equivalent density coefficient from the previous valid moment to avoid overflow due to the denominator approaching zero. This preset volume threshold can be determined based on the sensor's background noise calibration.

[0057] S14. Correcting the mass mapping and constructing the data queue. After obtaining the dynamic equivalent density coefficient, the main programmable logic controller (PLC) performs mass conversion calculations on the discrete material blocks whose volumes have been extracted. Specifically, the processor multiplies the calculated dynamic equivalent density coefficient by the discrete distribution volume of the material corresponding to the current discrete material block to obtain the specific mass value for each discrete material block. And determined by the specific quality values ​​corresponding to each discrete material block. Generate the corrected quality sequence corresponding to each discrete material block, where the first... The specific mass value corresponding to each discrete material block at the weighing end is denoted as: .

[0058] After completing the quality conversion calculation, the main programmable logic controller (PLC) constructs a first-in-first-out (FIFO) shift register queue in its internal storage area. The FIFO shift register queue consists of consecutively arranged memory address units, each corresponding to a discrete material block in physical space. The PLC sequentially stores the generated corrected quality sequence into the enqueue port of the FIFO shift register queue. For each discrete material block that the material feed conveyor belt travels the physical length of in the direction of the multi-temperature zone frying tank, the data pointer in the FIFO shift register queue shifts backward once synchronously. The specific quality values ​​of each discrete material block are then recorded. The arrangement order in the digital space maintains a corresponding mapping relationship with the actual distribution state of the materials on the material feeding conveyor belt.

[0059] See attached document Figure 3 In conjunction with step S20, by predicting the heat load demand of the material to be introduced into the oil system, when the predicted demand exceeds the remaining capacity of the heating system, the linear speed of the material feed conveyor belt is actively reduced. The specific implementation steps of the feedforward peak shaving speed regulation limit calculation include the following sub-steps:

[0060] S21. Physical Oil Inlet Zone Mapping and Block Sliding Window Construction. The inlet section of a multi-temperature zone fryer is the area where materials come into contact with high-temperature oil, causing moisture vaporization and temperature rise. When a large batch of materials passes through the inlet section of the multi-temperature zone fryer, it usually triggers a transient thermal load shock, causing the oil temperature to drop. The main programmable logic controller performs mesh belt spatial mapping based on the actual physical length of the inlet section of the multi-temperature zone fryer, dividing the actual physical length of the inlet section of the multi-temperature zone fryer by the dimensional spacing. The number of discrete material blocks corresponding to the inlet section of the multi-temperature zone frying tank is calculated. Based on this, in the first-in-first-out shift register queue, starting from the discrete material block that is about to enter the inlet section of the multi-temperature zone frying tank, a continuous storage address unit containing the number of discrete material blocks is delineated along the opposite direction of material transport, forming a block sliding window.

[0061] Meanwhile, considering the continuous conveying characteristics of the material feed conveyor belt, there is a fixed physical distance between the inlet section of the multi-temperature zone frying tank and the dynamic weighing device. Therefore, the first [item] about to enter the inlet section of the multi-temperature zone frying tank... The first inlet discrete material block and the first [unclear] located at the end of the dynamic weighing device's range Each discrete material block at the weighing end has a fixed sequence address offset in the first-in-first-out shift register queue. The sequence address offset is calculated by dividing the actual physical distance between the inlet section of the multi-temperature zone frying tank and the dynamic weighing device by the size spacing. The calculations yield the precise mapping between physical spatial location and digital queue index offset. The block sliding window slides synchronously with the data pointer shift of the first-in-first-out shift register queue, defining the material data about to be added in real time.

[0062] S22. Theoretical heat load power assessment. The main programmable logic controller extracts the corrected mass sequence within the block sliding window and obtains the specific mass values ​​corresponding to each discrete material block contained within the block sliding window. Since the heat required for the material to enter the inlet section of the multi-temperature zone frying tank consists of two parts: the sensible heat absorbed by the material itself as it heats up, and the latent heat absorbed by the vaporization of the surface coating moisture, the main programmable logic controller directly extracts the specific mass value. This is converted into the heat demand per unit time under rated linear velocity, and the theoretical heat load power is calculated. The formula for calculating the theoretical heat load power is as follows:

[0063] ;

[0064] in, This represents the theoretical thermal load power corresponding to the oil injection at rated linear velocity within the modified mass sequence of the block sliding window; This indicates that the current section is about to enter the inlet section of the multi-temperature zone frying tank. Sequence index of each discrete material block at the entrance; This indicates the number of discrete material blocks contained in the block sliding window; Indicates the first The specific mass value corresponding to each sliding window discrete material block; Indicates the rated linear velocity; Indicates a fixed dimensional spacing; This parameter represents the equivalent specific heat capacity of the crispy pork filling. This indicates the target set temperature fed back by the temperature sensor within the physical temperature zone corresponding to the inlet section of the multi-temperature zone fryer. This indicates the initial temperature of the material being fed into the feed conveyor belt; This represents the latent heat constant per unit mass required for the vaporization of moisture on the surface of a material.

[0065] This indicates the required temperature difference increase for the material; This represents the total heat energy per unit mass of material required to be absorbed when it enters the oil. This indicates the frequency of material feed conveyor belt passing through discrete material blocks at its rated linear speed. The equivalent mass flow rate of a discrete material block within a sliding window representing the target at a rated linear velocity; This indicates the range of blocks within the sliding window from the first... The first inlet discrete material block to the first The data corresponding to the discrete material blocks in each sliding window are accumulated.

[0066] Regarding the physical constants and temperature parameters in the above formula, the equivalent specific heat capacity parameter... With unit mass latent heat constant Based on the formulation characteristics of the processed materials and the moisture content of the coating, the values ​​were determined through preliminary experiments and pre-written as constants into the storage area of ​​the main control programmable logic controller; initial feed temperature The target temperature can be collected in real time by an infrared temperature sensor installed at the front end of the material feeding conveyor belt, or set according to the ambient temperature of the workshop. The rated linear speed is set by the operator through a human-machine interface based on the frying process recipe of the target product; The standard frying and cooking time for crispy pork and the total physical length of the multi-temperature zone frying tank are comprehensively calculated and set to achieve accurate quantitative evaluation of heat load power.

[0067] S23. Dynamic available power calculation of the heating system. While calculating the theoretical heat load power required by the material, the main programmable logic controller synchronously reads the real-time opening data of the proportional heating valves within the physical temperature zone corresponding to the inlet section of the multi-temperature zone fryer, and combines this real-time opening data with the design maximum rated heat power of the corresponding physical temperature zone. (The maximum rated thermal power of the corresponding physical temperature zone) The current output heat power is calculated based on the power limit of the heating element in each temperature zone as specified on the manufacturer's nameplate of the multi-temperature zone fryer. The mapping conversion between the opening degree of the proportional heating valve and the output thermal power can be achieved by those skilled in the art using existing linear proportional conversion or polynomial curve fitting methods. The mapping conversion process is a well-known technology in the field and will not be described in detail here.

[0068] Subsequently, the main programmable logic controller (PLC) uses the processor to determine the maximum rated thermal power for the corresponding physical temperature zone. Subtract current output heat power The net available compensation power limit at the current moment can be obtained directly by subtracting the numerical values. Net available compensation power limit It represents the maximum additional thermal power margin that the current physical temperature range can provide without exceeding the upper limit of the hardware heating capacity.

[0069] S24. Feedforward limiting speed physical constraint solution. The main programmable logic controller calculates the theoretical thermal load power and net available compensation power limit. Numerical comparison and judgment are performed. The theoretical heat load power is less than or equal to the net available compensation power limit. Under these conditions, it is shown that the remaining heating capacity of the current physical temperature zone is sufficient to cope with the thermal load impact generated by the material in the sliding window of the block, and the main control programmable logic controller maintains the rated linear speed unchanged.

[0070] When the theoretical heat load power exceeds the net available compensation power limit Under operating conditions, it is indicated that operating at the rated linear speed will lead to insufficient heat input and subsequent oil temperature collapse. The main programmable logic controller calculates the net available compensation power limit based on the energy conservation limit. As the maximum heating capacity constraint boundary, the reverse derivation can match the feedforward limiting speed of the heating load, ensuring that the actual heat load generated by the mass of material entering the multi-temperature zone frying tank per unit time after reducing the conveyor belt speed does not exceed the net available compensation power limit. .

[0071] Furthermore, to avoid a logical dead zone in subsequent division operations where the total mass within the sliding window is zero when the material feed conveyor belt is empty, the main programmable logic controller (PLC) performs a summation check on the corrected mass sequence within the sliding window before performing the division operation. When the total mass within the sliding window is close to zero or lower than a preset minimum safety threshold, the feedforward limiting speed is directly assigned to the rated linear speed. The minimum safety threshold is obtained through calibration using the sensor background noise data of the dynamic weighing device under no-load conditions. When the total mass is within the normal range, the formula for calculating the feedforward limiting speed is as follows:

[0072] ;

[0073] in, This represents the feedforward limit velocity obtained from the energy conservation limit solution; This represents the minimum value function, used to limit the solution speed from exceeding the rated linear velocity; Indicates the rated linear velocity; This represents the net available compensation power limit calculated at the current moment; Indicates the dimension spacing; This indicates that the current section is about to enter the inlet section of the multi-temperature zone frying tank. Sequence index of each discrete material block at the entrance; This indicates the number of discrete material blocks contained in the block sliding window; Indicates the first The specific mass value corresponding to each sliding window discrete material block; This parameter represents the equivalent specific heat capacity of the crispy pork filling. This indicates the target set temperature fed back by the temperature sensor within the physical temperature zone corresponding to the inlet section of the multi-temperature zone fryer. This indicates the initial temperature of the material being fed into the feed conveyor belt; This represents the latent heat constant per unit mass required for the vaporization of moisture on the surface of a material.

[0074] This indicates the required temperature difference increase for the material; This represents the total heat energy per unit mass of material required to be absorbed when it enters the oil. The absolute heat requirement for oil injection into a discrete material block within a sliding window represents the target. This represents the total absolute heat requirement for all discrete material blocks within the sliding window of the block; It represents the physical limit of thermal energy that the system can provide through a single fixed dimensional spacing under the net available compensated power limit.

[0075] See attached document Figure 4 In conjunction with step S30, during the actual continuous frying control process, the feedforward peak-shaving speed regulation limitation calculation causes dynamic changes in the linear velocity of the material feeding conveyor belt. To avoid displacement misalignment caused by fixed-time-period calculations and to overcome the heat transfer lag from the opening of the proportional heating valve to the actual rise in oil temperature, the system uses displacement coordinates instead of a time clock to perform forward predictive addressing, achieving time synchronization between material arrival and heat replenishment. The specific implementation steps of the multi-temperature zone causal feedforward mapping derivation include the following sub-steps:

[0076] S31. Absolute Coordinate System Construction and Queue Synchronization. The main programmable logic controller reads the actual position pulses fed back by the rotary encoder and accumulates these pulses using an internally configured high-speed counter. The total number of accumulated pulses is then multiplied by the physical displacement equivalent of a single pulse to obtain the absolute physical displacement integral of the material feed conveyor belt. The physical displacement equivalent of a single pulse is pre-calibrated based on the number of pulses per revolution of the rotary encoder and the circumference of the physical drive shaft.

[0077] After obtaining the absolute physical displacement integral, the main programmable logic controller uses it as the forced displacement update clock and divides it by the size spacing. The values ​​are converted into corresponding digital shift pulse steps, and the block coordinates of discrete material blocks within the FIFO shift register queue are updated using these converted digital shift pulse steps. Even if the operating frequency of the belt drive inverter changes dynamically, the update of the block coordinates of discrete material blocks in the FIFO shift register queue is synchronously correlated with the absolute physical displacement integral, thereby maintaining the correspondence between the digital queue sequence address and the actual physical material position within the multi-temperature zone frying tank.

[0078] S32. Establishment of the heat transfer hysteresis model. The multi-temperature zone fryer is divided into multiple independent physical temperature zones along the material conveying direction. After receiving the command, there is a physical response delay in the heating circuit of each physical temperature zone, where heat is transferred from the proportional heating valve to the internal oil through the heat exchange tube wall, causing a change in the oil temperature distribution.

[0079] By performing offline step response tests on each independent physical temperature zone, the preset heat transfer hysteresis time for each independent physical temperature zone is extracted. The step response test and pure hysteresis time parameter extraction of the target heating system are well-known techniques in the field and will not be elaborated upon here. The extracted heat transfer hysteresis time is pre-written into the non-volatile memory area of ​​the main programmable logic controller.

[0080] S33. Forward addressing based on a kinematic model. The main programmable logic controller (PLC) calculates the actual linear velocity at the current moment based on the actual position pulse changes within consecutive adjacent sampling periods and the control cycle time interval. To ensure that heat injection and material arrival coincide on the time axis, the PLC performs forward addressing based on the updated block coordinates and the preset heat transfer hysteresis time for each physical temperature zone in the multi-temperature zone frying tank.

[0081] The main control programmable logic controller (PLC) extracts the preset acceleration / deceleration slope parameters of the conveyor belt drive frequency converter. These parameters are calibrated by those skilled in the art based on the overall mechanical inertia of the material feeding conveyor belt and the conveyor belt drive module, and preset in the conveyor belt drive frequency converter. The PLC, combining the current actual operating linear velocity and the preset heat transfer hysteresis time for each physical temperature zone, calculates the predicted forward displacement of the material feeding conveyor belt after experiencing the corresponding heat transfer hysteresis time, based on uniformly accelerated linear motion logic. This predicted displacement is then added to the integral of the absolute physical displacement of the material feeding conveyor belt at the current moment, and the resulting target absolute displacement is compared in the first-in-first-out shift register queue.

[0082] During this addressing comparison process, the main programmable logic controller will... The actual physical span of each physical temperature zone relative to the inlet section of the multi-temperature zone fryer is spatially vector-superimposed with the forward predicted displacement, and the total predicted distance after superposition is divided by the dimensional spacing. The calculation yields the first [number]th [unit] relative to the inlet section of the multi-temperature zone frying tank. The dynamic index offset steps of each discrete material block at the entrance. Using this dynamic index offset steps, forward addressing can be used to determine the target material block that will arrive at the corresponding physical temperature zone after the heat transfer hysteresis time, thereby achieving precise spatial anchoring of each independent physical temperature zone in the digital queue, and combining the sequence addresses of the target material blocks into a sequence index set.

[0083] When determining the sequence index set of the target material block, the main programmable logic controller (PLC) compares the addressing result with the maximum storage depth of the first-in-first-out (FIFO) shift register queue. This maximum storage depth is obtained by dividing the effective total physical length of the material feed conveyor belt by a fixed dimensional spacing. If the sequence address determined by the addressing exceeds the physical coverage boundary of the FIFO shift register queue, the excess target address is forcibly constrained to the index range corresponding to the maximum storage depth to prevent array overflow from causing logic operation crashes.

[0084] S34. Causal Prediction Feedforward Heat Power Calculation. After addressing and positioning are completed, the main programmable logic controller (PLC) extracts the corrected mass sequence corresponding to the target material block from the first-in-first-out shift register queue based on the sequence index set, and calculates the causal prediction feedforward heat power corresponding to each physical temperature zone based on the corrected mass sequence. This is performed on the target material in the multi-temperature zone frying tank. When performing specific formula derivation for each physical temperature zone, the main programmable logic controller reads the specific mass value corresponding to the target material block from the corresponding corrected mass sequence as input parameters. The calculation formula for causal prediction feedforward heat power is as follows:

[0085] ;

[0086] in, Indicates that for the first Each physical temperature zone has a preset heat transfer hysteresis time; Indicates the first The specific quality value corresponding to each target material block; This represents the actual linear velocity calculated at the current moment. Indicates the dimension spacing; This parameter represents the equivalent specific heat capacity of the crispy pork filling. Indicates the first The target set temperature is fed back by the temperature sensor within the physical temperature zone; Indicates that the target material block has entered the first Expected inlet temperature parameters for each physical temperature zone (when) The initial feed temperature is used for the first temperature zone at the inlet; subsequent temperature zones use the target set temperature of the previous adjacent temperature zone. This indicates that the allocation is based on the process and assigned to the first... The proportion of latent heat constant per unit mass in each physical temperature region;

[0087] The overall time state variable refers to the future moment after the corresponding heat transfer hysteresis time from the current moment. This indicates that after experiencing the corresponding heat transfer hysteresis time, the [number]th [time] will be reached. A set of sequence indexes for target material blocks within a physical temperature zone; Indicates that for the first The causal prediction feedforward heat power obtained by solving the physical temperature zone; Indicates the material in the first The required temperature difference increase for each physical temperature zone; This represents the phased unit mass heat energy that the target material block needs to absorb in the corresponding temperature zone. This indicates the real-time block passage frequency of the material feed conveyor belt as it passes through discrete material blocks at its actual operating linear speed. This represents the equivalent mass flow rate of the target material block at the actual operating linear velocity. Represents a set of sequence indices The data corresponding to all target material blocks contained therein are accumulated and calculated.

[0088] Among the above parameters, the equivalent specific heat capacity parameter Share of latent heat constant per unit mass Based on the material composition characteristics of the crispy pork and the moisture evaporation ratio in each temperature zone, the results were obtained through offline experiments or by consulting a physical property handbook and preset; target temperature was set. The corresponding frying process formula is preset; the initial feeding temperature is estimated by the ambient temperature or obtained by the pre-positioned temperature probe, and both are used as system solidification constants in the control calculation.

[0089] See attached document Figure 5 In conjunction with step S40, simple feedback control is prone to dynamic control lag when faced with a large volume of material being injected into the oil. The system introduces causal predictive feedforward heat power and couples it with the basic feedback control. It pre-adjusts the amount of heat energy injected based on load changes to offset major disturbances and eliminate residual steady-state errors, thereby reducing oil temperature fluctuations caused by heat load shocks. The specific implementation steps of the cross-coupled heating control include the following sub-steps:

[0090] S41, Basic Feedback Control Instruction Generation. The main programmable logic controller reads in real time the actual temperature data of each independent physical temperature zone deployed in the multi-temperature zone fryer. (This refers to the first step in the process.) Each physical temperature zone is controlled by a programmable logic controller (PLC) that calculates the preset target temperature. Real-time temperature deviation value between actual temperature range data and actual temperature range data and the real-time temperature deviation value The proportional-integral-derivative (PID) operation is performed to generate the basic feedback control command. The preset proportional, integral, and derivative gain constants relied upon in the PID operation are obtained by technicians through on-site tuning using the critical proportional gain method. The calculation formula for the basic feedback control command is as follows:

[0091] ;

[0092] in, Indicates that for the first The basic feedback control command is generated by proportional-integral-differential operations on each physical temperature zone. This represents the preset proportional gain constant parameter; This represents the real-time temperature deviation value calculated at the current moment; This represents the preset integral gain constant parameter; This represents the total number of control calculation cycles experienced from the initial operation to the current moment; Indicates the first The discrete sampling time corresponding to each control calculation cycle; Indicates the first Temperature deviation values ​​obtained at discrete sampling times corresponding to each control calculation cycle; This represents the preset differential gain constant parameter; This indicates the time interval step between two adjacent control calculation cycles; This represents the historical temperature deviation value obtained at the time corresponding to the previous control calculation cycle;

[0093] This represents the cumulative temperature deviation from the initial moment to the current moment; This represents the proportional action term in proportional-integral-differential operations; This represents the integral term in proportional-integral-differential operations; This indicates the temperature deviation change between two adjacent control calculation cycles. This represents the differential action term in proportional-integral-differential operations. The specific discretization program implementation for proportional-integral-differential operations is a well-known technique and will not be elaborated further.

[0094] S42. Causal Prediction Feedforward Inverse Mapping Transformation. While executing the basic feedback control instruction calculations, the main programmable logic controller extracts the causal prediction feedforward thermal power calculated and output from previous calculation stages. And using a pre-calibrated and stored inverse mapping function, the causal prediction feedforward thermal power is calculated. Convert to opening increment command The inverse mapping function characterizes the digital conversion relationship between the thermal power output of the target physical temperature zone and the physical opening degree of the proportional heating valve. (Opening increment command) The calculation formula is as follows:

[0095] ;

[0096] in, Indicates that for the first The opening increment command obtained by the conversion of each physical temperature zone at the current moment; Indicates that for the first The causal prediction feedforward heat power obtained by solving the physical temperature zone; This indicates the maximum rated thermal power for the corresponding physical temperature zone. This indicates the physical full-scale constant state of the proportional heating valve. It represents the theoretical load ratio coefficient of the causal prediction feedforward heat power to the design maximum rated heat power of the corresponding physical temperature zone.

[0097] S43. Cross-coupling superposition and anti-saturation constraint. After the main programmable logic controller (PLC) obtains the basic feedback control command and the opening increment command, it performs algebraic superposition of the two in the digital logic operation space to generate a superposition result of dual control data including temperature deviation correction and active heat load prediction.

[0098] To prevent the superposition result from exceeding the physical stroke boundary of the proportional heating valve and causing integral saturation runaway, the main programmable logic controller (PLC) acquires the physical maximum allowable opening boundary (100%) and the physical minimum allowable opening boundary (0%), which are constantly calibrated by the mechanical valve core's physical limit. It then applies a saturation constraint function to numerically constrain the superposition result, calculating and generating the final heating valve opening command. The formula for calculating the final heating valve opening command is as follows:

[0099] ;

[0100] in, Indicates that for the first The final heating valve opening command is generated after constraining each physical temperature zone. This represents the saturation constraint function operator; Indicates basic feedback control commands; This indicates an increment instruction for the opening degree; This represents the physical minimum allowable opening boundary; Indicates the physical maximum allowable opening boundary; This represents the superposition result obtained by algebraically superimposing the basic feedback control command and the opening increment command.

[0101] When executing the saturation limit function: if the superposition result exceeds the physical maximum allowable opening boundary or is lower than the physical minimum allowable opening boundary, the final heating valve opening command will be clamped and assigned to the physical maximum allowable opening boundary or the physical minimum allowable opening boundary respectively; if the superposition result is between the two boundaries, the superposition result will be directly assigned to the final heating valve opening command.

[0102] To avoid integral saturation, during the control calculation cycle when the superposition result exceeds any physical opening boundary, the main programmable logic controller synchronously triggers an integral freeze command to suspend the integral action term in the basic feedback control command. The accumulation process retains only the proportional and derivative actions. This stops the ineffective integral growth and ensures that the system can immediately exit saturation and restore linear regulating response when reverse heat load or control demand occurs.

[0103] S44. Instruction Analog-to-Digital Conversion and Physical Execution Drive. After completing the numerical boundary constraint calculation, the main programmable logic controller (PLC) converts the final heating valve opening command into a standard industrial analog electrical signal (e.g., 4–20mA or 0–10V) and transmits it to the proportional heating valve corresponding to the physical temperature zone. The proportional heating valve dynamically adjusts the mechanical valve core displacement according to the industrial analog electrical signal to change the heating medium flow rate and perform heat replenishment for the corresponding physical temperature zone. The specific hardware implementation of the analog-to-digital conversion circuit and the proportional heating valve drive is well-known technology and will not be described further.

[0104] In conjunction with step S50, to address speed regulation interference caused by thermal load shocks during feedforward peak shaving speed regulation limitation calculations, the system introduces a time-temperature equivalent integral model to monitor the actual thermal history of the material. Through closed-loop arbitration and boundary safety control, it resolves multi-source control conflicts between energy constraints and quality consistency, including the following sub-steps:

[0105] S51. Cumulative Calculation of Shell Forming Heat Integral for Discrete Material Blocks Based on Residence Period. In the digital space, the main programmable logic controller, based on the pre-built first-in-first-out shift register queue, calculates the shell forming heat integral for each discrete material block that arrives at the discharge end of the multi-temperature zone fryer with physical transport, by accumulating the actual temperature data of the temperature zone at historical moments fed back by temperature sensors along the time dimension. The formula for calculating the shell forming heat integral is as follows:

[0106] ;

[0107] in, Indicates the first The shell forming heat integral corresponding to each heated discrete material block; The sequence index variable represents the heated discrete material block that arrives at the discharge end of the multi-temperature zone fryer; Indicates the first The moment when a heated discrete material block enters the multi-temperature zone frying tank; Indicates the first The current discharge time of a heated discrete material block leaving the multi-temperature zone fryer; This represents the shaping reaction rate constant corresponding to the material properties of crispy pork; Indicates the corresponding historical moment The The actual temperature record data of the physical temperature zone where each heated discrete material block is located; This indicates the preset physical gelatinization initiation threshold temperature;

[0108] This represents the difference between the actual temperature record data of the corresponding historical moment and the physical gelatinization initiation threshold temperature; This is a non-negative selection operation function that extracts the effective temperature difference portion that is greater than zero from the difference. This indicates the effective heat accumulation rate of the shell forming reaction; This indicates that the effective heat accumulation rate is calculated by time integration over the time interval of the material residence period.

[0109] The shaping reaction rate constant and the physical gelatinization initiation threshold temperature were obtained by fitting the water loss and center temperature penetration curves under laboratory isothermal conditions and were pre-stored in the non-volatile memory area of ​​the main control programmable logic controller. The main control programmable logic controller uses discrete time steps to realize the digital accumulation of the shell molding heat integral calculation formula. The discrete digital programming implementation using ladder diagrams or structured text code is a well-known technology in this field and will not be described in detail here.

[0110] S52. Finished Product Cookiness Characteristic Data Feedback, Basic Target Correction, and Integral Difference Extraction. The quality inspection module includes a discharge detection camera aimed at the formed material leaving the multi-temperature zone frying tank, collecting finished product cookiness characteristic data, including color difference and morphological characteristic values. The collected finished product cookiness characteristic data is processed by a low-pass filtering algorithm and then fed back to the main control programmable logic controller. This digital filtering program, which uses a first-order inertial filtering formula, is a well-known technology in the field and will not be described in detail here.

[0111] The basic maturity target is pre-calibrated and generated from the average sampling inspection characteristic parameters of historical qualified batches of products. The main control programmable logic controller corrects the preset basic maturity target in real time based on the finished product maturity characteristic data after low-pass filtering to obtain the standard maturity target value, and calculates the integral difference between the standard maturity target value and the shell molding heat integral as the basis data for maturity closed-loop evaluation.

[0112] S53. Calculation of under-integral state residence delay time and generation of ripeness compensation speed limit. When the main control programmable logic controller determines that the integral difference is greater than zero, that is, it evaluates that the shell forming heat integral of the discharged block is lower than the standard ripeness target value, and thus predicts that the system as a whole is showing an underheating trend, in order to make up for the under-integral gap and extend the residence heating time of the material still in the multi-temperature zone frying tank, the main control programmable logic controller uses a preset ripeness compensation ratio coefficient to convert the integral difference into the required residence delay time; wherein, the ripeness compensation ratio coefficient is obtained by linearly fitting the shell forming heat integral deviation and the corresponding process residence time compensation amount in each qualified production batch.

[0113] Meanwhile, to prevent excessive dehydration or coking of materials in the tank due to excessively large integral differences leading to excessively long calculated residence delay times, the system sets a maximum safe upper limit threshold for the maturity compensation ratio coefficient to prevent divergence, and also sets a time cutoff threshold for the converted residence delay time. Subsequently, the effective total physical length of the multi-temperature zone frying tank is divided by the sum of the standard process residence time and the residence delay time to calculate the maturity compensation speed limit. Since extending the residence time requires the conveyor belt to operate at a reduced speed, the maturity compensation speed limit in the control loop manifests as a maximum speed limit constraint to ensure the maturity quality of the product.

[0114] S54. Multivariable Target Operating Speed ​​Command Arbitration and Minimum Process Line Speed ​​Bottom Line Locking Mechanism. Given that the feedforward limit speed generated in step S20 and the maturity compensation speed limit generated in step S53 constitute concurrent requests to the same controlled actuator, to balance oil temperature collapse prevention and under-maturity compensation, the main programmable logic controller arbitrates the two speeds to generate a target operating speed command. The mathematical relationship of the target operating speed command is as follows:

[0115] ;

[0116] in, Indicates the target running speed command at the current moment; This represents the feedforward limit velocity obtained from the energy conservation limit solution; This represents the maturity compensation speed limit generated by the solution; This represents a selection operation function that compares the feedforward limiting speed and the maturity compensation speed limit and extracts the smaller value.

[0117] The main programmable logic controller (PLC) then determines the system's operating status. The system's preset minimum process line speed is determined by jointly evaluating the mechanical heat dissipation limit of the belt drive module and the minimum throughput requirement. Under normal operating conditions, if the target operating speed command obtained through arbitration is greater than or equal to the minimum process line speed, the main PLC will smooth the target operating speed command and directly issue it as the inverter's operating speed command.

[0118] S55, the defect traceability tag marking is linked with the externally configured front-end feeder and end-end sorting mechanism for intervention. When the target operating speed command falls below the minimum process line speed boundary condition, in order to prevent the production line from stopping, the main control programmable logic controller issues an inverter operating speed command locked to the minimum process line speed to the belt drive inverter.

[0119] Because the minimum process line speed forced by the belt drive frequency converter is greater than the target operating speed command, a shortage of material residence time occurs, leading to the risk of undercooked material. Accordingly, the main control programmable logic controller (PLC) marks the defect traceability tag of the target discrete material block under boundary conditions in the FIFO shift register queue and simultaneously outputs a control signal to the drive start / stop control terminal of the front-end feeder to limit the input heat load of the feed. Simultaneously, when the defect traceability tag in the FIFO shift register queue moves with the pointer to the discharge end of the multi-temperature zone fryer and aligns with the discharge detection camera, a command is issued to control the end-of-line sorting mechanism to perform targeted interception and defective product diversion of the target discrete material block marked with the defect traceability tag.

[0120] Reference Appendix Figure 6 To be continued Figure 8 To further disclose and deepen the technical features summarized in this invention, a specific implementation of the system's hardware and software features and collaborative control process is described in detail, taking a continuous frying production line for crispy pork with a capacity of 500 kg per hour in a food processing plant as an example.

[0121] In this embodiment, the main control programmable logic controller (PLC) is a Siemens S7-1500 series PLC with high-speed motion control and complex floating-point arithmetic capabilities. The industrial Ethernet protocol specifically uses the PROFINET fieldbus to achieve millisecond-level synchronization of the control cycle. The 3D laser profilometer uses a Keyence LJ-V7000 series line laser scanner with a scanning frequency set to 2kHz. The dynamic weighing device uses a high-speed parallel beam load cell array with C3 accuracy. The temperature sensor uses a Class A PT100 platinum resistance thermometer paired with a 4-20mA temperature transmitter to reduce signal attenuation over long distances. The proportional heating valve uses a pneumatic diaphragm regulating valve with an intelligent electric valve positioner, with a response resolution controlled within 0.1%. The discharge detection camera in the quality inspection module is an industrial CCD color area array camera with a gigabit network interface. The end-of-line sorting mechanism is specifically a flap-type pneumatic rejection device driven by a high-frequency solenoid valve. The front-end feeder is a variable frequency vibrating feeder hopper.

[0122] Parameter initialization is synchronized with hardware configuration. After the system is powered on, the main programmable logic controller sets the rated linear speed of the material feeding conveyor belt. Set to 0.05 m / s, with a fixed spacing. The value is set to 0.01m. Therefore, the shift time interval for each shift in the first-in-first-out shift register queue within the digital space is 0.2s. For the crispy pork product, the equivalent specific heat capacity parameter is measured offline. The latent heat constant is set at 2.8 kJ / (kg·℃). The target set temperature for the first temperature zone at the inlet of the multi-temperature zone fryer is set at 2260 kJ / kg. The initial feed temperature is set to 175℃. The temperature is set at 20℃. The first temperature zone corresponds to the maximum rated thermal power. The power is 120kW. After step response testing, the preset heat transfer lag time for the first temperature zone was determined. Write to the storage area, the value is 15 seconds.

[0123] Load array construction and dynamic weighing are integrated. After the feeder starts, the crispy pork enters the conveyor belt. The main control programmable logic controller receives the pulse sequence corresponding to a physical displacement of 0.01m generated by the rotary encoder and outputs an external trigger synchronization signal to the 3D laser profilometer to perform scanning. The collected point cloud data is uploaded to the PLC memory, and a second-order Butterworth digital low-pass filter algorithm is used to eliminate noise caused by oil stains on the belt surface. Then, the discrete distribution volume of the material is calculated by summing the data.

[0124] Synchronously, the underlying weighing sensor collects actual physical mass data every 10 milliseconds, and filters it using a sliding average algorithm with a time window containing 20 sampling points. The PLC uses this physical mass divided by the sum of the volumetric accumulations within the corresponding spatial interval to calculate the dynamic equivalent density coefficient (approximately 0.85 to 0.92 g / cm³). 3 (Dynamic fluctuations between them), thereby generating specific quality values, and generating corrected quality sequences for each discrete material block based on the specific quality values ​​of each discrete material block, which are then stored in the first-in-first-out shift register queue.

[0125] High load prediction and feedforward peak shaving limit calculation. When the operator pours excessive material into the vibrating feeder hopper, a sudden high load occurs. The main control programmable logic controller (PLC) extracts the corrected quality sequence through a block sliding window in the first-in-first-out shift register queue (this window covers 100 discrete material blocks equivalent to a 1-meter physical length in front of the multi-temperature zone fryer inlet). Calculations show that if operating at the rated linear velocity of 0.05 m / s, the theoretical thermal load power of the material about to enter the oil reaches 145 kW. At this time, the PLC reads the real-time opening data of the first temperature zone proportional heating valve, corresponding to the current output thermal power of 30 kW. Combining this with the design maximum rated thermal power of 120 kW, the net available compensation power limit is calculated. The power output is only 90kW. Since the theoretical heat load (145kW) is significantly greater than the available power (90kW), the PLC immediately initiates the energy conservation limit calculation, deriving the feedforward speed limit. The flow rate is set at 0.031 m / s to forcibly limit the material inflow rate into the oil, preventing the oil temperature from collapsing due to overheating in the first temperature zone.

[0126] Feedforward control with hysteresis compensation and inverse mapping drive. Due to the 15s heat transfer hysteresis time in the first temperature zone, the PLC does not wait for the actual temperature data of the temperature zone to drop from the temperature sensor, but instead performs kinematic forward addressing by combining the absolute displacement integral and the actual running linear velocity.

[0127] The PLC locates the target material block, which is 15 seconds away from the inlet of the first temperature zone, extracts its mass value, and extrapolates the feedforward thermal power. This feedforward power is directly converted into a 35% opening increment command through the built-in valve flow characteristic curve (equal percentage inverse mapping function).

[0128] Meanwhile, based on the small temperature deviation fed back by the PT100 in real time, incremental proportional-integral-derivative (PID) operations are performed to generate basic feedback control commands. The PLC algebraically superimposes the two commands, and after numerical clamping constraints from 0% to 100%, outputs a 4-20mA analog control signal, converted from the final heating valve opening command, to the electric valve positioner. This activates the steam heating circuit 15 seconds before the material actually arrives to absorb heat, thus offsetting the physical lag characteristic of the system.

[0129] The process involves integrating the maturity score and locking the baseline. As material passes through the oil tank, the PLC performs an integral calculation of the shell-forming heat for each material block based on temperature sampling every 100 milliseconds. Due to high load prediction and feedforward peak shaving limiting the calculation steps, the conveyor belt speed decreases, increasing the material residence time. The discharge detection camera detects that the surface color difference L value of some samples is low (darkened color), indicating a tendency for overheating. The PLC lowers the standard maturity target value to correct this, and the calculation reveals a negative integral difference.

[0130] The maturity compensation algorithm calculates a negative dwell time based on the integral difference to shorten the dwell time and outputs a maturity compensation speed limit. The speed is 0.045 m / s. The PLC is in... (0.031m / s) and Arbitration is performed at the minimum value between 0.045 m / s and 0.031 m / s to ensure thermal balance. If extreme heat loss causes the calculation speed to drop below the minimum process line speed of 0.02 m / s set by the mesh belt heat dissipation limit, the PLC will forcibly send a frequency converter running speed command locked by the minimum process line speed to the mesh belt drive frequency converter, write a hexadecimal defect traceability tag to the affected block in the memory queue, and simultaneously send a stop command to the vibrating feeder hopper through the PROFINET protocol. When the marked material reaches the discharge port, the solenoid valve of the pneumatic rejection device will be triggered to blow the uncooked material into the waste recycling bin.

[0131] To verify the reliability of this invention, a comparative experiment was conducted using the hardware platform built in the above embodiments. The experiment was set to a continuous operating cycle of 2 hours. For the first 30 minutes, a base load of 500 kg / h was maintained. At the 30-minute mark, the operating frequency of the feeding conveyor was artificially increased, creating a high-load impact condition of 800 kg / h for 3 minutes. Two independent tests were conducted on the same physical equipment, using both a traditional single-loop PID constant speed control program and the collaborative control program of this invention. The frying and cooking qualification rate was statistically analyzed using a discharge detection camera at the end of the production line combined with manual sampling, while the operating data of the underlying PLC was recorded using upper-level computer software.

[0132] Experimental and graphical data validated the control effectiveness of the present invention. (See attached...) Figure 6 As shown in the figure, the feed heat load changes with time. The horizontal axis is the running time (s) and the vertical axis is the feed load (kg / h). It can be seen that at t=30s, it enters a high load range that lasts for 30 seconds.

[0133] Observation Appendix Figure 7 The comparison of actual oil temperature changes and the data recorded in the background show that the horizontal axis of the graph represents the running time (s), and the vertical axis represents the oil temperature in the first temperature zone (°C). Under the traditional single-loop PID control mode shown in the legend, the conveyor belt speed remains constant at 0.05 m / s (as shown in the attached figure). Figure 8 The middle figure shows the curve of the traditional constant speed mode. A large amount of room temperature material exceeding the heating limit rushed into the oil tank, causing the oil temperature in the first temperature zone to deviate significantly from the baseline of the process target set value indicated by the figure. It dropped sharply from 175℃ to 165.8℃, with a maximum temperature drop of 9.2℃. It took 85 seconds for the temperature to rise again and stabilize at the set value range. During this period, the 48 kg of small crispy meat produced was identified by the camera as having problems such as the batter not being set and insufficient internal cooking and dehydration, resulting in a surge in the defect rate.

[0134] After applying the collaborative control system of this invention, the main programmable logic controller uses a first-in-first-out shift register queue to predict the imminent thermal load exceeding the limit. (See attached...) Figure 8The graph shows a comparison of the changes in the linear velocity of the conveyor belt. The horizontal axis represents the running time (s), and the vertical axis represents the linear velocity of the conveyor belt (m / s). The legend shows the curve of the adaptive feedforward speed limiting mode of this invention, indicating that before the sudden load oil inflow (t=28s), the linear velocity of the conveyor belt has been actively reduced and smoothly transitioned to 0.031m / s to implement flow limiting. At the same time, combined with the forward addressing logic, the opening of the heating valve is injected with a feedforward increment 15 seconds in advance.

[0135] As attached Figure 7 The figure in the middle illustrates the curves of the collaborative control of this invention. The oil temperature experiences only a slight disturbance, with the maximum drop compressed to 1.8℃ (minimum 173.2℃), and recovers to the steady-state reference of 175℃ within 20 seconds. Because the extended residence time compensates for the slight temperature drop, verification using a discharge detection camera combined with the heat integration of the shell forming process shows that the color characteristics of the crispy pork produced under impact conditions are highly consistent with those under constant load conditions. The overall defect rate of the machine has decreased from 4.5% in conventional testing to 0.2%.

[0136] The experimental data and the trend of the curves in the attached figure fully demonstrate that the queue prediction, feedforward peak shaving speed regulation and multivariate arbitration mechanism designed in this invention effectively eliminate the damage of sudden heat load to the temperature control system, resolve the contradiction between the equipment's energy supply limit and physical design requirements, and achieve a high degree of consistency in product quality.

Claims

1. A collaborative control system for parameters of a fried pork frying production line, characterized in that, include: The main programmable logic controller generates a corrected mass sequence and constructs a first-in-first-out shift register queue for discrete material blocks based on the three-dimensional point cloud data of the material collected by the three-dimensional laser profilometer and the actual physical mass data obtained by the dynamic weighing device. The actual position pulses fed back by the mesh belt drive module are converted into absolute physical displacement integrals, and the theoretical heat load power is calculated in combination with the corrected mass sequence. The feedforward limit speed is then calculated based on the theoretical heat load power. Simultaneously, based on the absolute physical displacement integral forward addressing, the corrected mass sequence is extracted and the causal prediction feedforward thermal power is calculated. The final heating valve opening command is generated by combining the causal prediction feedforward thermal power with the actual temperature zone data. The ripeness characteristic data of the finished product collected by the quality inspection module and the actual temperature data of the temperature zone are combined to calculate the ripeness compensation speed limit. The feedforward limit speed and the ripeness compensation speed limit are arbitrated to generate a target running speed command. The target running speed command is output to the mesh belt drive module and drives the material feeding mesh belt to run, thereby completing the coordinated control of the production line parameters.

2. The collaborative control system for parameters of a fried pork frying production line according to claim 1, characterized in that, The main programmable logic controller divides the effective physical length of the continuously running material feeding conveyor belt into multiple discrete material blocks according to a fixed size interval; The main control programmable logic controller receives the actual position pulse and converts the actual position pulse into an external trigger synchronization signal, which is then sent to the three-dimensional laser profilometer. The three-dimensional laser profilometer performs line laser scanning at fixed physical displacement intervals based on the external trigger synchronization signal to collect three-dimensional spatial coordinate data of the material surface and generate three-dimensional point cloud data of the material.

3. The collaborative control system for parameters of a fried pork frying production line according to claim 2, characterized in that, The three-dimensional laser profilometer is positioned above the material feeding conveyor belt, and the dynamic weighing device is positioned in the bottom support structure of the material feeding conveyor belt; The main programmable logic controller calculates the discrete distribution volume of the material corresponding to each discrete material block based on the three-dimensional point cloud data of the material, and generates a dynamic equivalent density coefficient by combining the actual physical mass data; multiplies the dynamic equivalent density coefficient with the discrete distribution volume of the material to generate a specific mass value corresponding to the discrete material block, and generates a corrected mass sequence corresponding to each discrete material block by combining the specific mass value and stores it in the first-in-first-out shift register queue.

4. The collaborative control system for parameters of a fried pork frying production line according to claim 3, characterized in that, It also includes a multi-temperature zone frying tank and a proportional heating valve. The multi-temperature zone frying tank is arranged in the downstream area of ​​the material feeding mesh belt, and the proportional heating valve is set in the heating circuit of each physical temperature zone. The final heating valve opening command is used to control the proportional heating valve. The main programmable logic controller is also used to delineate a block sliding window corresponding to the inlet section of the multi-temperature zone frying tank in the first-in-first-out shift register queue, and extract the corrected mass sequence within the block sliding window to calculate the theoretical heat load power; Read the real-time opening data of each of the proportional heating valves and combine it with the design maximum rated thermal power of the corresponding physical temperature zone to calculate the net available compensation power limit; When the theoretical heat load power exceeds the net available compensation power limit, the feedforward limiting speed is calculated in reverse based on the net available compensation power limit.

5. The collaborative control system for parameters of a fried pork frying production line according to claim 4, characterized in that, The theoretical heat load power is the sum of the product of the specific mass value, the ratio of the rated linear velocity to the fixed dimensional spacing, and the total heat energy per unit mass required for the material to enter the oil. The net available compensation power limit is the difference between the design maximum rated thermal power corresponding to the physical temperature zone and the current output thermal power corresponding to the real-time opening data. The feedforward limiting velocity is the smaller of the quotient obtained by multiplying the net available compensation power limit by the fixed dimensional spacing and dividing by the sum of absolute heat requirements, and the rated linear velocity.

6. The collaborative control system for parameters of a fried pork frying production line according to claim 4, characterized in that, The mesh belt drive module includes a rotary encoder, which feeds back the actual position pulses. The main programmable logic controller is also used to update the block coordinates of the first-in-first-out shift register queue based on the absolute physical displacement integral, and to obtain the preset heat transfer hysteresis time for each of the physical temperature zones. Based on the heat transfer hysteresis time, the updated block coordinates, and the actual operating linear velocity, forward addressing is performed to determine the target material block; Extract the corrected mass sequence corresponding to the target material block, and calculate the causal prediction feedforward heat power corresponding to each physical temperature zone. The causal prediction feedforward heat power corresponding to each physical temperature zone is the sum of the product of the specific mass value corresponding to the target material block, the ratio of the actual operating linear velocity to the fixed size spacing, and the phased unit mass heat energy that the material needs to absorb in the physical temperature zone.

7. The collaborative control system for parameters of a fried pork frying production line according to claim 6, characterized in that, It also includes temperature sensors, which are deployed in each of the physical temperature zones to acquire temperature data of the actual temperature zone; The main programmable logic controller is also used to calculate the real-time temperature deviation between the preset target temperature and the actual temperature zone data, perform proportional-integral-derivative operations to generate basic feedback control instructions, and convert the causal prediction feedforward thermal power into opening increment instructions through an inverse mapping function. The basic feedback control command and the opening increment command are algebraically superimposed to generate a superposition result. The superposition result is numerically bounded by a saturation constraint function, and the final heating valve opening command is output. When executing the saturation constraint function: If the superposition result exceeds the physical maximum allowable opening boundary or is lower than the physical minimum allowable opening boundary, the main control programmable logic controller clamps and assigns the final heating valve opening command to the physical maximum allowable opening boundary or the physical minimum allowable opening boundary respectively, and triggers an integral freeze command to suspend the accumulation operation of the integral action term in the basic feedback control command.

8. The collaborative control system for parameters of a fried pork frying production line according to claim 4, characterized in that, The quality inspection module includes a discharge inspection camera arranged at the discharge end of the multi-temperature zone frying tank, and the discharge inspection camera is aimed at the formed material leaving the multi-temperature zone frying tank. The main programmable logic controller is also used to calculate the shell forming heat integral by accumulating the actual temperature data of the temperature zone along the time dimension for each discrete material block that arrives at the discharge end of the multi-temperature zone frying tank. Wherein, the shell forming heat integral is the integral value of the effective heat accumulation rate calculated over time within the time interval of the material residence period, the effective heat accumulation rate is the product of the shaping reaction rate constant and the effective part of the temperature difference, and the effective part of the temperature difference is the part greater than zero in the difference between the actual temperature zone temperature data and the physical gelatinization initiation threshold temperature. The main control programmable logic controller corrects the preset basic maturity target based on the finished product maturity characteristic data to obtain a standard maturity target value, calculates the integral difference between the standard maturity target value and the shell forming heat integral, and calculates the maturity compensation speed limit based on the integral difference. The feedforward limit speed and the maturity compensation speed limit are compared, and the smaller value is extracted as the target running speed command.

9. The collaborative control system for parameters of a fried pork frying production line according to claim 8, characterized in that, It also includes a front-end feeder and an end-end sorting mechanism; When the target operating speed command falls below the preset minimum process line speed, the main control programmable logic controller sends an inverter operating speed command locked by the minimum process line speed to the mesh belt drive module. In the first-in-first-out shift register queue, a defect traceability tag is marked for the corresponding discrete material block. A feeding limit instruction is synchronously output to the front-end feeder to limit the input of feeding heat load. When the defect traceability tag moves with the data pointer to be aligned with the discharge detection camera, the end sorting mechanism is linked to perform directional interception and defective product diversion.

10. A method for coordinated control of parameters in a fried pork frying production line, characterized in that, A collaborative control system for parameters of a fried pork belly production line as described in any one of claims 1-9 includes the following steps: Based on the material's three-dimensional point cloud data and actual physical quality data, the feed heat load array is reconstructed to generate a corrected quality sequence corresponding to each discrete material block, and the corrected quality sequence is stored in the first-in-first-out shift register queue. In the first-in-first-out shift register queue, a block sliding window is defined to extract the corrected quality sequence and calculate the theoretical heat load power. Based on the comparison between the theoretical heat load power and the net available compensation power limit, the feedforward peak shaving speed regulation limit solution is performed to obtain the feedforward limit speed. The actual position pulse fed back by the mesh belt drive module is converted into an absolute physical displacement integral. The block coordinates are updated based on the absolute physical displacement integral. The target material block is determined by multi-temperature zone causal feedforward mapping based on the heat transfer hysteresis time. The causal prediction feedforward heat power corresponding to each physical temperature zone is calculated in combination with the corrected mass sequence. By combining the causal prediction feedforward thermal power with the actual temperature zone data, cross-coupled heating control is performed to generate the final heating valve opening command to adjust the heat input of the physical temperature zone. The shell forming heat integral is calculated by combining the actual temperature zone temperature data, and the temperature equivalent integral closed loop and arbitration boundary intervention are executed to determine the maturity compensation speed limit. The feedforward limit speed and the maturity compensation speed limit are arbitrated to generate the target running speed command.

Images