A food production equipment parameter self-adaptive regulation and control system and method

Abstract

The application relates to the technical field of food processing equipment control, and discloses a food production equipment parameter self-adaptive regulation and control system and method, which comprises an absolute value encoder, a plurality of physical temperature zones and a main control unit. The main control unit divides continuous material into virtual material slices, updates absolute spatial coordinates and instantaneous linear velocity; obtains environmental temperature and circulating fan operation frequency to calculate the actual dynamic heat enthalpy accumulation of the slices; when the slices reach the end of the current temperature zone, generates a feedforward compensation opening increment for the downstream temperature zone according to the heat enthalpy error; establishes a trigger point based on time hysteresis parameters and linear velocity to issue the increment in advance; and when issuing, expands the static dead zone constant of the current temperature zone controller into a dynamic dead zone threshold value according to a cross-coupling coefficient to freeze output change. The application solves the problems of heat response lag and airflow coupling interference of the multi-temperature zone equipment, and prevents control oscillation.

Description

Technical Field

[0001] This invention relates to the field of food processing equipment control technology, specifically to an adaptive control system and method for parameters of food production equipment. Background Technology

[0002] Multi-zone tunnel heating equipment is widely used in continuous food production processes. This type of equipment is typically divided into multiple physical temperature zones, each equipped with an independent heating control loop and temperature sensor, using feedback algorithms such as PID to maintain a constant ambient temperature in each zone.

[0003] However, existing multi-temperature zone food processing equipment has certain limitations in its control logic. Current control systems mostly rely on ambient temperature collected by fixed spatial position sensors as the sole feedback reference, failing to consider the thermodynamic state of the continuously moving material itself. The material's movement speed, heating time, and local convective heat transfer coefficient within the equipment are dynamically changing. Relying solely on ambient temperature cannot accurately reflect the actual heat absorbed by the material throughout the processing cycle, which can easily lead to deviations in the final product's processing maturity.

[0004] Meanwhile, the heating actuators of large-scale thermal equipment inherently exhibit thermal response lag. Traditional feedback control modes only begin adjusting the gas proportional valve or fan frequency after detecting a temperature deviation in the current temperature zone. Because heat transfer takes time, this reactive adjustment lag makes it difficult for the equipment's temperature output to match the real-time heat demand of materials during spatial movement.

[0005] Furthermore, in mechanical structures with continuous multi-temperature zones, there is usually no absolute physical seal between the zones. When the opening degree or fan frequency of a certain temperature zone changes due to control requirements, the resulting hot airflow will inevitably diffuse into adjacent temperature zones. This airflow intrusion will cause physical disturbances in the temperature sensor values ​​of adjacent temperature zones, leading the controllers of those zones to mistakenly believe that their own temperature is too high and incorrectly reduce the heating output. This cross-coupling interference between adjacent temperature zones can easily cause reciprocating oscillations and adjustments between multiple control loops, making it difficult to maintain a stable global temperature field for the entire system. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides an adaptive control system and method for parameters of food production equipment, which solves the problems of thermal response lag, neglect of the dynamic thermodynamic state of materials, and oscillation of the control system caused by airflow coupling interference between adjacent temperature zones in the control of multi-temperature zone food processing equipment in the prior art.

[0007] To address the above problems, the present invention provides the following technical solution: The first aspect of this invention provides an adaptive control system for parameters of food production equipment, characterized in that it comprises: Absolute encoder, multiple physical temperature zones, and main control unit; The main control unit stores: The slice tracking module is used to divide continuous material into virtual material slices and update the absolute spatial coordinates and instantaneous linear velocity of the virtual material slices based on the absolute encoder data. The enthalpy integration module is used to obtain the ambient temperature and circulating fan operating frequency of the virtual material slice based on the absolute spatial coordinates, and to calculate the actual dynamic enthalpy accumulation of the virtual material slice. The batch aggregation module is used to calculate the spatial moving average deviation based on the error between the actual dynamic enthalpy accumulation and the target enthalpy threshold when the virtual material slice reaches the end of the current physical temperature zone, and generate the feedforward compensation opening increment for the downstream adjacent physical temperature zone. The compensation trigger module is used to establish an absolute spatial coordinate trigger point based on the time hysteresis parameter and the instantaneous linear velocity. When the virtual material slice reaches the absolute spatial coordinate trigger point, it sends the feedforward compensation opening increment to the downstream adjacent physical temperature zone. The dead zone control module is used to expand the static dead zone constant of the current physical temperature zone controller into a dynamic dead zone threshold based on the cross-coupling coefficient when issuing the feedforward compensation opening increment to freeze the output change.

[0008] Preferably, the slice tracking module initializes a static parameter matrix, which includes a target enthalpy threshold array, a heat capacity response time bias matrix, a thermal dispersion field width matrix, and an airflow coupling matrix between adjacent temperature zones. The enthalpy integration module compares the absolute spatial coordinates of the virtual material slice with the physical installation positions of adjacent thermocouples, and calculates the equivalent ambient temperature by weighted average based on the relative position ratio. Extract the mapping function between the convective heat transfer coefficient and the fan operating frequency, calculate the convective heat transfer coefficient corresponding to the current absolute spatial coordinates, obtain the temperature difference value by subtracting the equivalent ambient temperature from the material base reference temperature constant, and multiply the convective heat transfer coefficient by the temperature difference value to obtain the instantaneous heat flux. The single-step heat absorption is calculated based on the instantaneous heat flux. If the single-step heat absorption is within the effective heat absorption range, the single-step heat absorption is added to the actual dynamic enthalpy accumulation.

[0009] Preferably, the batch aggregation module divides the width of the thermal dispersion field corresponding to the current physical temperature zone by a fixed physical length constant and performs an up rounding operation to calculate the capacity of the spatial sliding window; When the virtual material slice that has reached the end boundary of the current physical temperature zone is used as the center to trace forward and backward beyond the effective sequence range, the sliding window is truncated to the actual usable data boundary. Calculate the absolute value of the spatial moving average deviation. If the absolute value is less than the preset thermal enthalpy tolerance dead zone, then the feedforward compensation opening increment is set to zero. If the deviation is greater than or equal to the preset enthalpy tolerance dead zone, the deviation difference between the spatial moving average deviation and the preset enthalpy tolerance dead zone is calculated using the set feedforward proportional gain coefficient. Based on the positive or negative polarity of the spatial moving average deviation, the feedforward proportional gain coefficient is multiplied by the deviation difference to obtain the feedforward compensation opening increment.

[0010] Preferably, the compensation triggering module extracts the time hysteresis parameter corresponding to the downstream physical temperature zone in the thermal capacity response time bias matrix, and multiplies the instantaneous linear velocity with the time hysteresis parameter to obtain the spatial advance triggering distance; The absolute spatial coordinate trigger point is determined by subtracting the spatial advance trigger distance from the starting physical coordinates of the downstream physical temperature zone. Determine whether the space advance trigger distance is greater than the actual mechanical construction length of the current physical temperature zone. If the determination result is yes, forcibly cut off the space advance trigger distance and clamp it to the preset limit ratio of the actual mechanical construction length of the current physical temperature zone.

[0011] Preferably, the dead zone control module performs a spatial truncation procedure on the airflow coupling matrix between the adjacent temperature zones, forcibly overwriting the cross-coupling coefficient of temperature zones that are not directly adjacent in physical space to zero, retaining the self-feedback coefficient on the main diagonal and the mutual coupling data of the temperature zones that are adjacent on the upper and lower sides, and simplifying the airflow coupling matrix between the adjacent temperature zones into a tri-diagonal matrix. Extract the total opening change of adjacent upstream and downstream temperature zones in the current control cycle, and calculate the temperature disturbance by combining it with the simplified airflow coupling matrix between adjacent temperature zones. The dynamic dead zone threshold is calculated by multiplying the temperature disturbance by the safety margin factor and adding it to the static dead zone constant. If the dynamic dead zone threshold is greater than the preset maximum tolerance limit parameter, then the dynamic dead zone threshold is forcibly truncated and clamped to the maximum tolerance limit parameter.

[0012] A second aspect of the present invention provides a method for adaptive control of parameters of food production equipment, applied to the adaptive control system for parameters of food production equipment as described in any one of the first aspects of the present invention, characterized by comprising the following steps: The continuous material is divided into virtual material slices, and the absolute spatial coordinates and instantaneous linear velocity of the virtual material slices are updated based on the absolute encoder data. The ambient temperature and circulating fan operating frequency of the virtual material slice are obtained based on the absolute spatial coordinates, and the actual dynamic enthalpy accumulation of the virtual material slice is calculated. When the virtual material slice reaches the end of the current physical temperature zone, the spatial moving average deviation is calculated based on the error between the actual dynamic enthalpy accumulation and the target enthalpy threshold, and a feedforward compensation opening increment for the downstream adjacent physical temperature zone is generated. The absolute spatial coordinate trigger point is established based on the time hysteresis parameter and the instantaneous linear velocity. When the virtual material slice reaches the absolute spatial coordinate trigger point, the feedforward compensation opening increment is sent to the downstream adjacent physical temperature zone. When issuing the feedforward compensation opening increment, the static dead zone constant of the current physical temperature zone controller is extended to a dynamic dead zone threshold based on the cross-coupling coefficient, and the output change is frozen.

[0013] Preferably, in the step of dividing continuous material into virtual material slices, the main control unit instantiates a one-dimensional structure array containing a sequence number, a furnace entry timestamp, and a current position variable in sequence according to a fixed physical length constant; The product of the current instantaneous linear velocity and the scanning cycle is accumulated as the physical displacement increment into the absolute spatial coordinates of the previous cycle; When it is determined that the absolute spatial coordinates of a certain slice exceed the end boundary of the oven, the memory release mechanism is triggered to clear the structure data corresponding to the slice and reclaim the serial number.

[0014] Preferably, in the step of establishing the absolute spatial coordinate trigger point, the main control unit performs a moving average filtering process on the acquired instantaneous linear velocity to obtain the effective running linear velocity, and determines whether the effective running linear velocity is lower than a preset minimum maintenance threshold. If the value falls below the preset minimum maintenance threshold, the current spatiotemporal decoupling computation task is suspended and the feedforward instruction delivery port is locked.

[0015] Preferably, in the step of issuing the feedforward compensation opening increment, after the main control unit determines that the absolute spatial coordinates of the target virtual material slice have reached the absolute spatial coordinate trigger point, it simultaneously confirms that the pressure of the downstream temperature zone gas pipeline is within the preset normal operating range and that the combustion controller has not triggered the hardware interlock alarm, and activates the transmission port to issue the feedforward compensation opening increment to the gas proportional valve actuator in the downstream temperature zone.

[0016] Preferably, freezing the change in the output of the current physical temperature zone controller requires the following simultaneous conditions: The absolute value of the real-time deviation between the current actual ambient temperature and the set target temperature is within the dynamic dead zone threshold. The current temperature zone has not experienced a unidirectional monotonic temperature drift that lasts for a duration exceeding a preset anti-disturbance time threshold. The criteria for determining the unidirectional monotonic temperature drift are as follows: The temperature discrete difference sign remains consistent across multiple consecutive sampling periods, and the cumulative duration exceeds the preset anti-disturbance time threshold.

[0017] A second aspect of the present invention provides a method for adaptive control of parameters of food production equipment, comprising the following steps: The continuous material is divided into virtual material slices, and the absolute spatial coordinates and instantaneous linear velocity of the virtual material slices are updated based on the absolute encoder data. The ambient temperature and circulating fan operating frequency of the virtual material slice are obtained based on the absolute spatial coordinates, and the actual dynamic enthalpy accumulation of the virtual material slice is calculated. When the virtual material slice reaches the end of the current physical temperature zone, the spatial moving average deviation is calculated based on the error between the actual dynamic enthalpy accumulation and the target enthalpy threshold, and a feedforward compensation opening increment for the downstream adjacent physical temperature zone is generated. The absolute spatial coordinate trigger point is established based on the time hysteresis parameter and the instantaneous linear velocity. When the virtual material slice reaches the absolute spatial coordinate trigger point, the feedforward compensation opening increment is sent to the downstream adjacent physical temperature zone. When issuing the feedforward compensation opening increment, the static dead zone constant of the current physical temperature zone controller is extended to a dynamic dead zone threshold based on the cross-coupling coefficient, and the output change is frozen.

[0018] Preferably, in the step of dividing continuous material into virtual material slices, the main control unit instantiates a one-dimensional structure array containing a sequence number, a furnace entry timestamp, and a current position variable in sequence according to a fixed physical length constant; the product of the current instantaneous linear velocity and the scanning cycle is accumulated as the physical displacement increment into the absolute spatial coordinates of the previous cycle; when it is determined that the absolute spatial coordinates of a certain slice exceed the end boundary of the oven, the memory release mechanism is triggered to clear the structure data corresponding to the slice and reclaim the sequence number.

[0019] Preferably, in the step of establishing the absolute spatial coordinate trigger point, the main control unit performs a moving average filtering process on the acquired instantaneous linear velocity to obtain the effective running linear velocity, and determines whether the effective running linear velocity is lower than a preset minimum maintenance threshold; if it is lower than the preset minimum maintenance threshold, the current spatiotemporal decoupling calculation task is suspended and the feedforward instruction sending port is locked.

[0020] Preferably, in the step of issuing the feedforward compensation opening increment, after the main control unit determines that the absolute spatial coordinates of the target virtual material slice have reached the absolute spatial coordinate trigger point, it simultaneously confirms that the pressure of the downstream temperature zone gas pipeline is within the preset normal operating range and that the combustion controller has not triggered the hardware interlock alarm, and activates the transmission port to issue the feedforward compensation opening increment to the gas proportional valve actuator in the downstream temperature zone.

[0021] Preferably, freezing the output change of the current physical temperature zone controller requires the following conditions to be met simultaneously: the absolute value of the real-time deviation between the actual ambient temperature of the current temperature zone and the set target temperature is within the dynamic dead zone threshold; the current temperature zone does not experience unidirectional monotonic temperature drift that lasts for more than a preset anti-disturbance time threshold, wherein the unidirectional monotonic temperature drift is defined as: the temperature discrete difference sign remains consistent for multiple consecutive sampling cycles, and the cumulative duration exceeds the preset anti-disturbance time threshold.

[0022] This invention provides an adaptive control system and method for parameters of food production equipment. It has the following beneficial effects: 1. This invention divides continuous material into virtual material slices and calculates the actual dynamic enthalpy accumulation by combining their absolute spatial coordinates, ambient temperature and fan operating frequency. This changes the reference object of the control system from the single equipment ambient temperature to the actual heat absorption of the material itself, objectively restoring the heating state of the material during the production process and improving the accuracy of the equipment's assessment of the material's processing maturity.

[0023] 2. Based on the instantaneous linear velocity of the material and the time lag parameter of the downstream physical temperature zone, the present invention calculates the absolute spatial coordinate trigger point in reverse. When the material reaches the trigger point, the feedforward compensation opening increment is sent to the downstream temperature zone in advance, which offsets the inherent physical response delay of the thermal equipment itself. This makes the effective time of temperature adjustment in the downstream temperature zone relatively synchronized with the actual time of material entering the temperature zone, thus avoiding the fluctuation of processing quality caused by the response lag of the control system.

[0024] 3. While issuing feedforward compensation commands to downstream adjacent temperature zones, this invention calculates temperature disturbances by combining the airflow coupling matrix and the total opening change. It extends the static dead zone constant of the current physical temperature zone controller to a dynamic dead zone threshold to temporarily freeze output changes, thereby cutting off the misleading effect of hot air crossflow caused by multi-temperature zone connectivity on the control system, preventing repeated oscillations and adjustments between adjacent temperature zones, and maintaining the stability of the overall temperature control. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3A comparative curve of temperature response after applying a sudden increase in moisture to the present invention; Figure 4 This is a comparison chart of the probability density distribution of measured moisture content in the products produced by this invention. Detailed Implementation

[0026] 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.

[0027] See attached document Figure 1 This invention provides an adaptive control system for parameters of food production equipment, comprising: a main conveyor belt, an absolute encoder, multiple independently controlled physical temperature zones, and a main control unit.

[0028] An absolute encoder is mounted on the main conveyor belt drive shaft. Each physical temperature zone is equipped with a thermocouple, a circulating fan, a frequency converter controlling the circulating fan, and a proportional gas valve controlling the gas input. The main control unit establishes electrical communication connections with each sensor and actuator.

[0029] The main control unit's program memory is equipped with a slice tracking module, a thermal enthalpy integration module, a batch aggregation module, a compensation triggering module, and a dead zone control module.

[0030] The slice tracking module is used to initialize the static parameter matrix, which includes the target enthalpy threshold array for each physical temperature zone, the heat capacity response time bias matrix, the thermal dispersion field width matrix, and the airflow coupling matrix between adjacent temperature zones. Based on a fixed physical length constant, the slice tracking module divides continuous material into serialized virtual material slices, samples the pulse change rate of the absolute encoder to calculate the instantaneous linear velocity of the conveyor belt, performs time-domain integration on the instantaneous linear velocity, and updates the absolute spatial coordinates of each virtual material slice within the oven in real time.

[0031] The enthalpy integration module is used to read the ambient temperature and circulating fan operating frequency at the current location of each virtual material slice via thermocouples and frequency converters. The enthalpy integration module calls the mapping function between the convective heat transfer coefficient and the fan frequency to calculate the instantaneous heat flux, and discretizes and accumulates the instantaneous heat flux in the time domain to obtain the actual dynamic enthalpy accumulation of each virtual material slice.

[0032] The batch aggregation module determines the spatial sliding window capacity by combining the width matrix of the thermal dispersion field of the current physical temperature zone when the virtual material slice reaches the end boundary of the current physical temperature zone. The batch aggregation module calculates the spatial moving average deviation of the enthalpy error of the virtual material slice within the window and uses a control gain function to generate the feedforward compensation opening increment for the gas proportional valve of the downstream adjacent physical temperature zone. If the calculation determines that no compensation is needed, the feedforward compensation opening increment is set to zero.

[0033] The compensation triggering module extracts the time hysteresis parameter corresponding to the downstream physical temperature zone in the heat capacity response time bias matrix, and multiplies the instantaneous linear velocity by the time hysteresis parameter to obtain the spatial advance triggering distance. The compensation triggering module uses the initial physical coordinates of the downstream physical temperature zone minus the spatial advance triggering distance to establish the absolute spatial coordinate triggering point. When it is determined that the absolute spatial coordinates of the target virtual material slice have reached the absolute spatial coordinate triggering point, the compensation triggering module sends a feedforward compensation opening increment command to the gas proportional valve of the downstream physical temperature zone.

[0034] The dead zone control module is used to extract the proportional constant from the airflow coupling matrix to calculate the parasitic temperature rise prediction within the same calculation cycle as the issuance of the feedforward compensation opening increment command. The dead zone control module extends the static dead zone constant of the current physical temperature zone controller to a dynamic dead zone, freezing the output change of the current physical temperature zone controller as long as the actual temperature rise does not exceed the dynamic dead zone boundary. When the virtual material slice batch crosses the effective working area of ​​the downstream physical temperature zone, the dead zone control module cancels the feedforward compensation opening increment command and restores the static dead zone constant.

[0035] See attached document Figure 2 This invention provides a method for adaptive control of parameters in food production equipment, comprising the following steps: S10, the main control unit initializes the static parameter matrix. The static parameter matrix includes the target enthalpy threshold array for each physical temperature zone, the heat capacity response time offset matrix, the thermal dispersion field width matrix, and the airflow coupling matrix between adjacent temperature zones. S20: The main control unit divides the continuous material into serialized virtual material slices based on a fixed physical length constant, samples the pulse change rate of the absolute encoder to calculate the instantaneous linear velocity of the conveyor belt, and performs time domain integration on the instantaneous linear velocity to update the absolute spatial coordinates of each virtual material slice in the oven in real time. S30, the main control unit reads the ambient temperature and circulating fan operating frequency of each virtual material slice at its current position through thermocouples and frequency converters, calls the convective heat transfer coefficient and fan frequency mapping function to calculate the instantaneous heat flux, and discretizes and accumulates the instantaneous heat flux in the time domain to obtain the actual dynamic enthalpy accumulation. S40, when the virtual material slice reaches the end boundary of the current physical temperature zone, the main control unit determines the capacity of the spatial sliding window by combining the width matrix of the thermal dispersion field of the current physical temperature zone, calculates the spatial moving average deviation of the enthalpy error of the virtual material slice within the window, and uses the control gain function to generate the feedforward compensation opening increment of the gas proportional valve for the downstream adjacent physical temperature zone. S50, the main control unit extracts the time hysteresis parameter corresponding to the downstream physical temperature zone in the thermal capacity response time bias matrix, multiplies the instantaneous linear velocity with the time hysteresis parameter to obtain the spatial advance trigger distance, and uses the initial physical coordinates of the downstream physical temperature zone minus the spatial advance trigger distance to establish the absolute spatial coordinate trigger point. S60, when the absolute spatial coordinates of the target virtual material slice reach the absolute spatial coordinate trigger point, the main control unit sends a feedforward compensation opening increment command to the downstream physical temperature zone gas proportional valve. S70, the main control unit extracts the proportional constant in the airflow coupling matrix to calculate the parasitic temperature rise prediction, expands the static dead zone constant of the current physical temperature zone controller to a dynamic dead zone, freezes the output change of the current physical temperature zone controller when the actual temperature rise does not exceed the dynamic dead zone boundary, and cancels the feedforward compensation opening increment command and restores the static dead zone constant after the virtual material slice batch crosses the effective working area of ​​the downstream physical temperature zone.

[0036] In the actual production process of a continuous tunnel oven, materials are typically arranged in a continuous strip or dense array. To achieve accurate parameter calculations for the moving materials, based on the general principle of discretization modeling, the main control unit needs to convert the continuous physical object into a discrete data object. In this embodiment, the main control unit has a preset fixed physical length constant, which is used to divide the continuously entering materials into equal intervals at the logical level.

[0037] As a preferred approach, the value range of the fixed physical length constant is determined jointly based on the width of the thermal dispersion field of the downstream actuator and the available memory space of the main control unit, ensuring control accuracy while avoiding the risk of data overflow. The main control unit creates a one-dimensional array of structures in its own program memory or data block. When material moves into the oven inlet with the conveyor belt, the main control unit instantiates the corresponding structure in memory sequentially according to the physical length constant; each structure represents a virtual material slice. This structure encapsulates the slice's sequence number, the oven entry timestamp, and the current position variable. Through the configuration of the above memory data structures, the system establishes a following coordinate reference system based on the moving material entity.

[0038] To determine the precise position of each virtual material slice inside the oven in real time, the system relies on an absolute encoder to acquire the motion parameters of the conveyor belt. The high-speed counter module of the main control unit collects the pulse changes output by the absolute encoder at a set hardware interrupt cycle.

[0039] To improve the time alignment accuracy of multi-source data, the main control unit locks the instantaneous linear velocity value within the same interrupt cycle of acquiring pulse change, so as to eliminate high-frequency interference caused by mechanical vibration.

[0040] After acquiring the instantaneous linear velocity of the conveyor belt, the main control unit initiates kinematic tracking calculations for each virtual material slice. Based on the physical causal relationship in classical kinematics that displacement equals the integral of velocity over time, a target virtual material slice moving within the furnace is defined. The main control unit dynamically calculates the current absolute spatial coordinates of this target virtual material slice through time-domain integration. The specific integration formula is as follows: ; In the formula, Indicates the first The virtual material slice number is at the current moment. The absolute spatial coordinates of the distance from the oven entrance; Indicates the first The initial moment when the virtual material slice physical entity crosses the oven entrance; Indicates the conveyor belt in time variable The instantaneous linear velocity at that point; For continuous differential variables that are integrals over time; The sequence number index for the virtual material slice.

[0041] Considering that industrial programmable logic controllers and other main control units operate in a digital discrete control environment, the above continuous integral equations are transformed into discrete-time coordinate accumulation logic at the specific program operation level.

[0042] In this embodiment, the main control unit executes the tracking task with a fixed program scan cycle. Within each program scan cycle, the main control unit extracts the product of the current instantaneous linear velocity and the scan cycle to obtain the physical displacement increment within that cycle. The main control unit traverses the structure array in memory and accumulates the physical displacement increment into the absolute spatial coordinates of each valid virtual material slice in the previous cycle.

[0043] To ensure the integrity of the algorithm logic, the main control unit compares the absolute spatial coordinates of each virtual material slice with the total physical length of the oven in real time after the accumulation calculation. When it is determined that the absolute spatial coordinates of a virtual material slice exceed the end boundary of the oven, the main control unit automatically triggers the memory release mechanism, clears the structure data corresponding to the slice and reclaims the sequence number, thereby avoiding system memory overflow.

[0044] By using this discrete step-by-step displacement accumulation method, the main control unit can accurately map the real-time distribution of all effective virtual material slices in the furnace in physical space in memory, providing a reliable spatial reference for subsequent dynamic enthalpy calculation.

[0045] Based on the established servo coordinate reference system, the system needs to further calculate the actual heating process of the moving material inside the oven. Traditional tunnel ovens mostly employ static feedback regulation for a fixed physical space. These systems typically rely solely on instantaneous temperature data from a single temperature zone, failing to detect the accumulated heat deviation due to long-distance material transport. In this embodiment, the system uses an energy tracking algorithm from the perspective of the servo material, utilizing dynamic enthalpy integrals to accurately characterize the evolution of the material's internal thermodynamic state, thereby providing precise data support for subsequent cross-temperature zone feedforward decoupling control.

[0046] Based on the general calculation principle of the thermodynamic equation of state, the instantaneous heat absorption rate of a material depends on the temperature and flow field distribution of its microenvironment. Therefore, the main control unit needs to acquire the environmental state parameters of each virtual material slice in real time. Since the thermocouples and circulating fans inside the oven are typically distributed as discrete points, the main control unit must establish spatial matching logic between moving coordinates and fixed sensors. When a target virtual material slice is determined to have reached the target physical temperature zone, the main control unit extracts the current absolute spatial coordinates of that slice.

[0047] As a preferred approach, the main control unit does not rely solely on the readings of a single sensor. Instead, it employs a distance-weighted interpolation algorithm to dynamically compare the absolute spatial coordinates with the physical installation positions of two adjacent thermocouples. Based on the relative positional ratio of the slice to each thermocouple, the system performs a weighted average of the real-time readings from the two thermocouples to calculate the equivalent ambient temperature at the slice's current location. To ensure the completeness of the underlying division operation, the main control unit pre-determines the absolute difference between the virtual material slice's absolute spatial coordinates and the thermocouple's physical coordinates before performing the relative positional ratio calculation. If this difference approaches zero, the main control unit directly uses the actual reading of the thermocouple at that location, thus avoiding computational overflow anomalies caused by a zero denominator.

[0048] Simultaneously, the main control unit reads the operating frequency of the frequency converter controlling the circulating fan within this temperature range. Considering the inherent time difference between the analog sampling period of the thermocouple and the communication reading period of the frequency converter, the main control unit establishes a timestamp-based multi-source data alignment buffer queue. At the same time, the main control unit extracts the equivalent ambient temperature and frequency converter operating frequency with the same clock mark from the buffer queue to ensure the synchronization of operating conditions in the heat flux calculation.

[0049] After acquiring multi-source environmental data, the main control unit calculates the instantaneous heat flux. Considering that forced hot air convection is the main heat transfer path in modern tunnel ovens, the operating speed of the circulating fan directly determines the magnitude of the convective heat transfer coefficient. The main control unit's program memory pre-configures a mapping function between the convective heat transfer coefficient and the fan operating frequency.

[0050] Based on Newton's law of cooling, the intensity of heat transfer between a fluid and a solid surface is positively correlated with the temperature difference between them. Using this principle, the main control unit calculates the instantaneous heat flux of the virtual material slice at the current moment. The specific calculation formula is as follows: ; In the formula, Indicates the first The current moment of the virtual material slice. The instantaneous heat flux; This represents the built-in convective heat transfer coefficient mapping function; This indicates the inverter operating frequency associated with the absolute spatial coordinate position of the slice; This represents the equivalent ambient temperature at the current location, obtained through interpolation. This indicates the set material-based reference temperature constant.

[0051] In this embodiment, the basic reference temperature constant for the material is determined by weighting the initial ambient temperature measurement of the continuous material before it enters the oven, combined with the gelatinization or Maillard reaction phase transition reference temperature for special food processing, to ensure that the calculated temperature difference conforms to the actual physical heat exchange law. This formula clarifies the specific source of energy acquisition by the material, providing basic microdata for subsequent determination of whether the material's heat absorption meets the standard.

[0052] After obtaining the instantaneous heat flux, the system performs cumulative calculations using discrete time steps to obtain the complete heating history of the slice since it entered the furnace. Within each control scan cycle, the main control unit multiplies the currently calculated instantaneous heat flux by the duration of that scan cycle to obtain the single-step heat absorption. To prevent sudden readings caused by electromagnetic interference or loose wiring of individual thermocouples in industrial settings, which could lead to integral divergence, the main control unit performs outlier checks on the single-step heat absorption before accumulation. The main control unit has a built-in effective heat absorption range set based on process physical limits. When the single-step heat absorption is determined to exceed this effective range, the system automatically discards the current distorted data and maintains the effective value from the previous cycle.

[0053] In this embodiment, the effective heat absorption range is not a single fixed value, but a weighted judgment range dynamically adjusted by the main control unit based on the theoretical specific heat capacity of the material at different heating stages. This avoids a single extreme value causing normal rapid heating to be misjudged as abnormal, and ensures the multi-dimensionality and rationality of the output result judgment.

[0054] After verifying the data's validity, the main control unit adds the single-step heat absorption to the independent data volume corresponding to the target virtual material slice, thus obtaining the actual dynamic enthalpy accumulation. The specific integral calculation formula is as follows: ; In the formula, Indicates the first The virtual material slice number is at the current moment. The actual dynamic cumulative enthalpy; The index variable for the discrete accumulation step size; This represents the total number of program scan cycles experienced by the virtual material slice from the moment it enters the oven until the current moment. The fixed program scan cycle time is set by the main control unit. Through the above process, the system completes the data mapping from macroscopic ambient temperature to microscopic material internal energy state. This cumulative amount truly reflects the total thermal efficiency absorbed by the slice, providing a direct mathematical evaluation index for determining whether downstream temperature zone feedforward compensation is needed.

[0055] Based on the actual dynamic enthalpy accumulation of each virtual material slice output by the aforementioned dynamic enthalpy integration module, the main control unit can obtain the material's heating state at a microscopic level. In actual industrial control, there is a spatial scale difference between the physical output characteristics of the actuator and the microscopic data.

[0056] In this embodiment, the heating source of the tunnel oven is typically configured as a burner controlled by a gas proportional valve combined with a circulating fan. Due to thermodynamic and fluid dynamic characteristics, the spatial distribution of the heat flow field blown out by the circulating fan inside the furnace is not a perfectly isolated rectangular boundary, but rather exhibits a nonlinear distribution with a high temperature at the center and a gradual decrease towards both sides. If the main control unit directly drives the high-inertia gas proportional valve at high frequency for compensation based solely on the enthalpy deviation of a single, tiny virtual material slice, it will cause mechanical oscillations in the actuator and lead to divergence in the actual furnace temperature control curve.

[0057] To address the aforementioned spatial scale difference issue, the main control unit introduces a virtual batch aggregation mechanism based on physical field properties. In this embodiment, the thermal dispersion field width matrix is ​​pre-initialized in the main control unit's program memory. This matrix corresponds to and represents the first... The equivalent physical length of the burner's effective thermal radiation and forced convection heat transfer within each independent physical temperature zone, extending in absolute space. (The subscript is omitted as it is not part of the original text.) The serial number index represents the physical temperature zone, and its value range corresponds to the total number of physical temperature zones configured in the actual oven, thereby ensuring the accurate correspondence between spatial data and hardware equipment.

[0058] Regarding the determination logic of specific values ​​in the thermal dispersion field width matrix, as a preferred method, the thermal dispersion field width values ​​corresponding to each temperature zone are calibrated offline based on steady-state thermal field distribution tests of the equipment under full-load conditions. During the equipment commissioning phase, engineers collect spatial temperature decay curves extending from the central heat source of each temperature zone along the conveyor belt's running direction. The main control unit or host computer system extracts the two physical coordinate nodes corresponding to the temperature field decaying to a specific percentage threshold of the peak value, calculates the absolute spatial distance between the two nodes, and writes it as a specific value into the thermal dispersion field width matrix corresponding to that temperature zone. To ensure the rationality of parameter settings and the robustness of algorithm calculations, the value range of this specific percentage threshold is usually set to 30% to 50% of the central peak temperature. The basis for setting this value range is that the temperature boundary of this range can fully encompass the main effective heat energy generated by a single combustion action, while effectively eliminating ineffective low-heat areas at the edges that are easily disturbed by external cold airflow, thus avoiding the introduction of redundant interference.

[0059] The technical purpose of setting the thermal dispersion field width matrix is ​​to convert continuous spatial physical properties into a digital filtering benchmark. This thermal dispersion field width provides a physical scale reference for the main control unit to subsequently divide the spatial sliding window. Based on the above parameter configuration, the main control unit establishes the spatial mapping relationship between the actuator's physical action range and the virtual material slice sequence. Through this mapping, the system can aggregate discrete virtual material slice data into virtual batches that conform to the output capabilities of the equipment hardware. During the algorithm calculation process, low-pass filtering of the spatial dimension is achieved, making the generated control commands smoother and more in line with the objective laws of fluid physics.

[0060] Based on the aforementioned thermal dispersion field width matrix, the system needs to further convert the microscopic discrete slice data into macroscopic compensation commands. In industrial control theory, directly inputting the original deviation with high-frequency noise into the actuator usually leads to system oscillation or even control divergence. Therefore, this embodiment introduces a spatial sliding filter mechanism. This mechanism uses the thermal dispersion field width as a physical scale to batch aggregate the virtual material slice sequence, thereby filtering out single data distortions caused by local airflow disturbances or extreme values ​​of sensor measurements.

[0061] When the main control unit determines that the target virtual material slice has reached the end boundary of the current physical temperature zone, the system immediately triggers the deviation evaluation program. To determine the data scale for evaluation, the main control unit extracts the physical length value corresponding to the current physical temperature zone in the thermal dispersion field width matrix.

[0062] Since the fixed physical length constant used for segmentation in the initial stage of the system is a strictly positive scalar, the main control unit directly divides the width of the thermal dispersion field by this fixed physical length constant. To ensure the completeness of the low-level computation in the discrete digital environment, the main control unit performs an up-rounding operation on the result of this division operation, thereby calculating the spatial sliding window capacity. This capacity value clarifies the total number of virtual material segments included in a single deviation evaluation at the physical space level, effectively mapping the physical range of the actuator to the logical data queue.

[0063] After establishing the capacity of the sliding window, the main control unit uses the target virtual material slice that has just reached the end boundary of the temperature zone as the sequence center and traces adjacent virtual material slices in both directions. To prevent the underlying array index from going out of bounds during the tracing process, the main control unit performs boundary verification logic before extracting data. If the calculated forward and backward tracing indices exceed the range of the currently instantiated valid slice sequence in memory, the system automatically truncates the sliding window to the actual usable data boundary.

[0064] Subsequently, the main control unit extracts the actual dynamic enthalpy accumulation of all slices within the window range from memory and retrieves the target enthalpy threshold set for the current temperature zone from the static parameter matrix.

[0065] As a preferred approach, the target enthalpy threshold is determined offline based on the sum of the latent heat and sensible heat of the biochemical reaction phase change that the material needs to complete in this temperature range under a specific process formulation. The main control unit calculates the difference between the target enthalpy threshold and the actual dynamic enthalpy accumulation for each slice within the window, and then calculates the arithmetic mean of the differences for all slices to obtain the spatial moving average deviation.

[0066] The engineering purpose of the above steps is to transform the system's object of study from a single isolated particle into a batch of materials that conforms to the actual heat transfer physical field distribution, thereby effectively smoothing high-frequency disturbances in the microscopic calculation process.

[0067] After acquiring the spatial moving average deviation, the main control unit generates a feedforward compensation opening increment for the proportional valves of the gas in the downstream adjacent physical temperature zone based on this deviation value. To avoid mechanical wear caused by frequent fine-tuning of the combustion actuator due to small deviations, the main control unit incorporates a control gain function with static limits. The main control unit compares the absolute value of the spatial moving average deviation with the preset enthalpy tolerance dead zone.

[0068] The enthalpy tolerance dead zone is determined by weighting the inherent measurement error of the sensor with the acceptable range of process fluctuations, thereby ensuring the multidimensionality and rationality of the judgment results.

[0069] If the absolute value of the deviation is less than the enthalpy tolerance dead zone, the heating state of the current batch of material is determined to be within the process-allowed range, and the system directly sets the feedforward compensation increment to zero, while the downstream temperature zone maintains its original basic closed-loop control state. If the absolute value of the deviation exceeds the enthalpy tolerance dead zone, the main control unit calculates the specific compensation amount using the set proportional gain coefficient. The specific compensation amount calculation formula is as follows: ; In the formula, This indicates that the system generates and plans to distribute the data to the specified number of nodes. The feedforward compensation opening increment of the gas proportional valve for each physical temperature zone (i.e., downstream temperature zone); Indicates the first Feedforward proportional gain coefficients corresponding to each physical temperature zone; Indicates the first The spatial moving average deviation is calculated using the virtual material slice as the evaluation benchmark point. This is a sign function used to extract the positive or negative polarity of the deviation to determine whether the valve should be opened or closed. This is a preset enthalpy tolerance dead zone threshold. The difference between the spatial moving average deviation and the enthalpy tolerance dead zone threshold is calculated as: The system multiplies the feedforward proportional gain coefficient by the deviation difference based on the positive or negative sign of the spatial moving average deviation to obtain the feedforward compensation opening increment.

[0070] In this embodiment, the feedforward proportional gain coefficient is calibrated based on the inherent flow characteristic curve of the downstream temperature zone gas proportional valve, and its value is strongly correlated with the slope of the linear mapping from valve opening to gas heat release power. Through the above logic based on dead zone determination and proportional gain combination, the system not only achieves accurate quantitative evaluation of the underheating or overheating trend of the material, but also ensures the stability of the control commands of the downstream actuator and prevents over-adjustment.

[0071] Based on the feedforward compensation opening increment calculated by the aforementioned virtual batch aggregation and deviation evaluation mechanism, the main control unit needs to further determine the exact timing of issuing the control command. As a typical large-scale thermal equipment, the continuous tunnel oven's internal burners, circulating fans, and furnace insulation structure all exhibit significant physical thermal inertia.

[0072] Based on the principle of pure time delay in classical control theory, when the actuator receives the adjustment command, the physical changes in gas flow, the release of combustion heat energy, and the mass transfer of hot air convection all need to go through an objectively existing lag time before the actual adjustment effect can be reflected in the temperature field of the target space.

[0073] If the system only issues a compensation command when the material reaches the physical boundary of the downstream temperature zone, the actual compensation heat will not be synchronized with the target material batch on the time axis, thus causing control misalignment.

[0074] To address the decoupling issue between spatial movement and time lag, in this embodiment, a thermal capacity response time bias matrix is ​​pre-configured in the main control unit's program memory. The specific data structure of this matrix is ​​a one-dimensional array, with subscripts... This represents the sequence index of the physical temperature zone. Each element within this matrix corresponds to the index of the [number]th [temperature zone] within the oven. The thermodynamic time hysteresis property inherent in a special physical temperature region.

[0075] The physical meaning of this time hysteresis attribute is the absolute time span from the initial moment when the main control unit sends an opening change command to the gas proportional valve of the downstream physical temperature zone until the main equivalent ambient temperature in the physical temperature zone completes the dynamic transition and reaches a new steady-state range.

[0076] The system does not rely on a single theoretical estimation to determine the specific values ​​in the thermal capacity response time bias matrix. As a preferred approach, engineers perform step response tests on each independently controlled physical temperature zone during the offline commissioning phase of the equipment.

[0077] During the calibration process, under full-load standard ventilation conditions, the main control unit applies a fixed-amplitude step opening command to the gas proportional valve in a specific physical temperature zone and simultaneously starts an internal high-frequency timer. The main control unit continuously collects the instantaneous temperature feedback values ​​of the thermocouples within this temperature zone. Due to the sensor noise floor in industrial environments, the main control unit does not directly use the instantaneous difference between two adjacent cycles, but instead establishes a time difference window spanning multiple sampling cycles. The main control unit uses the temperature difference between the two ends of this time difference window divided by the total window duration as the result of the discrete derivative calculation; before the calculation, the system pre-determines whether the total window duration is greater than zero to prevent division-by-zero overflow caused by underlying hardware malfunctions.

[0078] To avoid misjudgments caused by a single extreme value, the main control unit establishes multi-dimensional steady-state judgment conditions. When the discrete derivative calculated by the system crosses the preset zero-reaching dead zone, and the actual ambient temperature stabilizes within the target steady-state tolerance band, the main control unit cuts off the timer and records the total transition time consumed. The values ​​of the aforementioned zero-reaching dead zone and target steady-state tolerance band are determined by a weighted average of the electrical noise floor amplitude of the on-site temperature sensor and the static error allowed by the special baking process.

[0079] The main control unit will then solidify the final calculated average transition time into the thermal capacity response time bias matrix for that physical temperature region. The specific parameters corresponding to that.

[0080] After establishing the aforementioned matrix, the main control unit can perform spatiotemporal decoupling operations using the extracted time parameters. When the system generates the feedforward compensation opening increment for the downstream physical temperature zone, the main control unit performs an increment operation based on the current temperature zone's sequence number index, thereby addressing and extracting the corresponding downstream temperature zone's time hysteresis parameter in the thermal capacity response time bias matrix.

[0081] To ensure the integrity of the underlying addressing logic, the main control unit will pre-determine whether the sequence number of the current temperature zone is equal to the sequence number of the last temperature zone at the end of the oven before performing the index increment operation. If it is determined that the target virtual material slice is currently in the last temperature zone, the system automatically locks the feedforward compensation trigger logic of the current scanning cycle and overwrites the feedforward compensation opening increment to zero, thereby completely avoiding the risk of main control unit crashing due to array out-of-bounds errors.

[0082] By introducing this matrix, the system transforms the originally isolated physical time delay parameter into an effective logical variable that can participate in kinematic calculations, laying a solid data foundation for subsequent accurate measurement of spatial advance triggering distance.

[0083] Based on the aforementioned extracted downstream physical temperature zone heat capacity response time bias parameters, the main control unit enters the inverse calculation process of spatiotemporal decoupling. There is an objective spatiotemporal misalignment in heat transfer within the continuous tunnel oven. To eliminate this misalignment, the main control unit needs to convert the thermal inertial delay in the time domain into a physical advance distance in the spatial domain. This step is based on the general principle of kinematics, namely, before the target batch of material arrives at the physical boundary of the downstream temperature zone, the system advances a physical distance that precisely offsets the thermodynamic transition time, and pre-issues the feedforward compensation command to the downstream actuator, thereby synchronizing the moment when the material batch enters the downstream temperature zone with the moment when a new steady-state thermal field is established in that temperature zone on the time axis.

[0084] To perform the aforementioned reverse calculation, the main control unit needs to acquire the kinematic parameters of the main conveyor belt in real time. The main control unit periodically reads the operating frequency of the drive inverter controlling the main conveyor belt via the industrial communication bus.

[0085] Considering the data jitter present in industrial field communication buses, the main control unit performs a moving average filter on the instantaneous linear velocity after acquisition to extract the effective operating linear velocity that represents the current stable operating condition. To ensure the integrity of the control logic under extreme conditions, the main control unit has a built-in low-speed minimum value verification program.

[0086] The system determines whether the effective operating line speed is lower than the minimum maintenance threshold allowed by the process. This minimum maintenance threshold is typically set to 5% to 10% of the rated operating line speed, and is set to prevent false triggering caused by mechanical jams or temporary shutdowns due to abnormalities in the upstream process. If the effective operating line speed is determined to be lower than this threshold, the system automatically suspends the current spatiotemporal decoupling calculation task and locks the feedforward command sending port to prevent excessive heat accumulation in the target temperature zone due to command issuance but material stagnation; the system only proceeds to the subsequent calculation stage when the effective operating line speed is within the normal process range.

[0087] After confirming the kinematic state is valid, the main control unit converts the time hysteresis attribute into spatial lead. In this embodiment, the main control unit calls the downstream temperature zone time hysteresis parameter generated in the preceding steps and multiplies it by the effective operating linear velocity obtained in the current scan cycle. The specific reverse calculation formula is as follows: ; In the formula, Indicates that for the first The spatial advance trigger distance is calculated in reverse from each physical temperature zone (i.e., the downstream temperature zone); This represents the first element extracted from the thermal capacity response time bias matrix. Time hysteresis parameters for each physical temperature zone; This represents the effective linear speed of the main conveyor belt after filtering. The spatial advance trigger distance determined by this formula has the physical meaning of the additional spatial span that the system must allow the target material to slide at the end of the current temperature zone in order to compensate for the lag time in the heat build-up of the downstream temperature zone.

[0088] After obtaining the preliminary calculation results, the main control unit performs amplitude limiting and truncation processing based on physical boundaries. As a preferred method, the main control unit retrieves the current... The system determines whether the calculated advance trigger distance for the current temperature zone is greater than the mechanical construction length of that zone. If the determination is yes, it indicates that due to the high conveyor belt speed or long-cycle thermal hysteresis, the theoretically calculated trigger point has exceeded the current temperature zone range. In this case, the main control unit forcibly cuts off the advance trigger distance and clamps it to a preset length limit ratio of the actual mechanical construction length of the current temperature zone. In a preferred embodiment of the invention, this length limit ratio is preferably 90% (it can also be set to 80%, 85%, or 95%, etc., depending on the physical characteristics of the actual equipment, and is not strictly limited).

[0089] It should be noted that when the aforementioned limiting cutoff is triggered, it means that the theoretically required feedforward lead has exceeded the spatial span of a single adjacent temperature zone. At this point, the clamping cutoff mechanism essentially triggers the system's underlying degradation safety protection mode. Its purpose is to prioritize the determinism of cross-zone addressing logic and the absolute stability of system operation, by tolerating short-term heat shortages in downstream temperature zones under this extreme condition, in exchange for the crash-free operation of the entire furnace control network. A 10% margin is reserved here to ensure that the command trigger point always falls within the logically permissible physical space of the current temperature zone, preventing underlying addressing out-of-bounds errors and execution logic chaos.

[0090] As the material conveyor belt continues to operate, the main control unit enters the dynamic comparison and command issuance phase. The main control unit tracks the absolute coordinates of the center slice of the target virtual batch in real time and calculates the remaining physical distance of the batch from leaving the current temperature zone based on the absolute spatial coordinates of the current temperature zone's end boundary. Within each program scan cycle, the system dynamically compares this remaining physical distance with the pre-triggered spatial distance after limiting. To avoid malfunctions due to reliance on a single spatial extreme value, the main control unit's command issuance module employs multi-dimensional weighted judgment logic.

[0091] When the system determines that the remaining physical distance is less than or equal to the pre-triggered spatial distance, and simultaneously confirms that the downstream gas pipeline pressure is within the normal operating range and the combustion controller has not triggered a hardware interlock alarm, the main control unit activates the transmitting port. After meeting these combined conditions, the main control unit converts the previously generated feedforward compensation opening increment into an analog quantity or communication message and sends it to the gas proportional valve actuator in the downstream temperature zone. Through this mechanism, the system achieves precise feedforward intervention in both physical space and thermodynamic time dimensions.

[0092] Based on the aforementioned spatiotemporal decoupling launch mechanism, the system achieves feedforward compensation along the longitudinal time axis for a single temperature zone. However, the continuous tunnel oven is physically a single, interconnected cavity. Based on fluid dynamics principles, the circulating fans in each temperature zone inevitably drive hot air axially along the furnace's longitudinal direction during forced convection heat transfer, causing cross-temperature airflow interference between adjacent zones. If the gas proportional valve in a certain temperature zone performs a large-scale opening adjustment, the excess heat generated will flow into adjacent temperature zones with the high-speed airflow, causing the actual temperature of adjacent zones to passively deviate from the set value. This physical airflow coupling effect severely weakens the stability of the independent closed-loop control loops of each physical temperature zone, easily triggering a chain reaction of oscillations in the entire furnace temperature field.

[0093] To quantify and decouple the aforementioned complex fluid disturbances, in this embodiment, an airflow coupling matrix is ​​pre-initialized in the main control unit's program memory. This matrix is ​​a two-dimensional square matrix, and its internal elements represent the steady-state coupling influence weights of heat output changes in the specific source physical temperature zone within the oven on the ambient temperature of the target physical temperature zone. The introduction of this matrix transforms the invisible gaseous fluid dynamics parameters into a digital static gain network that can be directly invoked by the control algorithm.

[0094] Regarding the determination logic of specific values ​​in the airflow coupling matrix, the system does not rely on a single theoretical fluid simulation, but is constructed based on the actual physical response of the physical equipment. As a preferred approach, engineers perform offline cross-calibration tests based on single-variable step disturbances during the no-load constant temperature commissioning phase of the equipment. During the test, the system maintains all circulating fans in the oven at their rated speed and brings the entire furnace to an initial thermal equilibrium steady state. Subsequently, the main control unit manually applies fixed opening increment commands to the gas proportional valves in the source physical temperature zone, while locking the opening of the gas proportional valves in the other temperature zones to maintain their current state.

[0095] The main control unit continuously monitors and records the steady-state change of the actual ambient temperature in the target temperature zone and the actual change of the opening of the gas proportional valve in the source temperature zone when the system reaches a new thermal equilibrium steady state. The system calculates the static gain ratio of the two by dividing the above steady-state temperature change by the actual change of the opening. To prevent the valve from mechanically jamming during the digital division operation, which could lead to the valve not actually operating and causing a zero-denominator overflow anomaly, the main control unit pre-determines whether the absolute value of the actual change of the opening of the gas proportional valve in the source temperature zone is greater than a preset mechanical action dead zone threshold before performing the division. This mechanical action dead zone threshold is set based on the inherent transmission gear clearance parameters of the gas proportional valve.

[0096] If the absolute value is determined to be less than or equal to the mechanical action dead zone threshold, the main control unit determines that the applied step disturbance has failed to overcome the mechanical resistance. The system actively abandons this record and prompts the engineers to reapply a step command with a larger amplitude to prevent the extraction of false noise gain. The system will only perform subsequent division calculations if and only if the change is greater than the threshold. The specific matrix element calculation formula is as follows: ; In the formula, Indicates the first The first physical temperature region (source temperature region) is related to the first... Airflow coupling coefficient for each physical temperature zone (target temperature zone); subscript and These represent the serial number indices of the target temperature zone and the source temperature zone, respectively. Indicates the first The steady-state change of actual ambient temperature measured by sensors before and after the disturbance test for each physical temperature zone; Indicates that it is artificially applied to the first The actual change in the opening degree of the gas proportional valve in each physical temperature zone; This represents the dimensionless wind pressure correction coefficient determined by the current operating frequency of the main circulation fan.

[0097] Specifically, this wind pressure correction factor Numerically, it is equal to the ratio of the real-time operating frequency of the circulating fan to the rated operating frequency of the equipment. Its purpose is to compensate for the physical influence of changes in fan exhaust volume on the absolute velocity of axial airflow under different production processes, so that the matrix elements still have accurate physical mapping capabilities under non-rated wind speeds.

[0098] After establishing the basic computational logic, the system needs further structural optimization of the airflow coupling matrix to adapt to real-time control requirements. Considering the heat dissipation characteristics of the airflow flowing longitudinally inside the tunnel oven, the heat crosstalk between physically distant temperature zones is very weak and can be ignored. To reduce the processing load of matrix multiplication operations in subsequent control cycles and to completely avoid ill-conditioned singularities in the airflow coupling matrix caused by the accumulation of small disturbances, the main control unit has a built-in spatial truncation program. The system forcibly separates physically non-adjacent temperature zones (i.e., those whose judgment conditions satisfy absolute values). The cross-coupling coefficient (at time) is directly overwritten as zero.

[0099] Through the aforementioned spatial truncation process, the system simplifies the complete airflow coupling matrix into a strictly tridiagonal matrix, retaining only the self-feedback coefficients on the main diagonal and the mutual coupling data of the adjacent temperature zones on its upper and lower sides. The establishment of this airflow coupling matrix enables the main control unit to proactively predict the accompanying heat impact of surrounding temperature zone adjustments on the current temperature zone, laying a solid data foundation for the subsequent implementation of dynamic dead zone widening and suppression of unnecessary feedback compensation actions.

[0100] Based on the previously established airflow coupling matrix, the main control unit enters the substantive execution phase of decoupling control. In traditional independent closed-loop control architectures, when adjacent temperature zones experience significant thermal adjustments due to feedforward or feedback mechanisms, the accompanying airflow crosstalk directly causes transient jumps in the sensor feedback values ​​of the current temperature zone. If the closed-loop controller of the current temperature zone immediately performs reverse adjustment in response to this external disturbance, it can easily trigger control competition between adjacent temperature zones and system-level thermal oscillations. This control step is based on the general technical principle of dynamic tolerance adjustment.

[0101] Specifically, the system proactively extracts external disturbances and temporarily relaxes the temperature feedback control dead zone of the current temperature zone. Within the expected disturbance range, the system forcibly maintains the current actuator opening unchanged, thereby achieving passive decoupling at the physical level. To implement dynamic dead zone compensation, the main control unit collects the operating status of the gas proportional valves in adjacent physical temperature zones in real time during each control cycle.

[0102] In this embodiment, the main control unit extracts the upstream and downstream adjacent temperature zones (i.e., the first... and The system calculates the feedforward compensation opening increment and feedback regulation opening change of the gas proportional valve (for each temperature zone), and then sums their algebraic values ​​as the total opening change. Combining this with the corresponding cross-coupling coefficients in the airflow coupling matrix, the system calculates the expected intrusion into the current temperature zone. Temperature disturbance in each physical temperature zone.

[0103] To ensure the safety and integrity of the underlying addressing logic, the main control unit pre-determines the index sequence number of the current temperature zone before performing algebraic and accumulation operations. Is it located at the physical boundaries of the device? If determined... As the first physical temperature zone, the system automatically cuts out the appropriate temperature range. The index is used for addressing and the variable is set to zero; if the determination is made... For the terminal physical temperature zone, the system automatically trims the parameters. The index is used to address and the variable is set to zero, thereby completely avoiding array out-of-bounds crashes caused by addressing a non-existent physical temperature range.

[0104] To avoid sudden dead-zone changes caused by a single extreme value, the main control unit multiplies the calculated temperature disturbance by a safety margin coefficient and adds it to the system's inherent static dead-zone to calculate the real-time dynamic dead-zone threshold. The specific formula for calculating the dynamic dead-zone is as follows: ; In the formula, Indicates that for the first The dynamic dead zone threshold calculated in real time for each physical temperature zone; This represents the static dead zone constant that is pre-set based on the allowable steady-state error according to the process formula; This represents the system's preset safety margin coefficient; Represents the first in the airflow coupling matrix The adjacent temperature zone for the first The airflow coupling coefficient for the current temperature zone; Indicates the first The total opening change of the gas proportional valves in adjacent temperature zones during the current control cycle.

[0105] Considering the inherent time lag in physical airflow conduction, simply widening the dead zone during the current valve operation cycle will not cover the subsequent actual temperature jumps. Therefore, the main control unit, after calculating the temperature change caused by... The predicted temperature disturbance is then pushed into a delay buffer queue based on the system clock. The delay time parameter of this queue is set to the average physical transport time of airflow across adjacent temperature ranges. The system extracts the delay disturbance from the output of this queue at the current moment to calculate the current dynamic dead zone threshold. Alternatively, the main control unit can configure a time decay hold function for the amplified dynamic dead zone, so that it remains at a high level throughout the expected thermal disturbance period, thereby achieving precise alignment of control commands and physical disturbances on the time axis.

[0106] Safety margin coefficient in the above formula The value range is usually set to 1.2 to 1.5. Its physical setting aims to cover the nonlinear turbulent disturbance components in the fluid dynamics model that are not fully linearized, and to ensure that the dynamically relaxed dead zone can completely cover the actual temperature jump.

[0107] After calculating the dynamic dead zone threshold, the system must guard against the risk of the dead zone being infinitely amplified and causing runaway. As a preferred approach, the main control unit is configured with a maximum tolerance limit parameter. The main control unit compares the calculated dynamic dead zone threshold with this maximum tolerance limit parameter.

[0108] If the calculated dynamic dead zone threshold exceeds the limiting parameter, the system forcibly truncates it and clamps it to the highest tolerable limiting parameter. This limiting mechanism is designed to ensure the boundary safety of the control system, preventing irreversible process defects such as severe scorching or under-baking of the baking materials due to prolonged disconnection from the closed-loop control protection network in the temperature zone.

[0109] After establishing a safe and effective dynamic dead zone, the main control unit directly applies it to the state-closed-loop control logic of the current temperature zone. The system continuously calculates the absolute value of the real-time deviation between the actual ambient temperature and the set target temperature in the current temperature zone. The main control unit then compares this real-time deviation absolute value with the updated dynamic dead zone threshold. Multi-dimensional weighted judgment is performed. When the system determines that the absolute value of the real-time deviation is within the dynamic dead zone threshold, and simultaneously confirms that the current temperature zone has not experienced a unidirectional monotonic temperature drift that lasts for more than the set time threshold, the main control unit determines that the current temperature fluctuation is a reasonable crosstalk caused by adjacent temperature zones.

[0110] Regarding the underlying logic for determining the aforementioned unidirectional monotonic temperature drift, the main control unit continuously records the actual temperature deviation values ​​of multiple past sampling periods and calculates the discrete difference between the deviation values ​​of adjacent periods. The system determines that a true unidirectional monotonic temperature drift has occurred only when the signs of the discrete differences for multiple consecutive periods remain consistent (i.e., representing continuous positive divergence or continuous negative divergence) and the cumulative duration exceeds the anti-disturbance time threshold set by the system.

[0111] After satisfying the aforementioned composite decoupling conditions of no drift and within the dead zone, the system automatically freezes the feedback control calculation for the current temperature zone, maintaining the gas proportional valve opening unchanged from the previous cycle. Conversely, if the absolute value of the real-time deviation exceeds the dynamic dead zone threshold, or if the system detects a prolonged unidirectional physical temperature rise / fall trend even though the temperature is within the dead zone, the main control unit immediately unfreezes the system and reactivates the conventional closed-loop algorithm for active intervention. Through this dynamic dead zone mechanism, the system filters out spurious deviation signals caused by airflow coupling, significantly reducing mechanical wear and unnecessary control competition of the actuators while ensuring the stability of the overall furnace macroscopic temperature field. It should be noted that the conventional algorithms, data processing flows, and underlying hardware control logic not elaborated in detail in this paper are all well-known technologies in the field. Those skilled in the art can implement them in conjunction with existing mature solutions. To highlight the core logic of this solution, they will not be described in detail here.

[0112] Specific application examples: To further illustrate the technical solution of the present invention, the following describes the actual production conditions of a continuous cookie tunnel oven.

[0113] The continuous tunnel oven is set to a total length of 30 meters, divided into 5 independent physical temperature zones (Zone 1 to Zone 5) from the entrance to the exit. The rated linear speed of the conveyor belt is 1.5 m / min (i.e., 0.025 m / s). The fixed physical length constant set within the main control unit is 0.1 meters, meaning that a virtual material slice (e.g., a slice) is generated for every 0.1 meters the conveyor belt travels. ).

[0114] Phase 1: Enthalpy Tracking and Feedforward Generation When the cookie dough (sliced) has a high moisture content When the dough enters Zone 2 (the main baking proofing zone, set to a target temperature of 180°C), it absorbs a significant amount of heat due to the low initial temperature of the dough batch. When slicing... Upon reaching the end of Zone 2, the batch aggregation module of the main control unit sums the enthalpy of the slices within the sliding window. The calculation reveals that the actual cumulative dynamic enthalpy of this batch is only 380 kJ / kg, lower than the target enthalpy threshold of 420 kJ / kg set by the process, resulting in an enthalpy deviation of -40 kJ / kg. Based on the proportional gain coefficient, the main control unit calculates that a 5% increase in the gas proportional valve opening is needed in the downstream Zone 3 (dehydration and shaping zone) as feedforward compensation.

[0115] Phase Two: Spatiotemporal Decoupling and Precise Triggering Due to the thermal inertia of the burner and hot air system in Zone3, the time hysteresis parameter of Zone3 is extracted as 40 seconds based on the thermal capacity response time bias matrix. The main control unit calculates the spatial advance trigger distance: rice.

[0116] Therefore, when slicing When there is 1.0 meter remaining to the entrance of Zone 3 (at which point the physical entity is still sliding at the end of Zone 2), the main control unit sends a +5% opening command to Zone 3 in advance. When slicing... When you actually enter Zone3, the additional thermal field of Zone3 has just been established, achieving precise alignment between the thermal time axis and the material space axis.

[0117] Phase 3: Airflow Coupling and Dead Zone Freezing After the gas valve in Zone 3 was opened by 5%, excess hot air flowed longitudinally into Zone 4 along the furnace. The main control unit extracted the airflow coupling matrix and found that the airflow coupling coefficient from Zone 3 to Zone 4 was 0.3. The main control unit calculated that the predicted temperature disturbance that Zone 4 would face was +1.5℃.

[0118] Subsequently, the main control unit temporarily widened the original static dead zone (±1.0℃) of Zone4 unidirectionally to +2.5℃ along the positive half-axis. When the actual temperature probe of Zone4 briefly rose by 1.2℃ due to air leakage, because it fell within the widened dynamic dead zone, the closed-loop controller of Zone4 was frozen, and its gas proportional valve remained stationary. (To be sliced) After leaving, the thermal field recovers and the dynamic dead zone shrinks, completely avoiding the accidental valve closure and subsequent temperature fluctuations caused by external disturbances in Zone4.

[0119] Experimental verification and effect comparison: The experiment set up two control modes for comparison: Control group (traditional mode): Each physical temperature zone adopts independent traditional PID closed-loop temperature control, and valve adjustment is based solely on the real-time feedback of a single thermocouple in this temperature zone. Experimental group (modulation of this invention): The above-mentioned adaptive control system based on dynamic enthalpy integral, spatiotemporal decoupling feedforward and airflow coupling dynamic dead zone is enabled.

[0120] After 30 minutes of continuous full-load operation of the oven, a disturbance was artificially applied: the moisture content of the continuously entering materials was instantly increased by 15% (simulating abnormal mixing of materials in the upstream dough belt during actual production). The test results are as follows: Comparison of furnace temperature field stability: by Figure 3 As can be seen, in the traditional mode (shown by the dotted line), a sudden increase in moisture causes a sharp drop in the temperature of the current temperature zone, and the closed-loop PID controller violently opens the valve. Subsequently, the delayed excess heat is carried by the airflow into the downstream adjacent temperature zone, causing severe chain oscillations throughout the furnace. The temperature overshoot of the adjacent temperature zone is as high as ±8.0℃, and it takes nearly 15 minutes to recover to a steady state.

[0121] In the mode of this invention (shown by the solid line), due to the early intervention of the enthalpy integral feedforward, the downstream temperature zone is pre-supplemented with the latent heat required by the dough; at the same time, the dynamic dead zone mechanism shields the airflow crosstalk brought by the upstream. The maximum fluctuation of the entire furnace temperature field is strictly limited to within ±1.8℃, and it smoothly transitions to a new steady state within 3.5 minutes, significantly improving the system's ability to resist large hysteresis disturbances.

[0122] Comparison of mechanical action frequency of actuators: Extracting continuous 4 hours of operating data, under the traditional mode, the gas proportional valves in each temperature zone are affected by mutual competition and high-frequency noise in the space, with an average action frequency as high as 145 times / hour, which can easily cause physical wear of the valve body transmission gears.

[0123] This invention utilizes a spatial sliding window filtering and dynamic dead zone freezing mechanism to filter out invalid micro-disturbances, significantly reducing the average valve actuation frequency to 32 times / hour, a reduction of 77.9%, and extending the service life of the core actuators of the equipment.

[0124] Consistency of end-product quality: Sampled biscuits were tested and compared. Figure 4 It can be seen that the product moisture distribution under the traditional model (dashed envelope) is relatively discrete with a large variance, resulting in a defect rate (over-burning or under-burning) of 4.2%.

[0125] In the mode of this invention (solid line envelope thickened), based on energy tracking and compensation accurate to the microscopic level of virtual slices, the total enthalpy of all materials exiting the furnace is highly consistent. Its probability density curve exhibits a distinctly slender characteristic, with data highly concentrated around the target center value set by the process (e.g., 3.0%), and a very small standard deviation. Statistics show that the defect rate is significantly reduced to below 0.3%, and this invention achieves high-quality uniformity in large-scale continuous food production.

Claims

1. An adaptive control system for parameters of food production equipment, characterized in that, include: Absolute encoder, multiple physical temperature zones, and main control unit; The main control unit stores: The slice tracking module is used to divide continuous material into virtual material slices and update the absolute spatial coordinates and instantaneous linear velocity of the virtual material slices based on the absolute encoder data. The enthalpy integration module is used to obtain the ambient temperature and circulating fan operating frequency of the virtual material slice based on the absolute spatial coordinates, and to calculate the actual dynamic enthalpy accumulation of the virtual material slice. The batch aggregation module is used to calculate the spatial moving average deviation based on the error between the actual dynamic enthalpy accumulation and the target enthalpy threshold when the virtual material slice reaches the end of the current physical temperature zone, and generate the feedforward compensation opening increment for the downstream adjacent physical temperature zone. The compensation trigger module is used to establish an absolute spatial coordinate trigger point based on the time hysteresis parameter and the instantaneous linear velocity. When the virtual material slice reaches the absolute spatial coordinate trigger point, it sends the feedforward compensation opening increment to the downstream adjacent physical temperature zone. The dead zone control module is used to expand the static dead zone constant of the current physical temperature zone controller into a dynamic dead zone threshold based on the cross-coupling coefficient when issuing the feedforward compensation opening increment to freeze the output change.

2. The adaptive control system for parameters of food production equipment according to claim 1, characterized in that, The slice tracking module initializes a static parameter matrix, which includes a target enthalpy threshold array, a heat capacity response time bias matrix, a thermal dispersion field width matrix, and an airflow coupling matrix between adjacent temperature zones. The enthalpy integration module compares the absolute spatial coordinates of the virtual material slice with the physical installation positions of adjacent thermocouples, and calculates the equivalent ambient temperature by weighted average based on the relative position ratio. Extract the mapping function between the convective heat transfer coefficient and the fan operating frequency, calculate the convective heat transfer coefficient corresponding to the current absolute spatial coordinates, obtain the temperature difference value by subtracting the equivalent ambient temperature from the material base reference temperature constant, and multiply the convective heat transfer coefficient by the temperature difference value to obtain the instantaneous heat flux. The single-step heat absorption is calculated based on the instantaneous heat flux. If the single-step heat absorption is within the effective heat absorption range, the single-step heat absorption is added to the actual dynamic enthalpy accumulation.

3. The adaptive control system for parameters of food production equipment according to claim 2, characterized in that, The batch aggregation module divides the width of the thermal dispersion field corresponding to the current physical temperature zone by a fixed physical length constant and performs an upward rounding operation to calculate the capacity of the spatial sliding window. When the virtual material slice that has reached the end boundary of the current physical temperature zone is used as the center to trace forward and backward beyond the effective sequence range, the sliding window is truncated to the actual usable data boundary. Calculate the absolute value of the spatial moving average deviation. If the absolute value is less than the preset thermal enthalpy tolerance dead zone, then the feedforward compensation opening increment is set to zero. If the deviation is greater than or equal to the preset enthalpy tolerance dead zone, the deviation difference between the spatial moving average deviation and the preset enthalpy tolerance dead zone is calculated using the set feedforward proportional gain coefficient. Based on the positive or negative polarity of the spatial moving average deviation, the feedforward proportional gain coefficient is multiplied by the deviation difference to obtain the feedforward compensation opening increment.

4. The adaptive control system for parameters of food production equipment according to claim 2, characterized in that, The compensation triggering module extracts the time hysteresis parameter corresponding to the downstream physical temperature zone in the thermal capacity response time bias matrix, and multiplies the instantaneous linear velocity with the time hysteresis parameter to obtain the spatial advance triggering distance. The absolute spatial coordinate trigger point is determined by subtracting the spatial advance trigger distance from the starting physical coordinates of the downstream physical temperature zone. Determine whether the space advance trigger distance is greater than the actual mechanical construction length of the current physical temperature zone. If the determination result is yes, forcibly cut off the space advance trigger distance and clamp it to the preset limit ratio of the actual mechanical construction length of the current physical temperature zone.

5. The adaptive control system for parameters of food production equipment according to claim 2, characterized in that, The dead zone control module performs a spatial truncation procedure on the airflow coupling matrix between the adjacent temperature zones, forcibly overwriting the cross coupling coefficient of temperature zones that are not directly adjacent in physical space to zero, retaining the self-feedback coefficient on the main diagonal and the mutual coupling data of the temperature zones that are adjacent on the upper and lower sides, and simplifying the airflow coupling matrix between the adjacent temperature zones into a tri-diagonal matrix. Extract the total opening change of adjacent upstream and downstream temperature zones in the current control cycle, and calculate the temperature disturbance by combining it with the simplified airflow coupling matrix between adjacent temperature zones. The dynamic dead zone threshold is calculated by multiplying the temperature disturbance by the safety margin factor and adding it to the static dead zone constant. If the dynamic dead zone threshold is greater than the preset maximum tolerance limit parameter, then the dynamic dead zone threshold is forcibly truncated and clamped to the maximum tolerance limit parameter.

6. A method for adaptive control of parameters of food production equipment, applied to the adaptive control system for parameters of food production equipment as described in any one of claims 1-5, characterized in that, Includes the following steps: The continuous material is divided into virtual material slices, and the absolute spatial coordinates and instantaneous linear velocity of the virtual material slices are updated based on the absolute encoder data. The ambient temperature and circulating fan operating frequency of the virtual material slice are obtained based on the absolute spatial coordinates, and the actual dynamic enthalpy accumulation of the virtual material slice is calculated. When the virtual material slice reaches the end of the current physical temperature zone, the spatial moving average deviation is calculated based on the error between the actual dynamic enthalpy accumulation and the target enthalpy threshold, and a feedforward compensation opening increment for the downstream adjacent physical temperature zone is generated. The absolute spatial coordinate trigger point is established based on the time hysteresis parameter and the instantaneous linear velocity. When the virtual material slice reaches the absolute spatial coordinate trigger point, the feedforward compensation opening increment is sent to the downstream adjacent physical temperature zone. When issuing the feedforward compensation opening increment, the static dead zone constant of the current physical temperature zone controller is extended to a dynamic dead zone threshold based on the cross-coupling coefficient, and the output change is frozen.

7. The adaptive control method for parameters of food production equipment according to claim 6, characterized in that, In the step of dividing continuous material into virtual material slices, the main control unit instantiates a one-dimensional structure array containing a sequence number, furnace entry timestamp, and current position variable in sequence according to a fixed physical length constant. The product of the current instantaneous linear velocity and the scanning cycle is accumulated as the physical displacement increment into the absolute spatial coordinates of the previous cycle; When it is determined that the absolute spatial coordinates of a certain slice exceed the end boundary of the oven, the memory release mechanism is triggered to clear the structure data corresponding to the slice and reclaim the serial number.

8. The adaptive control method for parameters of food production equipment according to claim 6, characterized in that, In the step of establishing the absolute spatial coordinate trigger point, the main control unit performs a moving average filtering process on the acquired instantaneous linear velocity to obtain the effective running linear velocity, and determines whether the effective running linear velocity is lower than the preset minimum maintenance threshold. If the value falls below the preset minimum maintenance threshold, the current spatiotemporal decoupling computation task is suspended and the feedforward instruction delivery port is locked.

9. The adaptive control method for parameters of food production equipment according to claim 6, characterized in that, In the step of issuing the feedforward compensation opening increment, after the main control unit determines that the absolute spatial coordinates of the target virtual material slice have reached the absolute spatial coordinate trigger point, it simultaneously confirms that the pressure of the downstream temperature zone gas pipeline is within the preset normal operating range and that the combustion controller has not triggered the hardware interlock alarm, and activates the transmission port to issue the feedforward compensation opening increment to the gas proportional valve actuator in the downstream temperature zone.

10. The adaptive control method for parameters of food production equipment according to claim 6, characterized in that, To freeze the current physical temperature zone controller output change, the following conditions must be met simultaneously: The absolute value of the real-time deviation between the current actual ambient temperature and the set target temperature is within the dynamic dead zone threshold. The current temperature zone has not experienced a unidirectional monotonic temperature drift that lasts for a duration exceeding a preset anti-disturbance time threshold. The criteria for determining the unidirectional monotonic temperature drift are as follows: The temperature discrete difference sign remains consistent across multiple consecutive sampling periods, and the cumulative duration exceeds the preset anti-disturbance time threshold.

Images