Vanadium-nitrogen alloy semi-finished product sintering double push plate kiln multi-temperature zone temperature control system

Abstract

This invention discloses a multi-temperature zone temperature control system for a double-pusher plate kiln in the sintering of vanadium-nitrogen alloy semi-finished products, belonging to the field of industrial kiln temperature control technology. The system's functional carrier module integrates a stress wave sensor and a heat flow sensor in its variable specific heat carrier, simultaneously acquiring thermal stress acoustic emission and pyroelectric current signals of the material during cooling. After signal processing and digitization, the signal extraction module analyzes and fuses these signals to generate a cooling fingerprint feature vector containing four key features. The physical information neural network of the inversion module outputs a thermodynamic instability criterion based on this vector. The control module uses this criterion to determine and classify risks, generating instructions to dynamically adjust the carrier's heat absorption ratio and release micro-amplitude heat waves for precise temperature compensation. The early warning module provides linked display and alarm functions. This invention achieves essential perception and active control of the thermodynamic state during the cooling process, effectively suppressing nitrogen back diffusion and improving product consistency.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology for double-pusher kilns, and more specifically, to a multi-temperature zone temperature control system for a double-pusher kiln used for sintering vanadium-nitrogen alloy semi-finished products. Background Technology

[0002] Vanadium-nitrogen alloys, as a type of steel microalloying additive, derive their core function from the stable and uniform vanadium nitride phase within the alloy. The mainstream production process is the carbothermic reduction nitriding method, in which a mixed billet containing vanadium oxides and a carbonaceous reducing agent undergoes sequential heating, high-temperature isothermal sintering, and subsequent cooling in a nitrogen atmosphere within a double-pusher kiln.

[0003] During the high-temperature sintering stage, nitrogen atoms diffuse into the alloy surface to form a nitride layer. When the billet enters the cooling zone, if the cooling rate is inappropriate, especially if the cooling rate is too slow in the high-temperature section near the isothermal zone, it can trigger a severe nitrogen back-diffusion phenomenon. The mechanism is as follows: there is a chemical potential gradient driven by the temperature difference between the billet surface and the core. When the surface layer fails to quickly freeze the nitrogen atoms in the lattice due to slow cooling, the nitrogen atoms will gain enough kinetic energy to migrate back to the interior, where the temperature is higher and the nitrogen concentration is lower. This ultimately leads to a composition inversion layer in the alloy, where the nitrogen content increases from the surface to the interior. The composition inversion layer causes uneven nitrogen distribution within the alloy. After dissolving in the steel, coarse and uneven nitrides will prematurely precipitate in the localized supersaturated nitrogen concentration areas. These coarse particles not only lose their pinning strengthening effect but may also become the origin of microcracks. This results in batch-to-batch performance inconsistencies and is one of the key bottlenecks affecting the yield and quality stability of high-end vanadium-nitrogen alloys.

[0004] However, existing technologies and research generally focus on the kinetics of reduction and nitriding reactions in the high-temperature range, while lacking sufficient attention and effective control methods for the cooling process, especially the stability control of the high-temperature nitrided layer during the cooling phase. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides a multi-temperature zone temperature control system for a double-pusher kiln used for sintering vanadium-nitrogen alloy semi-finished products, comprising:

[0006] The functional carrier module includes a variable specific heat carrier and its internal thermal inert gas, stress wave sensor, and heat flow sensor.

[0007] The cooling zone is divided into N consecutive thermal inert zones along the material travel direction. Each thermal inert zone is independently laid with a variable specific heat carrier, and the interior of the variable specific heat carrier is filled with thermal inert gas.

[0008] The variable specific heat transfer fluid is made of porous ceramic material with high specific heat capacity and good thermal stability, and has a loose and porous plate-like structure.

[0009] In the preparation of the variable specific heat transfer fluid, aluminum nitride piezoelectric ceramic fibers are embedded in a three-dimensional mesh structure as an embedded piezoelectric fiber network, and stress wave sensors are also embedded therein. The stress wave sensors are used to sensitively capture the high-frequency thermal stress acoustic emission signals generated by the release of thermal stress due to the material above the variable specific heat transfer fluid bearing surface during cooling and contraction. The embedded piezoelectric fiber network transmits the thermal stress acoustic emission signals to the next module.

[0010] A heat flow sensor is integrated on the upper surface of the variable specific heat carrier in contact with the material. The heat flow sensor uses a ceramic coating with a stable pyroelectric effect uniformly prepared on the carrier surface as the core sensing element, and includes an internal electrode connected to it.

[0011] The dynamic interfacial heat flow generated by the temperature difference between the material and the variable specific heat carrier acts on the ceramic coating. Based on the pyroelectric effect, the ceramic coating converts the dynamic interfacial heat flow into a weak pyroelectric current signal. The pyroelectric current signal is then transmitted to the next module via the built-in electrodes of the heat flow sensor.

[0012] The variable specific heat carrier, while carrying materials, provides a channel for uniform flow and heat exchange of thermal inert gas. The interior and bottom of the variable specific heat carrier are connected to a closed thermal inert gas circulation and thermal management unit.

[0013] The thermal inert gas circulation and thermal management unit is the actuator that realizes the variable specific heat function, and it uses an inert gas with stable chemical properties and high heat capacity.

[0014] The gas drive component of the thermal inert gas circulation and thermal management unit includes a high-temperature resistant gas pump, a precision flow control valve, and a gas flow reversing valve. These components are used to drive and precisely control the flow rate and direction of the gas within the system.

[0015] The variable specific heat carrier is equipped with a high-temperature heat storage chamber and a cooling chamber. The high-temperature heat storage chamber heats the flowing gas and maintains it at a set high temperature by means of electric auxiliary heating or recovery of waste heat from the kiln. The cooling chamber cools the flowing gas and maintains it at a set low temperature by means of an external cooling system.

[0016] High-temperature resistant metal bellows are used to connect all components to the gas interface at the bottom of the variable specific heat carrier, forming a completely sealed circulation loop to ensure no gas leakage.

[0017] The overall heat absorption ratio of the variable ratio heat carrier is achieved by adjusting the state of the gas flowing through it. Specifically, there are two basic working modes, which can be switched steplessly.

[0018] One operating mode is a high heat absorption ratio mode, used for cooling enhancement. A gas pump drives cryogenic gas to flow unidirectionally through the microchannels inside the carrier at a high flow rate.

[0019] Forced convection heat exchange occurs between the low-temperature gas and the high-temperature carrier skeleton, as well as the material heat transferred through thermal conduction. The gas is heated, thereby efficiently carrying away the heat from the material. At this point, the carrier exhibits a high heat absorption ratio, i.e., a strong cooling capacity.

[0020] Another operating mode is the low heat absorption ratio mode, used to generate micro-wave heat. An airflow reversing valve switches the flow path. An air pump drives the gas in the high-temperature heat storage chamber to be pulsedly injected into the carrier microchannels at a specific flow rate and duration.

[0021] The high-temperature gas heats the carrier frame, transferring heat directionally and rapidly to the material above through heat conduction and radiation. At this point, the carrier's heat absorption ratio decreases significantly or even becomes negative, i.e., it releases heat, forming a controllable micro-wave of heat to achieve precise thermal compensation to the bottom of the material.

[0022] By continuously adjusting the gas flow rate, temperature, and the switching ratio between the two modes, the overall heat absorption ratio of the variable specific heat carrier can be continuously and linearly dynamically adjusted between maximum heat absorption and maximum heat release.

[0023] The signal processing module synchronously acquires, amplifies, filters, and digitizes the thermal stress acoustic emission signals and pyroelectric current signals generated in each thermal inert zone.

[0024] The thermal stress acoustic emission signal channel is connected to an embedded piezoelectric fiber network to receive thermal stress acoustic emission signals. The cutoff frequency of the high-pass filter is set to 100kHz to effectively filter out kiln mechanical vibration and power frequency interference. The amplifier gain is adjustable, and the sampling rate is not less than 1MHz to preserve the rich frequency domain characteristics of the thermal stress acoustic emission signal.

[0025] The pyroelectric current signal channel connects to the built-in electrodes of the heat flux sensor to receive the pyroelectric current signal. It is equipped with a low-noise current amplifier and a high-precision analog-to-digital converter, with a sampling rate of at least 1 kHz to accurately capture current waveform details.

[0026] The synchronization controller ensures that the thermal stress acoustic emission signal and the pyroelectric current signal acquisition timestamp are strictly synchronized in the same zone and at the same time, providing a foundation for subsequent feature fusion.

[0027] The synchronization controller achieves strict synchronization of the sampling start point by sending a unified hardware trigger clock signal to the analog-to-digital converters of the two signal channels; and adds a microsecond-level timestamp to all data frames with the same time base.

[0028] All signal transmissions in the signal processing module use shielded twisted-pair cables or optical fibers to ensure long-term operational stability in harsh industrial environments.

[0029] The synchronized and digitized dual signal streams are transmitted in real time to the feature extraction module as the original data source for calculating the cooling fingerprint feature vector.

[0030] The feature extraction module performs real-time analysis on the digitized signal flow and extracts feature vectors that are strongly correlated with cooling dynamics, known as cooling fingerprints.

[0031] For thermal stress acoustic emission signals, the signal power spectrum is calculated by fast Fourier transform at each time window interval, and the centroid eigenvalue of the dominant frequency distribution is calculated. The Shannon entropy of the thermal stress acoustic emission signal amplitude within the time window is calculated and used as the energy disorder eigenvalue to quantify the randomness and statistical distribution characteristics of stress release events.

[0032] For pyroelectric current waveform signals, within the same time window, the decay segment curve of the pyroelectric current is fitted using an exponential decay function model, and the time constant of its dominant decay process is extracted as a characteristic value of thermal flux relaxation characteristics, reflecting the rapidity of interfacial heat flux establishment.

[0033] The intensity of the decay component under a specific time constant is obtained by fitting the current decay curve, which is the characteristic value of the thermal flux relaxation characteristic, reflecting the relaxation characteristics of the thermal flux.

[0034] The above four feature values ​​are normalized and combined in a fixed order to generate the standardized cooling fingerprint feature vector FP_k(t) of the k-th partition of the cooling zone at time t, which is: [major frequency distribution centroid feature value, energy disorder feature value, heat flow establishment abruptness feature value, heat flow relaxation characteristic feature value];

[0035] This feature vector, as the output of this module, is sent in real time to the inversion module for further processing.

[0036] The time window is a fixed sliding window of two seconds, the length of which is determined by a combination of the typical heat transfer time of the material in the cooling zone and the stability of the signal.

[0037] The inversion module has a built-in lightweight physical information neural network model that has been pre-trained. The lightweight physical information neural network model takes the real-time generated cooling fingerprint feature vector FP_k(t) as input and outputs a thermodynamic instability criterion, which is a scalar value.

[0038] The pre-training steps for the lightweight physical information neural network model are as follows:

[0039] During kiln commissioning or specific experimental phases, the system operates under various known and precisely controlled cooling rate conditions, covering a full range of conditions from excessively fast and ideal to excessively slow, and collects cooling fingerprint data of each thermal inertia zone under various conditions as training samples.

[0040] For each sample, set its training target value. Defined as the ratio of the actual cooling rate to the safe cooling rate required by the process. Under ideal conditions, this ratio is 1; when cooling is too slow, the ratio is less than 1; when cooling is too fast, the ratio is greater than 1.

[0041] The required safe cooling rate is the critical minimum cooling rate determined through process testing that ensures no back diffusion of nitrogen occurs in the vanadium-nitrogen alloy nitriding layer.

[0042] In production, the trained physical information neural network model receives the real-time cooling fingerprint feature vector FP_k(t) and calculates the thermodynamic instability criterion. .

[0043] This indicates that the cooling kinetics of the thermally inert zone are ideal;

[0044] This indicates that the cooling kinetics are too slow; the smaller the value, the higher the risk.

[0045] The control module is used to implement hierarchical early warning and adaptive feedback control based on the thermodynamic instability criterion TDIC_k(t), and it performs the following steps:

[0046] The control module stores and manages warning thresholds for the risk of excessively slow temperature drop. The initial value of this threshold is derived from a large amount of historical safety process data. Analysis of the statistical lower limit.

[0047] Continuous comparison of real-time data across partitions Compared with the above warning threshold Based on the comparison results and trends, the partition status is determined in real time:

[0048] When 1 When the k-th partition is in a safe state, it is determined that the k-th partition is in a safe state.

[0049] when At that time, the k-th partition is determined to be in a state of risk due to excessively slow temperature drop.

[0050] Once a risk state is determined, structured, tiered control instructions are generated based on the risk level.

[0051] Based on real-time deviation Based on the duration of deviation, the risk level is divided into mild risk, moderate risk, and severe risk.

[0052] The risk level classification is based on:

[0053] Mild risk: And duration ;

[0054] Moderate risk: Or mild risk persists ;

[0055] Severe risk: or moderate risk persists .

[0056] in, , , , This is a threshold preset based on process safety margin.

[0057] The generated structured instruction data includes the target partition number k, the current... Value, risk level, and a core regulatory intensity coefficient. .

[0058] Regulation intensity coefficient It is a normalized value between 0 and 1, determined by a preset value. =f(Δ) is calculated from the nonlinear mapping relationship; the larger Δ is, the better. The larger the value, the stronger the regulatory effort required.

[0059] The generated control commands are sent to the upper-level core controller in real time. The controller then processes the received commands. and regulation intensity coefficient By using a preset mapping relationship, the required heat absorption ratio adjustment for the variable specific heat carrier in the k-th partition and its adjacent partitions is calculated. Subsequently, the control system instructs the lower carrier to release stored heat, generating a micro-thermal wave, thereby actively and precisely balancing the material temperature difference and ultimately achieving the control objective of suppressing nitrogen back diffusion.

[0060] By adjusting the speed of the high-temperature pump driving the circulation of thermal inert gas and the opening of the reversing valve, the flow rate and duration of the thermal inert gas flowing into the carrier of this zone can be controlled, thereby achieving controllable heat release.

[0061] The early warning module receives and responds to the risk status assessment results issued by the control module, providing operators with tiered status displays and early warnings. The early warning levels are strictly correlated with the aforementioned risk levels, as detailed below:

[0062] Level 3 warning, when First time below In the monitoring interface, the partition identifier turns yellow and flashes, and a log is recorded, indicating that the cooling rate of the kth partition has begun to decrease, and it is recommended to pay attention to the trend.

[0063] Level 2 warning: When the risk status persists for more than 30 seconds, or the real-time deviation is detected... When the preset lower limit of the medium risk threshold is reached, an audible and visual alarm is triggered. The zone identifier turns orange and stays lit, and a real-time data window for that zone automatically pops up, indicating that the cooling of zone k is too slow and micro-thermal wave compensation has been automatically activated. Please confirm the process parameters.

[0064] Level 1 Warning: When the system determines that it has entered a severe risk level, the highest level audible and visual alarm is triggered, and the zone indicator turns red. It is recommended to pay attention immediately or prepare for maintenance. The warning indicates that the cooling of zone k is severely insufficient and the risk of nitrogen back diffusion is high! Maximum compensation has been executed. Please check the equipment immediately or prepare for maintenance.

[0065] Level 2 and above alerts require operator confirmation via the interface; the confirmation action will be recorded. When this zone... The value continued to recover to Once the above conditions are met and the situation remains stable for a set period of time, the warning status will automatically reset, and the interface indicators will return to normal.

[0066] The beneficial effects of the multi-temperature zone temperature control system for a double-pusher kiln in the sintering of vanadium-nitrogen alloy semi-finished products of this invention are as follows:

[0067] Existing technologies and research generally focus on the kinetics of reduction and nitriding reactions at high temperatures, but lack effective means to control the stability of the high-temperature nitrided layer during cooling. This makes it difficult to suppress nitrogen back-diffusion caused by slow cooling, ultimately leading to inherent defects such as inverted composition layers and uneven performance in the product. This invention innovatively abandons traditional indirect temperature measurement methods. It directly captures the acoustic emission of thermal stress and the pyroelectric signals at the interface from the material using a variable specific heat carrier, fusing them to form a cooling fingerprint. Furthermore, it utilizes a physical information neural network to directly deduce the thermodynamic instability criteria, thus achieving a direct, essential, and sensitive judgment of the cooling rate state. Secondly, it pioneers an active and precise control mechanism based on multi-source information fusion. When the system determines that a certain zone is cooling too slowly, it can accurately calculate the control intensity, inject micro-amplitude thermal waves into the carrier to implement local thermal compensation, and coordinately adjust the heat absorption ratio of adjacent zones to form an intelligent closed-loop control. This invention significantly improves the active intervention capability in the sintering cooling process, especially the stability of the high-temperature nitrided layer and nitrogen back-diffusion, fundamentally ensuring the homogeneity and batch stability of high-end vanadium-nitrogen alloy products. Attached Figure Description

[0068] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0069] Figure 1 This is a flowchart of the method for the vanadium-nitrogen alloy semi-finished product sintering double-pusher kiln multi-temperature zone temperature control system of the present invention;

[0070] Figure 2 This is a schematic diagram of the multi-temperature zone temperature control system of the double-pusher plate kiln for sintering vanadium-nitrogen alloy semi-finished products according to the present invention.

[0071] Figure 3 This is a flowchart illustrating the dynamic control of heat absorption ratio in the multi-temperature zone temperature control system of the double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products according to the present invention.

[0072] Figure 4 This is a flowchart illustrating the calculation of the thermodynamic instability criterion in the multi-temperature zone temperature control system of the double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products according to the present invention. Detailed Implementation

[0073] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0074] This invention provides a multi-temperature zone temperature control system for a double-pusher kiln used for sintering vanadium-nitrogen alloy semi-finished products, comprising:

[0075] The functional carrier module includes a variable specific heat carrier and its internal thermal inert gas, stress wave sensor, and heat flow sensor.

[0076] The cooling zone is divided into N consecutive thermal inert zones along the material travel direction. Each thermal inert zone is independently laid with a variable specific heat carrier, and the interior of the variable specific heat carrier is filled with thermal inert gas.

[0077] It should be noted that:

[0078] Thermal inertia zones refer to several independent cooling control units artificially divided along the material flow direction within the cooling zone of a double-pusher kiln. Each thermal inertia zone is structurally isolated and thermally exhibits low coupling with external thermal disturbances. Internally, it primarily exchanges heat with the material through a variable specific heat carrier, without directly participating in the overall kiln heating or combustion process. By dividing the cooling zone into multiple thermal inertia zones, the cooling behavior of each zone can be made relatively independent and finely controlled, thus providing a basis for zone-level cooling status perception and local thermal compensation.

[0079] The variable specific heat transfer fluid is made of porous ceramic material with high specific heat capacity and good thermal stability, and has a loose and porous plate-like structure.

[0080] In the preparation of the variable specific heat carrier, aluminum nitride piezoelectric ceramic fibers are pre-embedded in a three-dimensional mesh structure as an embedded piezoelectric fiber network, and stress wave sensors are embedded therein.

[0081] The stress wave sensor is used to sensitively capture the high-frequency thermal stress acoustic emission signal generated by the release of thermal stress when the material above the variable ratio heat carrier surface cools and contracts. The embedded piezoelectric fiber network transmits the thermal stress acoustic emission signal to the next module.

[0082] A heat flow sensor is integrated on the upper surface of the variable specific heat carrier in contact with the material. The heat flow sensor uses a ceramic coating with a stable pyroelectric effect uniformly prepared on the carrier surface as the core sensing element, and includes an internal electrode connected to it.

[0083] The dynamic interfacial heat flow generated by the temperature difference between the material and the variable specific heat carrier acts on the ceramic coating. Based on the pyroelectric effect, the ceramic coating converts the dynamic interfacial heat flow into a weak pyroelectric current signal. The pyroelectric current signal is then transmitted to the next module via the built-in electrodes of the heat flow sensor.

[0084] The variable specific heat carrier, while carrying materials, provides a channel for uniform flow and heat exchange of thermal inert gas. The interior and bottom of the variable specific heat carrier are connected to a closed thermal inert gas circulation and thermal management unit.

[0085] The thermal inert gas circulation and thermal management unit is the actuator that realizes the variable specific heat function, and it uses an inert gas with stable chemical properties and high heat capacity.

[0086] The gas drive component of the thermal inert gas circulation and thermal management unit includes a high-temperature resistant gas pump, a precision flow control valve, and a gas flow reversing valve. These components are used to drive and precisely control the flow rate and direction of the gas within the system.

[0087] The variable specific heat carrier is equipped with a high-temperature heat storage chamber and a cooling chamber. The high-temperature heat storage chamber heats the flowing gas and maintains it at a set high temperature by means of electric auxiliary heating or recovery of waste heat from the kiln. The cooling chamber cools the flowing gas and maintains it at a set low temperature by means of an external cooling system.

[0088] High-temperature resistant metal bellows are used to connect all components to the gas interface at the bottom of the variable specific heat carrier, forming a completely sealed circulation loop to ensure no gas leakage.

[0089] The overall heat absorption ratio of a variable specific heat transfer fluid is achieved by adjusting the state of the gas flowing through it. Specifically, there are two basic operating modes, which can be seamlessly switched:

[0090] One operating mode is a high heat absorption ratio mode, used for cooling enhancement. A gas pump drives cryogenic gas to flow unidirectionally through the microchannels inside the carrier at a high flow rate.

[0091] Forced convection heat exchange occurs between the low-temperature gas and the high-temperature carrier skeleton, as well as the material heat transferred through thermal conduction. The gas is heated, thereby efficiently carrying away the heat from the material. At this point, the carrier exhibits a high heat absorption ratio, i.e., a strong cooling capacity.

[0092] Another operating mode is the low heat absorption ratio mode, used to generate micro-wave heat. An airflow reversing valve switches the flow path. An air pump drives the gas in the high-temperature heat storage chamber to be pulsedly injected into the carrier microchannels at a specific flow rate and duration.

[0093] The high-temperature gas heats the carrier frame, transferring heat directionally and rapidly to the material above through heat conduction and radiation. At this point, the carrier's heat absorption ratio decreases significantly or even becomes negative, i.e., it releases heat, forming a controllable micro-wave of heat to achieve precise thermal compensation to the bottom of the material.

[0094] By continuously adjusting the gas flow rate, temperature, and the switching ratio between the two modes, the overall heat absorption ratio of the variable specific heat carrier can be continuously and linearly dynamically adjusted between maximum heat absorption and maximum heat release.

[0095] When the control module determines that there is a risk of excessively slow temperature drop in the k-th partition and calculates the control intensity coefficient... back:

[0096] The control module sends instructions to the controller of the gas circulation control unit in that zone, including the target mode, gas temperature, flow rate setting, and duration.

[0097] The controller adjusts the reversing valve to connect the high-temperature heat storage chamber and drives the gas pump to pump the high-temperature gas into the variable specific heat carrier in the kth partition at a flow rate of Q_k for T_k seconds.

[0098] High-temperature gas flows through the carrier, generating micro-thermal waves that locally heat the material. Simultaneously, the system can adjust the gas flow rate in adjacent zones to collaboratively optimize the thermal field.

[0099] System monitoring follow-up The upward trend can be used to determine whether to terminate or adjust this heat release action.

[0100] In this embodiment, the gas flow and heat exchange response are fast, meeting the requirements for real-time control. Precise control of gas velocity and energy, along with the control intensity coefficient, can be achieved through the flow valve. It can establish accurate mapping relationships. Simultaneously, the gas, as a medium, makes the internal temperature field of the carrier more uniform, which is beneficial for uniform heating of the material. The inert gas loop is closed, does not interfere with the kiln process atmosphere, and is safe and reliable. The vibration frequency generated by the gas flow is usually much lower than the acoustic emission signal frequency band and can be effectively filtered out by a filter, without affecting the signal acquisition of the piezoelectric fiber network.

[0101] The signal processing module synchronously acquires, amplifies, filters, and digitizes the thermal stress acoustic emission signals and pyroelectric current signals generated in each thermal inert zone.

[0102] The thermal stress acoustic emission signal channel is connected to an embedded piezoelectric fiber network to receive thermal stress acoustic emission signals. The cutoff frequency of the high-pass filter is set to 100kHz to effectively filter out kiln mechanical vibration and power frequency interference. The amplifier gain is adjustable, and the sampling rate is not less than 1MHz to preserve the rich frequency domain characteristics of the thermal stress acoustic emission signal.

[0103] The pyroelectric current signal channel connects to the built-in electrodes of the heat flux sensor to receive the pyroelectric current signal. It is equipped with a low-noise current amplifier and a high-precision analog-to-digital converter, with a sampling rate of at least 1 kHz to accurately capture current waveform details.

[0104] The synchronization controller ensures that the thermal stress acoustic emission signal and the pyroelectric current signal acquisition timestamp are strictly synchronized in the same zone and at the same time, providing a foundation for subsequent feature fusion.

[0105] The synchronization controller achieves strict synchronization of the sampling start point by sending a unified hardware trigger clock signal to the analog-to-digital converters of the two signal channels; and adds a microsecond-level timestamp to all data frames with the same time base.

[0106] All signal transmissions in the signal processing module use shielded twisted-pair cables or optical fibers to ensure long-term operational stability in harsh industrial environments.

[0107] The synchronized and digitized dual signal streams are transmitted in real time to the feature extraction module as the original data source for calculating the cooling fingerprint feature vector.

[0108] The feature extraction module performs real-time analysis on the digitized signal flow and extracts feature vectors that are strongly correlated with cooling dynamics, known as cooling fingerprints.

[0109] For thermal stress acoustic emission signals, the centroid eigenvalue of the dominant frequency distribution is calculated at each time window interval; the Shannon entropy of the thermal stress acoustic emission signal amplitude within that time window is calculated and used as the energy disorder eigenvalue to quantify the randomness and statistical distribution characteristics of stress release events.

[0110] For pyroelectric current waveform signals, within the same time window, the decay segment curve of the pyroelectric current is fitted using an exponential decay function model, and the time constant of its dominant decay process is extracted as a characteristic value of thermal flux relaxation characteristics, reflecting the rapidity of interfacial heat flux establishment.

[0111] The dominant frequency distribution centroid eigenvalue, energy disorder eigenvalue, thermal flux relaxation characteristic eigenvalue, and thermal flux build-up abruptness eigenvalue characterize the cooling dynamics from different dimensions: the dominant frequency distribution centroid reflects the main frequency band of thermal stress release, which drifts when cooling is uneven; energy disorder characterizes the dispersion of stress release events, with higher entropy values ​​indicating greater cooling instability; thermal flux build-up abruptness directly reflects the strength of the interfacial temperature difference drive and is an instantaneous manifestation of the cooling rate; and thermal flux relaxation characteristic eigenvalue reflects thermal inertia and heat transfer resistance. Their fusion (FP) can comprehensively and essentially describe the thermo-mechanical coupling process of material cooling, providing an informational basis for accurately inverting instability criteria.

[0112] The above four feature values ​​are normalized and combined in a fixed order to generate the standardized cooling fingerprint feature vector FP_k(t) of the k-th partition of the cooling zone at time t, which is: [major frequency distribution centroid feature value, energy disorder feature value, heat flow establishment swiftness feature value, heat flow relaxation characteristic feature value]; this feature vector is the output of this module and is sent to the inversion module in real time for further processing.

[0113] The calculation steps for the energy disorder eigenvalue are as follows:

[0114] Within a time window of length T, a discrete sequence of digitized and filtered thermal stress acoustic emission signals is received from the signal processing module: , where M is the total number of sampling points within the window. The instantaneous amplitude of the thermal stress acoustic emission signal, which has been digitized and filtered, is obtained from the first sampling moment within a specific time window of length T. This represents the instantaneous amplitude of the thermal stress acoustic emission signal, which has been digitized and filtered, acquired from the Mth sampling moment within the same time window.

[0115] Calculate the absolute amplitude at each sampling point in the signal sequence to obtain a non-negative amplitude sequence:

[0116] Find the minimum value in sequence A and maximum value Amplitude range Divide the signal into N equal intervals, where N is the number of histogram bins, which can be preset according to the signal characteristics, for example, N=256.

[0117] The number of data points falling within each interval in the amplitude sequence A is counted to obtain the total count. .

[0118] Dividing the count of each interval by the total number of sampling points M yields an estimate of the probability of that amplitude interval occurring:

[0119] Based on the definition of Shannon entropy in information theory, and using the obtained probability estimate... For all that satisfy Summing the interval i, we obtain the energy disorder eigenvalue of the signal within that time window. :

[0120] ;

[0121] The steps for calculating the eigenvalues ​​of thermal flux relaxation characteristics are as follows:

[0122] Within a time window synchronized with acoustic emission analysis, the received signal processing module has digitized the pyroelectric current signal sequence. ,in, This represents the Lth pyroelectric current signal value.

[0123] In sequence In this process, locate a complete single-peak event. Find the peak point of the event and the end point where the current falls back to near the baseline noise level after the peak.

[0124] Extract the subsequence from the peak point m to the end point n as the decay segment data to be fitted: ,in, The subsequence ending at point n corresponds to the time series as follows: ,in The sampling interval is denoted as .

[0125] A single exponential decay model was used to fit the decay segment data: .in, It is the fitted current value at time t. It is the baseline current offset that decays and tends to stabilize. τ is the amplitude coefficient of the decay process, τ is the thermal relaxation time constant to be determined, i.e., this eigenvalue, and t is the time counted from the decay start point, i.e., the peak point. This represents an exponential function with the natural constant as its base.

[0126] Using nonlinear least squares method to evaluate the parameters Perform the optimal estimate.

[0127] by The initial value is set as the average of several points at the end of the decay segment. The initial value is set to The initial value of τ is set as an empirical value based on historical data.

[0128] Adjust parameters iteratively , to fit the curve Compared with measured data Minimize the sum of squared residuals between them.

[0129] When the change in parameters or the change in residuals is less than the preset tolerance, the fitting is complete, and the optimal parameter estimate is obtained, which includes the target value τ.

[0130] The time constant τ obtained from the fitting is output as the characteristic value of the thermal relaxation properties of that time window.

[0131] The spectrum is obtained by performing a fast Fourier transform on the thermal stress acoustic emission signal sequence within the time window. The weighted average of the squares of the amplitudes of each frequency component is calculated, with the weights being the corresponding frequency values. The resulting weighted average is the centroid characteristic value of the dominant frequency distribution of the signal energy.

[0132] Identify the rising edge phase before the peak in the pyroelectric current signal waveform, calculate the first derivative sequence of the current with respect to time during this phase, and take the maximum absolute value of the derivative sequence, i.e., the maximum instantaneous slope. This value is used as the characteristic value of the heat flow establishment abruptness.

[0133] The time window is a fixed sliding window of two seconds, the length of which is determined by a combination of the typical heat transfer time of the material in the cooling zone and the stability of the signal.

[0134] Using the above method, the feature extraction module simultaneously extracts four-dimensional features characterizing event distribution, process randomness, transient intensity, and dynamic decay from two complementary physical dimensions: microscopic mechanical acoustic emission and macroscopic heat transfer and pyroelectricity. This multi-dimensional and integrated feature design can comprehensively and essentially characterize the thermo-mechanical coupling dynamic state of materials during the cooling process, laying the foundation for subsequent accurate diagnosis.

[0135] The inversion module has a built-in lightweight physical information neural network model that has been pre-trained. The lightweight physical information neural network model takes the real-time generated cooling fingerprint feature vector FP_k(t) as input and outputs a thermodynamic instability criterion, which is a scalar value.

[0136] The pre-training steps for the lightweight physical information neural network model are as follows:

[0137] During kiln commissioning or specific experimental phases, the system operates under various known and precisely controlled cooling rate conditions, covering a full range of conditions from excessively fast and ideal to excessively slow, and collects cooling fingerprint data of each thermal inertia zone under various conditions as training samples.

[0138] For each sample, set its training target value. Defined as the ratio of the actual cooling rate to the safe cooling rate required by the process. Under ideal conditions, this ratio is 1; when cooling is too slow, the ratio is less than 1; when cooling is too fast, the ratio is greater than 1.

[0139] The required safe cooling rate is the critical minimum cooling rate determined through process testing that ensures no back diffusion of nitrogen occurs in the vanadium-nitrogen alloy nitriding layer.

[0140] The input to the lightweight physical information neural network model is a four-dimensional, normalized cooling fingerprint feature vector FP_k(t). The output of the model, namely the thermodynamic instability criterion, is a continuous scalar value with a theoretical range of approximately (0,+∞), which typically falls within the neighborhood of [0.5,1.5] under ideal process control.

[0141] During training, a physical constraint loss is incorporated. The physical constraint loss function includes a monotonic constraint loss, which is constructed as follows: the partial derivative of the output thermodynamic instability criterion of the lightweight physical information neural network model with respect to the features in the input feature vector FP_k(t) that are positively correlated with the cooling rate is negative. This forces the model to learn the monotonically increasing relationship between the thermodynamic instability criterion and such features, ensuring that its prediction conforms to the basic physical law that the weaker the cooling driving force, the higher the risk of instability.

[0142] The lightweight physical information neural network model adopts a fully connected structure with fewer layers and optimized number of neurons. After training, it is pruned or quantized to meet the real-time requirements of the production line while reducing the consumption of computing resources and making it easy to deploy on industrial controllers.

[0143] In production, the trained lightweight physical information neural network model receives the real-time cooling fingerprint feature vector FP_k(t) and calculates the thermodynamic instability criterion. .

[0144] Thermodynamic instability criterion The calculation steps are as follows:

[0145] The trained lightweight physical information neural network model takes the real-time generated cooling fingerprint feature vector FP_k(t) as input. After entering the lightweight physical information neural network model, this vector goes through multiple hidden layers in sequence: each neuron in each layer receives the output of all neurons in the previous layer, multiplies it by the fixed weights obtained from training, adds a fixed bias, and then performs a nonlinear transformation through an activation function, passing it forward layer by layer.

[0146] Finally, in the output layer, a single neuron maps the result of the previous layer to a continuous scalar value, which is the thermodynamic instability criterion for the current partition at time t. .

[0147] This indicates that the cooling kinetics of the thermally inert zone are ideal;

[0148] This indicates that the cooling kinetics are too slow; the smaller the value, the higher the risk.

[0149] The control module is used to determine the thermodynamic instability criterion. To achieve hierarchical early warning and adaptive feedback control, the following steps are performed:

[0150] The control module stores and manages warning thresholds for the risk of excessively slow temperature drop. The initial value of this threshold is derived from a large amount of historical safety process data. Analysis of the statistical lower limit.

[0151] Continuous comparison of real-time data across partitions Compared with the above warning threshold Based on the comparison results and trends, the partition status is determined in real time:

[0152] When 1 When the k-th partition is in a safe state, it is determined that the k-th partition is in a safe state.

[0153] when At that time, the k-th partition is determined to be in a state of risk due to excessively slow temperature drop.

[0154] Once a risk state is determined, structured, tiered control instructions are generated based on the risk level.

[0155] Based on real-time deviation Based on the duration of deviation, the risk level is divided into mild risk, moderate risk, and severe risk.

[0156] The risk level classification is based on:

[0157] Mild risk: And duration ;

[0158] Moderate risk: Or mild risk persists ;

[0159] Severe risk: or moderate risk persists .

[0160] in, , , , This is a threshold preset based on process safety margin.

[0161] The generated structured instruction data includes the target partition number k, the current... Value, risk level, and a core regulatory intensity coefficient. .

[0162] Regulation intensity coefficient The calculation is a dynamic process based on the real-time deviation degree Δ and considering the duration of the deviation. The following calculation model, incorporating nonlinear transformations and time accumulation effects, is used, and the calculation formula is as follows:

[0163] ;

[0164] in, It is the degree of real-time deviation. , Calculation is triggered at time 0. This is the saturation deviation, a constant preset based on process and equipment capabilities, representing the typical maximum risk deviation that can be addressed. When At this point, the basic computation term reaches 1. γ is the nonlinear shaping exponent, a constant greater than 1, typically γ∈[1.5,3]. It controls... Follow The shape of the growth curve. It is the duration of the deviation from the current risk status, that is, from First time below Start timing. η is the saturation time, a preset constant representing the typical maximum time threshold for allowing risk to persist. η is the time-cumulative gain coefficient, used to adjust the additional control intensity due to continued risk. min(1.0,...) is the limiting function, ensuring the final... Not exceeding 1.0, that is, not exceeding the system's maximum control capacity.

[0165] The control module calculates the control intensity coefficient according to the following steps within each control cycle. :

[0166] Get current Dynamic threshold and the duration of deviation from the current risk status. .

[0167] Calculate the degree of real-time deviation If Δ≤0, then set If the value is 0, the calculation ends; otherwise, it continues.

[0168] Calculate the basic control item When Δ is very small, this term is extremely small because γ > 1, resulting in very mild initial control. As Δ increases, this term increases rapidly, achieving a strong response to large deviations.

[0169] calculate This parameter increases linearly with the duration of the risk, and is used to address situations where Δ has not worsened but the risk has not been eliminated in the long term, reflecting the concept of integral control.

[0170] Add the two items together, and then use the min(1.0,...) function to restrict the result to the interval [0,1.0] to obtain the final result. .

[0171] Regulation intensity coefficient The parameters in the calculation formula need to be predetermined and fixed in the system in the following ways:

[0172] Determination: In the process safety limit test, an excessively slow cooling rate is artificially created until the product reaches the critical point for quality defects. Record the value corresponding to this critical state. ,but The saturation deviation is related to the maximum acceptable risk and the maximum controllability.

[0173] Determining γ: Through system simulation or backtesting of historical data, a trade-off is struck between controlling stability, avoiding overshoot oscillations, and response speed. Typically, debugging begins with γ=2, observing the system's adjustment process under different Δ values, and selecting the optimal value.

[0174] Determination of η: Set based on production cycle time and experience. η is a fine-tuning parameter used to set the maximum additional increment of control intensity under sustained risk.

[0175] The control module generates control commands and sends them to the controller in real time. The controller then processes the received commands. and regulation intensity coefficient By using a preset mapping relationship, the required heat absorption ratio adjustment for the variable specific heat carrier in the k-th partition and its adjacent partitions is calculated.

[0176] The controller receives structured control commands from the control module, which include: the target partition number k and the real-time thermodynamic instability criterion. and the calculated control intensity coefficient .

[0177] The controller has a preset control intensity coefficient. Adjustment amount of heat absorption ratio to core zone The mapping function is typically a linear or piecewise linear function. This function maps values ​​between 0 and 1. The value is mapped to the heat absorption ratio from the maximum heat absorption state. to maximum heat release state A specific adjustment amount between .

[0178] The calculation formula can be:

[0179] ;

[0180] in, =0 corresponds to maintaining the maximum heat absorption state. =1 corresponds to switching to the maximum heat dissipation state. When the value is in the middle, the corresponding adjustment amount is proportional.

[0181] To prevent the control measures from causing new temperature unevenness in space, the system coordinates the adjustment of adjacent zones based on a heat conduction model. The adjustment amount for adjacent zones... It is usually set as a decay ratio of the current zone adjustment amount, or dynamically calculated according to a preset diffusion function, with the aim of smoothing the overall thermal field.

[0182] The controller will adjust the calculated heat absorption ratio ( , This is converted into physical control parameters that can be directly executed by the thermal management unit.

[0183] The gas temperature is determined by switching the gas flow source. The gas switched to the high-temperature heat storage chamber is used to reduce the heat absorption ratio, while the gas switched to the cooling chamber is used to increase the heat absorption ratio.

[0184] The gas volumetric flow rate is controlled by adjusting the speed of the high-temperature gas pump and the opening of the precision flow valve. The higher the flow rate, the greater the intensity of forced convection heat transfer, and the greater the power of heat absorption or release.

[0185] The duration of action is the length of time the valve maintains the above opening degree and sustains the gas state.

[0186] Temperature, flow rate, and time together determine the total energy input to and output from the carrier and material system, thereby precisely achieving the target heat absorption ratio adjustment. .

[0187] After receiving the physical control parameters, the thermal management unit immediately drives the actuator:

[0188] The airflow reversing valve switches to the corresponding high-temperature or low-temperature gas source path according to the command; the high-temperature gas pump and flow regulating valve are adjusted to the target speed and opening to output a precise flow rate; the timing control maintains the above state within the set duration; the high-temperature or low-temperature gas flows through the porous channel of the variable specific heat carrier, and undergoes intense heat exchange with the carrier skeleton through forced convection, rapidly changing the carrier temperature, thereby quickly affecting the direction and magnitude of heat flow between the carrier and the material above, and realizing dynamic adjustment of the heat absorption ratio.

[0189] Within the set time window for the compensation action, the system continuously monitors newly generated data in the partition. Numerical value.

[0190] like If the temperature rises above the safety threshold: the compensation is deemed effective, and the control module will reduce or terminate the compensation command in the next cycle, i.e., lower the temperature. .

[0191] like If the price fails to recover or continues to decline: the control module will recalculate and potentially increase its position in the next cycle based on the updated deviation. This value enhances the regulation intensity, increases the flow rate, or extends the action time, forming a closed-loop control process of perception, decision-making, execution, and re-perception.

[0192] Within the set time window following the issuance of the control command, the module continuously monitors the partition. The changing trend. If its value rises back to... If the above conditions are met, the regulation is deemed effective and the market enters a maintenance phase; if the market does not recover or even continues to decline, adjustments will be made in the next cycle based on new data. The risk level and regulatory intensity may be reassessed and potentially upgraded.

[0193] To maintain the sensitivity and accuracy of the system's judgments over the long term, a threshold is required. It features adaptive fine-tuning: the system performs periodic evaluations based on long-term operational data and final product quality feedback. If multiple consecutive batches of products... Lower than the current If no quality defects are found by then, the system will determine the value based on the data from these batches. An adaptive downward adjustment is made to dynamically match the long-term slow changes in kiln performance.

[0194] The early warning module receives and responds to the risk status assessment results issued by the control module, providing operators with tiered status displays and early warnings. The early warning levels are strictly correlated with the aforementioned risk levels, as detailed below:

[0195] Level 3 warning, when First time below In the monitoring interface, the partition identifier turns yellow and flashes, and a log is recorded, indicating that the cooling rate of the kth partition has begun to decrease, and it is recommended to monitor the trend.

[0196] Level 2 warning: When the risk status persists for more than 30 seconds, or the real-time deviation is detected... When the preset lower limit of the medium risk threshold is reached, an audible and visual alarm is triggered. The zone identifier turns orange and remains lit, and a real-time data window for that zone automatically pops up, prompting: Zone k is cooling too slowly. Micro-amplitude thermal wave compensation has been automatically activated. Please confirm the process parameters.

[0197] Level 1 Warning: When the system determines that it has entered a severe risk level, the highest level audible and visual alarm will be triggered, and the zone indicator will turn red. It is recommended to pay attention immediately or prepare for maintenance. Note: Zone k has severely insufficient cooling and a high risk of nitrogen back diffusion! Maximum compensation has been executed. Please check the equipment immediately or prepare for maintenance.

[0198] Level 2 and above alerts require operator confirmation via the interface; the confirmation action will be recorded. When this zone... The value continued to recover to Once the above conditions are met and the situation remains stable for a set period of time, the warning status will automatically reset, and the interface indicators will return to normal.

[0199] The central monitoring screen displays a virtual kiln diagram, using colors to map each zone in real time. The values, along with trend curves, make the thermodynamic state of the entire cooling zone readily apparent.

[0200] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-temperature zone temperature control system for a double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products, characterized in that, include: The functional carrier module is laid in each thermal inertia zone of the cooling area, carrying materials and integrating sensing and thermal management functions; The signal processing module, connected to the functional carrier module, synchronously acquires, amplifies, filters, and digitizes the thermal stress acoustic emission signals and pyroelectric current signals from each thermally inert zone; The feature extraction module, connected to the signal processing module, performs real-time analysis on the digitized thermal stress acoustic emission signal and pyroelectric current signal, extracts and generates a cooling fingerprint feature vector characterizing the cooling dynamics of the thermally inert partition. The inversion module, with a built-in physical information neural network model, is used to receive the cooling fingerprint feature vector and calculate and output a quantized thermodynamic instability criterion. The control module assesses risk based on thermodynamic instability criteria and generates tiered control commands. The early warning module receives the risk assessment results and executes corresponding multi-level early warnings; The control module generates control commands that drive the functional carrier module to perform corresponding thermal management actions.

2. The multi-temperature zone temperature control system for a double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products according to claim 1, characterized in that, The functional carrier module is a variable specific heat transfer medium.

3. The multi-zone temperature control system for a sintered double pusher plate kiln for vanadium-nitrogen alloy semi-product according to claim 2, characterized in that, The signal processing module includes a thermal stress acoustic emission signal channel and a pyroelectric current signal channel that operate independently and synchronously. The thermal stress acoustic emission signal channel is equipped with a high-pass filter with a cutoff frequency of 100kHz and operates at a sampling rate of not less than 1MHz. The pyroelectric current signal channel is equipped with a low-noise current amplifier and operates at a sampling rate of not less than 1kHz.

4. The multi-zone temperature control system for sintering double push plate kiln of vanadium-nitrogen alloy semi-product according to claim 1, characterized in that, The feature extraction module operates within a sliding time window of a preset length. The extracted features include: the centroid feature value of the main frequency distribution and the energy disorder feature value from the thermal stress acoustic emission signal, and the heat flow establishment abruptness feature value and the heat flow relaxation characteristic feature value from the pyroelectric current signal. The centroid feature value of the main frequency distribution, the energy disorder feature value, the heat flow establishment abruptness feature value, and the heat flow relaxation characteristic feature value are normalized and combined into a cooling fingerprint feature vector.

5. The multi-zone temperature control system for sintering double push plate kiln of vanadium-nitrogen alloy semi-product according to claim 1, characterized in that, The physical information neural network model in the inversion module is obtained through pre-training. The training samples are cooling fingerprint feature vectors collected under known cooling rate conditions. The training target value is the ratio of the actual cooling rate to the process safety cooling rate. Furthermore, a physical constraint loss function based on thermodynamic laws is added during the training process.

6. The multi-zone temperature control system for sintering double push plate kiln of vanadium-nitrogen alloy semi-product according to claim 1, characterized in that, The control module determines the risk status by comparing the real-time thermodynamic instability criterion with the dynamic early warning threshold, and classifies the risk level according to the degree of deviation and duration; based on the risk level, it generates a structured instruction containing a control intensity coefficient, which is calculated according to the degree of deviation through a preset nonlinear mapping function.

7. The multi-zone temperature control system of a sintered double pusher plate kiln for vanadium-nitrogen alloy semi-product according to claim 1, characterized in that, The early warning module includes a three-level early warning system linked to the risk level: when a risk is first determined to exist, an alert is triggered, which includes a color change of the interface icon and log recording; when the risk persists or escalates, a warning is triggered, which includes an audible and visual alarm and a pop-up data window. When a situation is determined to be of severe risk, the highest level of audible and visual alarms and maintenance recommendation alarms will be triggered.

8. The multi-temperature zone temperature control system for a double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products according to claim 2, characterized in that, The heat absorption ratio of the variable ratio heat transfer fluid is actively adjustable. The gas circulation control unit actively and continuously changes the temperature and flow rate of the thermal inert gas flowing through the microchannels inside the variable ratio heat carrier according to the control command, and dynamically adjusts the rate at which the variable ratio heat carrier absorbs heat from the material during the cooling process, that is, actively adjusts the heat absorption ratio.

9. The multi-temperature zone temperature control system for a double-pusher kiln for sintering vanadium-nitrogen alloy semi-finished products according to claim 8, characterized in that, The high-temperature heat storage unit is filled with heat storage material and equipped with a heating device to heat the flowing inert gas and maintain it at a constant high temperature; the low-temperature cooling unit is equipped with a heat exchanger to cool the flowing inert gas and maintain it at a constant low temperature.

10. The multi-zone temperature control system for a sintered double pusher plate kiln for vanadium-nitrogen alloy semi-product according to claim 6 or 9, characterized in that, When the control module determines that a certain zone has a risk and calculates the control intensity coefficient, the gas circulation control unit receives the instruction, the reversing valve switches to the passage connecting the high-temperature thermal storage unit, and the gas pump pumps high-temperature inert gas into the variable specific heat carrier at a flow rate proportional to the control intensity coefficient for a preset time to generate a micro-wave of heat that matches the risk level.

Images