Full-automatic high-temperature heat sink sintering furnace based on hydrogen energy power generation SOFC and control method

Abstract

The invention relates to a full-automatic high-temperature heat sink sintering furnace based on a hydrogen energy power generation SOFC. Comprising a furnace body with a sintering cavity, a controllable valve body arranged at a first opening, a movable feeding table bearing a ceramic plate, a heating pipe arranged around a material frame, an adjusting mechanism for driving the heating pipe to be close to or away from the material frame and a temperature sensor arranged around the material frame, wherein the extending direction of the heating pipe is parallel to the moving direction of the feeding table; and the controller is connected with the components. According to the sintering furnace, the position of the heating pipe and on-off of the valve body are adjusted in a linkage mode through the controller based on the detection value of the temperature sensor, local temperature difference real-time compensation and glue discharging disturbance control are achieved, and therefore sintering consistency and heat efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to a sintering furnace, in particular to a full-automatic high-temperature heat sink sintering furnace based on hydrogen energy power generation SOFC and a control method. BACKGROUND

[0002] With the increasing demand for high-reliability metal paste ceramic plates in new energy vehicle power systems, distributed energy storage, and high-end electronic packaging industries, related sintering processes are evolving towards "high temperature, long cycle, batch, and full automation". To reduce carbon emissions and improve energy self-sufficiency, some production lines have begun to introduce solid oxide fuel cells (SOFC) as a stable power source and waste heat complementary means to support continuous operation and energy efficiency coordination in the high-temperature section.

[0003] The existing sintering furnace usually arranges the electric heating element fixedly on the wall or the periphery of the rack, and the temperature detection relies on a small number of thermocouples or infrared point measurements. The sensing position is usually set at the representative point or air flow channel of the furnace cavity. In addition, in the degassing stage, in order to ensure that the organic components are fully discharged, the exhaust valve often needs to be opened.

[0004] However, the above operation is easy to introduce non-uniform convection and break the radiation balance, resulting in local temperature drop and temperature difference expansion in different directions, thereby affecting the degassing effect and product quality. Therefore, it is urgent to propose a full-automatic high-temperature heat sink sintering furnace to solve the above problems. SUMMARY

[0005] The purpose of the present application is to provide a high-temperature heat sink sintering furnace that can adjust the relative position between the heating pipe and the rack based on the multi-point temperature detection results on the periphery of the rack, and control the valve body in parallel to improve the temperature field stability in the degassing stage and optimize the thermal coupling consistency in the high-temperature section.

[0006] The technical scheme adopted by the present application to solve the above problems is: a full-automatic high-temperature heat sink sintering furnace based on hydrogen energy power generation SOFC for sintering of metal paste ceramic plates, comprising a furnace body, a valve body, a feeding table, a heating mechanism, an adjusting mechanism and a controller, the furnace body comprising a sintering cavity and a first opening and a second opening communicating with the sintering cavity; the valve body is arranged at the first opening and is controlled to be opened and closed; the feeding table is used for placing a rack carrying a ceramic plate and is controlled to move to drive the rack to enter or leave the sintering cavity through the second opening, the moving direction of the feeding table being defined as a first direction; the heating mechanism is arranged in the sintering cavity, the heating mechanism comprising a plurality of heating pipes controlled to be heated, each of the heating pipes being configured to be arranged around the rack after the rack enters the sintering cavity, and the extension direction of the heating pipe being parallel to the first direction; the adjusting mechanism comprises an adjusting end controlled to move and transmissionally connected with the heating pipe to drive the heating pipe to move towards the direction of approaching or moving away from the rack; a plurality of temperature sensors are arranged on the side of the furnace body and have detection ends extending into the interior of the sintering cavity, each of the detection ends being arranged around the side of the rack after the rack enters the sintering cavity to obtain a surrounding temperature detection value; the controller is connected with the heating mechanism, the adjusting mechanism and the temperature sensors, and is configured to control the adjusting mechanism and the valve body to operate based on the temperature detection value to adjust the relative position of the heating pipe and the rack and the opening and closing of the valve body.

[0007] In particular, a control method for the full-automatic high-temperature heat sink sintering furnace as described above, the metal paste ceramic plate comprising a green body, the green body being a blank body formed by ceramic powder of the ceramic plate combined under the action of organic binder and / or solvent contained in the metal paste and not subjected to sintering densification treatment.

[0008] The sintering furnace further comprises a gas pressure sensor and a gas component sensor, the gas pressure sensor being arranged on a gas path outside the first opening and located between the valve body and the air exchange fan to obtain the atmospheric pressure in the sintering cavity; the gas component sensor being arranged on the gas path outside the first opening to obtain the gas component in the sintering cavity.

[0009] The control method comprises:

[0010] obtaining the thickness of the green body, the effective heat exchange area of the green body, the mass of the green body, the specific heat of the green body, the initial glue content of the green body, the initial porosity of the green body, the calibrated relationship of the pore permeability of the green body, the calibrated relationship of the allowable pore pressure threshold of the green body, the calibrated curve of the gas viscosity in the sintering cavity and the gas density in the sintering cavity, the atmospheric pressure in the sintering cavity, the maximum allowable temperature difference, the upper limit of the hardware temperature rise slope, the surface radiation emissivity of the heating pipe, the Stefan-Boltzmann constant, the view factor-distance calibrated curve and the thermogravimetric curve.

[0011] According to the thickness of the green body, the effective heat exchange area of the green body, the mass of the green body, the specific heat of the green body, the initial glue content of the green body, the initial porosity of the green body, the calibration relationship of the pore permeability of the green body, the calibration relationship of the allowable pore pressure threshold of the green body, the temperature calibration curve of the gas viscosity in the sintering cavity and the temperature calibration curve of the gas density in the sintering cavity, the atmosphere body pressure in the sintering cavity, the maximum allowable temperature difference, the upper limit of the hardware temperature rise slope, the surface radiation emissivity of the heating pipe, the Stefan-Boltzmann constant, the view factor-distance calibration curve, and the thermogravimetric curve, a target temperature curve is determined;

[0012] According to the target temperature curve and each temperature sensor, the adjusting mechanism and the heating mechanism are controlled to operate.

[0013] The beneficial effects of the embodiment in the present application are:

[0014] 1. Since the detection values of the multiple temperature sensors distributed on the side of the furnace body and extending into the interior of the sintering cavity are collected by the controller, the temperature distribution condition of the side of the rack can be sensed in real time, the controller controls the adjusting mechanism based on the temperature detection values, so that the heating pipe can move towards the direction close to or away from the rack, thereby dynamically changing the radiation heat exchange geometric relationship between the heating pipe and the rack, and realizing real-time compensation for the local temperature difference; at the same time, the controller can also control the on-off state of the valve body arranged at the first opening, so as to adapt to the linkage control requirement of the atmosphere flow and temperature disturbance in the glue discharging stage, therefore, the non-uniformity problem of the temperature field caused by the fixed heating assembly, few temperature measurement points, poor representative arrangement, and temperature disturbance easily caused by exhaust operation in the prior art is effectively solved, especially the key problems such as radiation imbalance and local temperature drop in the glue discharging stage cannot be compensated quickly, thereby realizing the technical effects of sufficient glue discharging, controlled temperature difference, balanced temperature field, and heat efficiency improvement in the sintering process of the metal paste ceramic plate, and significantly improving the consistency of the product and the automation control level of the process, which is suitable for long-period, high-temperature, and high-reliability production requirements. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a schematic structural view of a sintering furnace shown in an embodiment of the present application.

[0016] Figure 2 is a schematic sectional view of a sintering furnace shown in an embodiment of the present application.

[0017] Figure 3 is a schematic structural view of the interior of a sintering furnace shown in an embodiment of the present application.

[0018] Figure 4 is a schematic top view of a sintering furnace shown in an embodiment of the present application.

[0019] Figure 5 is a schematic structural view of a guide plate shown in an embodiment of the present application.

[0020] Wherein: 10, furnace body; 110, sintering cavity; 20, valve body; 30, feeding table; 40, heating mechanism; 410, heating pipe; 50, adjusting mechanism; 510, center pipe; 520, guide plate; 521, first guide groove; 530, driving plate; 531, driving groove; 540, adjusting frame; 541, frame body; 542, driving shaft; 5421, annular protrusion; 543, limiting piece; 5431, third guide groove; 550, moving piece; 551, lifting rod; 552, clamping piece; 5521, second guide groove; 60, temperature sensor; 70, air exchanger. DETAILED DESCRIPTION

[0021] The specific embodiments of the present application will be further described in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not used to limit the scope of the present application.

[0022] Reference Figures 1 to 5The embodiment provides a full-automatic high-temperature heat sink sintering furnace, which takes a solid oxide fuel cell as a cooperative supply source of electric energy and waste heat, and is used for sintering of a metal paste ceramic plate. The sintering furnace comprises a furnace body 10, a valve body 20, a feeding table 30, a heating mechanism 40, an adjusting mechanism 50, a plurality of temperature sensors 60 and a controller. The furnace body 10 comprises a sintering cavity 110 and a first opening and a second opening in communication with the sintering cavity 110, the first opening and the second opening are respectively arranged on opposite sides of the inner wall of the sintering cavity 110, and the center line of the first opening and the center line of the second opening are collinear; the valve body 20 is arranged at the first opening and is controlled to operate; the feeding table 30 is used for placing a rack for carrying a ceramic plate and is controlled to move, so that the rack enters or leaves the sintering cavity 110 through the second opening, the moving direction of the feeding table 30 is defined as a first direction, and the center line of the first opening is parallel to the first direction; the heating mechanism 40 is arranged in the sintering cavity 110, the heating mechanism 40 comprises a plurality of heating pipes 410 controlled to heat, each heating pipe 410 is arranged around the rack after the rack enters the sintering cavity 110, and the extension direction of the heating pipe 410 is parallel to the first direction, wherein the rack comprises a plurality of material placing plates arranged perpendicularly to the first direction and arranged in sequence along the first direction; the adjusting mechanism 50 comprises an adjusting end controlled to move and in transmission connection with the heating pipe 410, so as to drive the heating pipe 410 to move towards the direction close to or away from the rack, and the number of the adjusting end is a plurality of and corresponds to each heating pipe 410 one by one; the plurality of temperature sensors 60 are arranged on the side of the sintering furnace and have detection ends extending into the interior of the sintering cavity 110, each detection end is arranged around the side of the rack after the rack enters the sintering cavity 110, so as to obtain a surrounding temperature detection value, each temperature sensor 60 is divided into a plurality of groups and arranged in parallel to each other, all the temperature sensors 60 in each group are located in the same plane perpendicular to the first direction, each group of temperature sensors 60 is arranged around the rack after the rack enters the sintering cavity 110, and the spacing between adjacent two temperature sensors 60 in each group is equal; the controller is connected with the heating mechanism 40, the adjusting mechanism 50 and the temperature sensors 60, and is configured to control the adjusting mechanism 50 and the valve body 20 to operate based on the temperature detection value, so as to adjust the relative position of the heating pipe 410 and the rack and the on-off of the valve body 20.

[0023] Specifically, the high-temperature heat sink sintering furnace provided in the embodiment takes the furnace body 10 as the main body, a closed sintering cavity 110 is formed in the furnace body 10, the outer shell of the furnace body 10 adopts a composite structure of a high-temperature-resistant heat insulation layer and a force-bearing framework, and the inner lining is a stack of refractory and low-thermal-conductivity materials to reduce heat leakage and steady-state heat insulation. The furnace body 10 is connected to the outside through two openings, a controllable on-off valve body 20 is arranged at the first opening, which is used for discharging volatile substances and adjusting the pressure difference of the cavity, and the second opening is used as a loading and unloading port. The loading table 30 is arranged outside the furnace body 10 and moves linearly along a defined first direction through a guide and drive assembly to realize the smooth entry and exit of the material rack carrying ceramic plates into and out of the sintering cavity 110 through the second opening. The heating mechanism 40 is located in the sintering cavity 110 and adopts a plurality of heating pipes 410 controlled to be heated to surround the material rack, the extension direction of the heating pipes 410 is parallel to the movement direction of the loading table 30, and a radiation and convection channel layout consistent with the workpiece transport direction is formed. In the embodiment, replaceable support and heat insulation gaskets are arranged between the heating pipes 410 and the furnace lining to facilitate maintenance and thermal resistance balancing. The adjusting mechanism 50 is connected with the heating pipes 410 through a moving end and is used to realize the approach and departure of the heating pipes 410 relative to the material rack. A plurality of temperature sensors 60 are arranged on the side of the furnace body 10 and penetrate into the sintering cavity 110 through temperature-resistant sealing elements, and the detection ends are arranged around the side of the material rack after the material rack enters the furnace. The controller is electrically connected with the heating mechanism 40, the adjusting mechanism 50, the temperature sensors 60 and the valve body 20 to form a closed-loop control unit for temperature field sensing and execution. In order to adapt to the solid oxide fuel cell power supply scene of hydrogen energy power generation, the furnace body 10 is provided with interfaces for power supply and waste heat exchange units and is provided with grounding and safety interlocking.

[0024] After the loading table 30 sends the material rack carrying the metal slurry ceramic plates into the sintering cavity 110 along the first direction and completes the sealed closure, the controller calls the target temperature curve matched with the batch process to perform partition control on the temperature of the side of the material rack with the surrounding detection value of the temperature sensor 60 as the feedback reference. Specifically, the controller schedules the heating pipes 410 to output and drives the adjusting mechanism 50 to change the relative position of the heating pipes 410 and the material rack according to the deviation of the real-time temperature of each detection end and the target temperature curve, so that the radiation view factor and the near-wall convection path are quickly corrected, thereby continuously tracking the target temperature curve in the degassing stage. At the same time, the controller controls the opening and closing and opening degree of the valve body 20 at the first opening in linkage with the temperature distribution in the cavity and the volatile release rate to keep the discharge flux and temperature stability coordinated, when local temperature drop or temperature difference expansion trend is monitored, preferentially heat compensation is performed through joint adjustment of the position and power of the heating pipes 410, and slight correction of the opening degree of the valve body 20 is performed to suppress the disturbance caused by non-uniform convection, so that the temperature field of the side of the material rack closely follows the target temperature curve and stably evolves, and it is ensured that the organic components are fully and uniformly discharged.

[0025] The heating pipe 410 can be a metal tubular electric heating element or a ceramic-based electric heating element, and a radiation shaping sheath can be added to the outer surface to optimize the view factor. The heating pipe 410 can be arranged in a rectangular annular array or a multi-row parallel array to adapt to different specifications of the rack and loading modes. The valve body 20 can adopt a butterfly type, gate type or straight-through regulating type structure, and can be configured with a corrosion-resistant lining and a temperature-resistant seal to improve the stability of the glue discharging stage. The drive of the loading table 30 can adopt a lead screw, cable winding or linear actuation scheme, and the guide can adopt a rolling and sliding composite structure to balance positioning accuracy and temperature resistance. The furnace lining level can be adjusted to a combination of composite brick lining or fiber module and dense lining according to the process temperature and thermal shock requirements, so as to balance the heat preservation performance, structural strength and maintenance convenience. The waste heat utilization of the hydrogen energy power supply unit can be coupled to the gas inlet preheating or external drying link through a heat exchanger to realize energy recovery.

[0026] In this embodiment, by using the zone closed-loop control based on the multi-point temperature detection of the furnace body 10, the radiation geometry adaptive layout composed of the heating pipe 410 and the adjusting mechanism 50 surrounding the rack, and the linkage control of the valve body 20 in the glue discharging and steady state stages, the technical problems of uneven temperature field, difficult to suppress non-uniform convection disturbance caused by glue discharging, and local temperature drop and temperature difference cannot be compensated in time caused by fixed arrangement of heating components in the prior art are effectively solved, thereby realizing the technical effects of sufficient glue discharging, controlled temperature difference, balanced temperature field and energy utilization rate improvement, significantly enhancing the consistency of batch products and improving the overall flow automation level.

[0027] In some embodiments, the temperature sensor 60 is a thermocouple sensor, and the extension direction of the thermocouple sensor is perpendicular to the first direction.

[0028] Further, in some embodiments, the adjusting mechanism 50 further comprises a center tube 510, a guide plate 520, a driving plate 530, a plurality of adjusting frames 540 and a plurality of moving pieces 550. The center tube 510 is arranged on the inner wall of the sintering cavity 110 on the side of the first opening, the axis of the center tube 510 is coaxial with the center line of the first opening, one end of the center tube 510 is connected with the first opening, and the other end of the center tube 510 is in communication with the sintering cavity 110; the guide plate 520 is fixedly sleeved outside the center tube 510, a plurality of groups of first guide grooves 521 are arranged on the guide plate 520 in an equidistant and spaced manner with the center of the guide plate 520 as the center, and the extension direction of the first guide grooves 521 is configured to point to the direction close to the center tube 510 or the direction away from the center tube 510; the driving plate 530 is rotatably sleeved outside the center tube 510 and is controlled to rotate, a plurality of groups of driving grooves 531 are arranged on the driving plate 530 in an equidistant and spaced manner with the center of the driving plate 530 as the center, the number of the driving grooves 531 is the same as that of the first guide grooves 521 and each driving groove 531 corresponds to a first guide groove 521, and the axis of the driving plate 530 is collinear with the axis of the guide plate 520; each adjusting frame 540 corresponds to each group of first guide grooves 521 and each group of driving grooves 531, the adjusting frame 540 comprises at least one frame body 541 and a driving shaft 542 arranged at one end of the frame body 541, the number of all the frame bodies 541 is the same as that of the heating pipes 410 and each frame body 541 carries a corresponding heating pipe 410, the driving shaft 542 passes through the first guide groove 521 away from the frame body 541 and is aligned with the driving groove 531, and the end of the driving shaft 542 away from the frame body 541 is the adjusting end; each moving piece 550 is movably connected with each driving shaft 542 and is controlled to move along the first direction, so as to drive the adjusting frame 540 to move along the first direction synchronously when moving along the first direction. When the adjusting mechanism 50 operates, the moving piece 550 drives the adjusting frame 540 to move towards the direction close to the first opening, so that the driving shaft 542 is inserted into the driving groove 531, and then when the driving plate 530 is controlled to rotate, the inner wall of the driving groove 531 abuts against the driving shaft 542, so as to force the driving shaft 542 to move along the extension direction of the first guide groove 521.

[0029] Specifically, the adjusting mechanism 50 is arranged at the side of the first opening of the furnace body 10, the center pipe 510 is made of high-temperature-resistant metal or ceramic composite material, is fixed axially along the inner wall of the furnace body 10, the inner cavity of the pipe body is communicated with the sintering cavity 110, and the end close to the outside of the furnace is sealingly connected with the first opening. The outer surface of the pipe is provided with a temperature-resistant sealing gasket and a positioning ring. The guide plate 520 is an annular disc part coaxially sleeved outside the center pipe 510 and connected with the force skeleton of the furnace body 10 through a heat insulation support to reduce the thermal stress conduction. The guide plate 520 is provided with a first guide groove 521 in the form of an annular array with the center of the guide plate 520 as the center. The groove can be a straight groove, and the extension direction of the groove is oriented or away from the center pipe 510 to realize radial guidance. The driving plate 530 is a rotatable annular disc part coaxially sleeved outside the center pipe 510, limited by a temperature-resistant bearing and a positioning snap ring, and provided with a driving groove 531 corresponding to the first guide groove 521 on the driving plate 530. The edge of the driving groove 531 is subjected to wear-resistant hardening treatment to bear the face contact or line contact of the end of the driving shaft 542. Each adjusting frame 540 is in the form of an arc segment or a door-shaped support structure, one end of which is the frame body 541 supporting the heating pipe 410, provided with a temperature-resistant heat insulation pad and an elastic clamping part 552 to adapt to the thermal expansion of the heating pipe 410, and the other end is provided with the driving shaft 542. The driving shaft 542 passes through the first guide groove 521 and the driving groove 531 along the normal line, and the outer end thereof serves as an adjusting end. The moving part 550 is arranged in the first direction, one end of which passes through a through hole provided on the furnace body 10 to the outside of the furnace body 10, and the end located outside the furnace body 10 is connected with the extension end of the driving part arranged outside the furnace body 10, thereby providing a linear advancing or retreating stroke in the first direction. The rotation of the driving plate 530 is driven by the actuating unit arranged outside the furnace body 10. The driving plate 530 is provided with a guide shaft coaxial therewith, and the furnace body 10 is provided with an arc-shaped groove at the corresponding position to allow the guide shaft to pass through. One end of the guide shaft extends to the outside of the furnace body 10 through the arc-shaped groove and is connected with the brake unit arranged outside the furnace body 10. Under the control of the brake unit and the cooperation of the actuating unit, the guide shaft moves along the arc-shaped groove, thereby driving the driving plate 530 to rotate around the axis of the center pipe 510, and realizing the zoned linkage driving of the adjusting frame 540.

[0030] Further, to ensure the air tightness and safety of the cavity in the glue discharging stage, a follow-up sealing structure is arranged outside the furnace body 10 and the extension part of the guide shaft respectively, and the through hole and the arc-shaped groove are sealingly constrained in a following manner, so that the guide shaft and the moving part 550 keep a relatively sealed state in the reciprocating and rotating process, and the leakage of organic volatile matter and hot gas is avoided and the thermal load of the external components is reduced.

[0031] After the rack is positioned into the furnace through the second opening, the controller performs closed-loop tracking on the temperature of each partition according to the target temperature curve of the glue discharging stage. When geometric radiation compensation needs to be performed on a circumferential partition, the moving piece 550 is advanced in the first direction, causing the corresponding adjusting rack 540 to move as a whole toward the first opening side, and the end of the drive shaft 542 is inserted into the drive groove 531 and abuts against the inner wall thereof; then the drive plate 530 is controlled to rotate by a set angular displacement, the drive groove 531 exerts a tangential component force on the drive shaft 542 and normal constraint thereon, and the drive shaft 542 is forced to move along the extension direction of the first guide groove 521, because the first guide groove 521 is directed toward the center pipe 510, so the adjusting rack 540 is folded inward along the radial direction or spread outward, causing the corresponding heating pipe 410 to approach or move away from the rack. Through the stroke of the moving piece 550 and the angular displacement of the drive plate 530, linkage adjustment of the radial displacement and the circumferential partition can be realized, forming synchronous or differential pose control of a single or multiple heating pipes 410. When reset, the drive plate 530 returns to the initial angle, the moving piece 550 is reversely retreated in the first direction, and the drive shaft 542 exits the drive groove 531 and is limited by the guide plate 520. The whole process can be quickly executed before and after the appearance of the glue discharging atmosphere and temperature disturbance, so that the radiation view factor of the rack periphery and the near-wall convection path maintain a dynamic relationship consistent with the target temperature curve.

[0032] In this embodiment, the composite mechanism of the guide plate 520 coaxially arranged with the center pipe 510 and the rotatable drive plate 530, the geometric limiting and mechanical transmission of the first guide groove 521 and the drive groove 531, and the coordinated technical means of axial insertion and decoupling realized by the moving piece 550, effectively solve the technical problems in the prior art that the relative position of the heating pipe 410 to the rack is difficult to quickly and accurately adjust during the glue discharging stage, the single-partition compensation coupling is large, and the repeated positioning accuracy is poor, and further realize predictable and reproducible regulation and control of the circumferential partition and the radial gap, so that the temperature on the periphery of the rack is more consistent with the target temperature curve and the temperature disturbance and non-uniform convection caused by glue discharging are significantly inhibited.

[0033] It should be noted that the position adjustment of each heating pipe 410 relative to the rack is achieved on the premise that the clamped position of the rack on the loading table 30 is known. The loading table 30 is provided with a reference surface for constraining the posture, a positioning pin and a limiting block, so that the rack forms a repeatable positioning relationship when loaded; after the loading table 30 enters the sintering cavity 110, the relative reference of the loading table 30 and the furnace body 10 is established through the combination of the entry detection and the stroke coding, and the initial position of the loading table 30 relative to each heating pipe 410 is directly given by the calibration relationship table. The controller determines the initial gap and relative position between each heating pipe 410 and the rack before action, and maps the stroke of the moving part 550 and the angular displacement of the driving plate 530 to the radial and circumferential displacement instructions of the heating pipe 410. In the degassing stage, the controller takes the initial alignment as the zero point, combines the target temperature curve and the real-time temperature feedback to execute the position closed loop and the power fine tuning, so that each heating pipe 410 of each sub-area approaches or moves away from the rack according to the predetermined path, thereby ensuring the repeatability and tracking accuracy of the position adjustment process.

[0034] Further, in some embodiments, referring to Figure 3 , the moving part 550 includes a lifting rod 551 and a clamping part 552. The lifting rod 551 is controlled to move along a first direction, and the axis of the lifting rod 551 is parallel to the first direction; the clamping part 552 is arranged at one end of the lifting rod 551 and between the driving plate 530 and the guide plate 520, and a second guide groove 5521 is formed through the clamping part 552, and the extension direction of the second guide groove 5521 is consistent with the first guide groove 521. The circumferential side of the driving shaft 542 is provided with an annular protrusion 5421, and the annular protrusion 5421 is located between the clamping part 552 and the driving plate 530. When the adjustment mechanism 50 operates, the clamping part 552 moves along the first direction with the lifting rod 551 and abuts against the annular protrusion 5421, so as to drive the adjustment frame 540 to move along the first direction and make the driving shaft 542 inserted into the driving groove 531. The adjustment frame 540 further includes a limiting end arranged on the frame body 541 close to the second opening side. The sintering furnace further includes a limiting part 543 arranged on the inner wall of the sintering cavity 110 and corresponding to each adjustment frame 540, and the limiting part 543 is provided with a third guide groove 5431, and the extension direction of the third guide groove 5431 is consistent with the extension direction of the first guide groove 521. When the adjustment frame 540 moves along the extension direction of the first guide groove 521, the limiting end is inserted into the third guide groove 5431, so that the side of the adjustment frame 540 close to the second opening can only move along the extension direction of the first guide groove 521.

[0035] Specifically, the axis of the lifting rod 551 is parallel to the first direction, and the rod body passes through the furnace wall through hole with a temperature-resistant guide sleeve and a sealing element to realize linear reciprocating and maintain air tightness. It should be noted that in the present embodiment, the movement mode of the lifting rod 551 is consistent with the driving mode of the aforementioned moving part 550, and both are provided with a linear reciprocating stroke by the extendable end of the driver arranged outside the furnace body 10. The lifting rod 551 is detachably coupled with the extendable end of the driver via the furnace wall through hole, and a follow-up sealing and guide sleeve are arranged at the through hole to ensure air tightness and coaxiality; a thermal insulation pad and a flexible coupling element are arranged between the driving end and the lifting rod 551 to reduce the influence of heat conduction and assembly deviation on stroke accuracy. The controller adopts a unified stroke calibration and zero point reset strategy for the two types of execution units, and issues displacement instructions based on the partition error criterion of the same target temperature curve, thereby realizing cooperative control of position closed loop and power fine tuning in the glue discharging stage, and ensuring the consistency of the action mode, response characteristics and safety interlocking of the lifting rod 551 and the moving part 550. The clamping part 552 is located between the driving plate 530 and the guide plate 520, and is an annular or saddle-shaped bearing. A second guide groove 5521 is formed on the side facing the guide plate 520, and the extension direction of the second guide groove 5521 is consistent with that of the first guide groove 521, which is used to provide smooth guidance when the driving shaft 542 is inserted and withdrawn. The driving shaft 542 is arranged at the outer end of the adjusting frame 540, and the circumferential surface close to the driving plate 530 is provided with an annular protrusion 5421. The annular protrusion 5421 is located between the clamping part 552 and the driving plate 530, and the side surface of the protrusion is hardened or embedded with a wear-resistant ring to withstand the thrust of the clamping part 552 and the lateral reaction force of the driving plate 530. Each adjusting frame 540 carries a corresponding heating pipe 410, and the frame body 541 forms a limiting end close to the second opening. On the inner wall of the sintering cavity 110 of the furnace body 10 on the second opening side, a limiting part 543 corresponding to each adjusting frame 540 is arranged, and a third guide groove 5431 is formed on the limiting part 543. The extension direction of the third guide groove 5431 is consistent with that of the first guide groove 521, which is used to constrain and guide the movement of the side of the adjusting frame 540 close to the second opening. The driving plate 530 is a rotatable annular disc part coaxially sleeved outside the center pipe 510, and the disc face is provided with a driving groove 531 corresponding to the first guide groove 521; the guide plate 520 is fixed to the center pipe 510, and the first guide grooves 521 are uniformly distributed and paired with the driving grooves 531. The feeding table 30 provides a reference position for the material rack, and the initial position of the feeding table 30 relative to each heating pipe 410 after entering the furnace is known, which is used as the zero point of position control.

[0036] When the glue discharging stage starts, the controller calls the target temperature curve and reads the temperature signals arranged around the rack, selects the circumferential partition that needs to be compensated according to the temperature difference deviation criterion of each partition, and issues the pose adjustment instruction. The lifting rod 551 advances in the first direction, the clamping piece 552 moves forward and aligns the annular protrusion 5421 at the outer end of the drive shaft 542 through the second guide groove 5521, and after abutting, the adjustment frame 540 is pushed in the first direction as a whole, so that the end of the drive shaft 542 is inserted into the corresponding drive groove 531 and abuts against the inner wall thereof. Subsequently, the drive plate 530 is controlled to rotate by a set angular displacement, the drive groove 531 exerts a tangential component force and a normal constraint on the drive shaft 542, and the drive shaft 542 is forced to move in the extension direction of the first guide groove 521, thereby driving the adjustment frame 540 to move radially close to or away from the rack. The limiting end of the adjustment frame 540 close to the second opening is inserted into the third guide groove 5431 throughout the process, and can only move in the same direction as the first guide groove 521, so as to avoid unwanted tilting or twisting. After the pose is in place, the controller makes a slight adjustment to the heating power of the partition in combination with the temperature feedback, so as to make the temperature on the circumferential side of the rack close to the target temperature curve. When it is necessary to remove the mechanical coupling of the partition, the drive plate 530 returns to the initial angular position, the lifting rod 551 reverses and retreats, the clamping piece 552 disengages from the annular protrusion 5421, the drive shaft 542 exits the drive groove 531 and is limited by the guide plate 520, and the partition is reset.

[0037] Before lifting and advancing, the system checks the positioning state of the feeding table 30 and the rack, the zero point and the stroke margin of the lifting rod 551, and the alignment of the clamping piece 552 and the drive shaft 542; after being inserted into the drive groove 531, the system estimates the position according to the calibrated relationship between the angular displacement of the drive plate 530 and the radial displacement of the adjustment frame 540, and links the disturbance of the opening of the valve body 20 to ensure that the temperature fluctuation during the insertion period is controllable; in the reset stage, the drive shaft 542 is monitored to completely exit and the limiting end returns to the neutral zone in the third guide groove 5431, and then the interlocking of the partition is released. The whole process takes the known initial position as the reference to realize the cooperative closed loop of the position loop and the temperature loop.

[0038] The structure is used for high-temperature and volatile conditions in the glue discharging stage. The lifting rod 551, the clamping piece 552, the drive shaft 542 and the guide pair need to be made of temperature-resistant, wear-resistant and low-adhesion materials or surface treatment, and the sealing assembly is suitable for heat cycle and organic gas atmosphere. The installation area should have good ventilation and tail gas purification conditions, and maintain a safe distance from the energy supply and heat exchange unit. The clamping precision and repeated positioning capability of the feeding table 30 should meet the positioning consistency required for partition compensation.

[0039] The clamping piece 552 can adopt an open ring type, a dovetail fitting type or a roller support type structure to reduce pushing friction and inhibit wear; the second guide groove 5521 and the first guide groove 521 can be designed as straight lines to realize linear or nonlinear displacement response; the annular protrusion 5421 can be replaced with a detachable wear-resistant sleeve ring for easy maintenance; the limiting piece 543 can adopt an embedded type, an integral machining type or an adjustable pre-tightening type structure to adapt to different guide precision and thermal expansion compensation strategies; the lifting rod 551 is realized by electric actuation, gas actuation or liquid actuation, and the end can be provided with a flexible connecting piece to absorb thermal displacement; a buffer pad and a stroke limit can be added between the driving plate 530 and the guide plate 520 to avoid abnormal impact.

[0040] In the embodiment, by adopting the axial insertion mechanism composed of the lifting rod 551 and the clamping piece 552, the low-friction insertion and withdrawal realized by the cooperation of the second guide groove 5521 and the first guide groove 521, the stable force transmission realized by the annular protrusion 5421, and the one-way constraint of the adjusting frame 540 by the third guide groove 5431 of the limiting piece 543, the technical problems of the prior art, such as coupling of partition pose adjustment, difficult insertion and alignment, radial and axial motion cross talk and poor repeatability of positioning, are effectively solved, and predictable and reproducible regulation based on the known initial position is realized, so that the position adjustment of the heating pipe 410 relative to the rack is faster and more stable, the temperature on the side of the rack is more in line with the target temperature curve, and the temperature disturbance and non-uniform convection caused by glue discharge are significantly inhibited.

[0041] In some embodiments, referring to Figure 1 , the sintering furnace further comprises an air exchange fan 70 arranged on the air path outside the first opening and located outside the valve body 20, and the air exchange fan 70 is connected with the controller and controlled to operate.

[0042] Specifically, the sintering furnace is provided with an air path assembly outside the first opening, and the air path comprises a valve body 20, a connecting pipe section and an air exchange fan 70 in sequence. The air exchange fan 70 is installed on a support frame outside the furnace body 10 and connected with the pipe section through a damping pad and a flexible temperature-resistant joint to absorb equipment vibration and thermal expansion displacement. The air exchange fan 70 is preferably a fan body and a protective cover made of high-temperature-resistant and corrosion-resistant materials, and the motor adopts an explosion-proof and insulating structure. A wear-resistant lining is arranged between the impeller and the shell to reduce the erosion of concentrated volatile substances to the flow channel. In order to ensure air tightness and convenient maintenance, temperature-resistant sealing pads and quick locking pieces are arranged at the flanges of the air path, and tail gas purification units and silencing assemblies can be connected in parallel at the outlet side of the fan. The controller is connected with the air exchange fan 70 through a frequency conversion or speed regulation module, collects temperature and pressure difference signals from the furnace body 10 and the air path, and cooperatively controls the fan speed and the opening of the valve body 20 as adjustable execution quantities.

[0043] After the loading table 30 loads the material into the furnace and completes the sealing, the controller calls the target temperature curve of the glue discharging stage, performs closed-loop adjustment on the temperature difference of each partition according to the surrounding temperature measurement signal, adopts linkage compensation of the heating pipe 410 power and the relative position, and realizes passive guiding and discharging of volatile substances in the small opening mode of the valve body 20 and maintains the cavity pressure difference. The air exhaust fan 70 remains in the closed state in this stage, and the start-stop and speed command of the fan in the control logic are shielded to avoid introducing additional convection disturbance to affect the tracking of the target curve by the temperature on the side. When the glue discharging and subsequent temperature rising and holding steps are completed, the system enters the cooling stage and the temperature drops below the set safety threshold, the controller unblocks and slowly starts the air exhaust fan 70, establishes a stable exhaust flow through the gradual speed curve combined with the valve body 20 opening, and tracks the target cooling curve, while resetting the heating pipe 410 position to a neutral safety gap to suppress local hot spots and thermal stress gradient. During the cooling process, the fan speed and the valve body 20 opening are adjusted according to the pressure difference and temperature slope, and the interlocking strategy of fan speed reduction and valve body 20 closing is executed in abnormal conditions to ensure the air tightness and safety boundary.

[0044] The controller sets the stage state machine, and the fan control amount is locked in the glue discharging stage. Only when the sintering process is completed and the unlocking criteria of temperature and gas conditions are met, the fan can be powered on and the speed adjustment is allowed; the valve body 20 assumes the only gas path opening execution responsibility in the glue discharging stage, and is decoupled from the fan; the cooling stage enables the coordinated control curve of the fan and the valve body 20, and keeps the interlocking with the loading table 30 lock, the furnace door state and the tail gas purification pressure loss. To ensure position control consistency, the heating pipe 410 position is reset and zero point is checked before the fan is started, to ensure that the macro flow field generated by the fan will not be superimposed on the fine radiation geometry compensation link.

[0045] In this embodiment, by adopting the segmented control technology of shielding the fan in the glue discharging stage, relying only on the valve body 20 to realize passive guiding and discharging, and using the heating pipe 410 power and position to compensate the temperature difference in the glue discharging stage, and then starting the fan and the valve body 20 to track the cooling curve in the cooling stage, the non-uniform convection and temperature disturbance caused by the fan exhaust in the glue discharging stage in the prior art are effectively solved, and the stable tracking of the target temperature curve in the glue discharging stage and the efficient controlled cooling in the cooling stage are realized, so that the product consistency and process safety are simultaneously improved.

[0046] Further, it needs to be explained that the metal slurry ceramic plate includes a green body, which refers to a blank body formed by the ceramic powder of the ceramic plate combined under the action of the organic binder and / or solvent contained in the metal slurry and without sintering densification treatment. The sintering furnace further includes a gas pressure sensor and a gas component sensor, the gas pressure sensor is arranged on the gas path outside the first opening and between the valve body 20 and the air exchange fan 70, so as to obtain the atmospheric body pressure in the sintering cavity 110; and the gas component sensor is arranged on the gas path outside the first opening, so as to obtain the gas component in the sintering cavity 110. In order to improve the final sintering quality of the metal slurry ceramic plate, based on the above scheme, a control method of the sintering furnace is further provided, which includes the following steps:

[0047] Step S100: obtaining the thickness L of the green body, the effective heat exchange area A of the green body, the mass m of the green body p , the specific heat C of the green body p , the initial glue content m of the green body b0 , the initial porosity φ0 of the green body, the calibrated relationship of the pore permeability k of the green body perm (φ), the calibrated relationship of the allowable pore pressure threshold p of the green body max (T, φ), the calibrated curve of the gas viscosity μ(T) in the sintering cavity 110 and the calibrated curve of the gas density ρ v (T) in the sintering cavity 110, the atmospheric body pressure p in the sintering cavity 110 bulk , the maximum allowable temperature difference ΔT allow , the upper limit of the hardware temperature rising slope r max , the surface radiation emissivity ε of the heating pipe 410, the Stefan-Boltzmann constant σ, the apparent factor-distance calibrated curve F(r) and the thermogravimetric curve m b (T).

[0048] Step S200: according to the thickness L of the green body, the effective heat exchange area A of the green body, the mass m of the green body p , the specific heat C of the green body p , the initial glue content m of the green body b0 , the initial porosity φ0 of the green body, the calibrated relationship of the pore permeability k of the green body perm (φ), the calibrated relationship of the allowable pore pressure threshold p of the green body max (T, φ), the temperature calibrated curve of the gas viscosity μ(T) in the sintering cavity 110 and the temperature calibrated curve of the gas density ρ v (T) in the sintering cavity 110, the atmospheric body pressure p in the sintering cavity 110 bulk , the maximum allowable temperature difference ΔT allow, the hardware temperature rise slope upper limit r max , the heating pipe 410 surface radiation emissivity ε, the Stefan-Boltzmann constant σ, the view factor-distance calibration curve F(r), and the thermal weight curve m b (T) determining a target temperature curve;

[0049] Step S300: controlling the adjusting mechanism 50 and the heating mechanism 40 to operate according to the target temperature curve and each temperature sensor 60.

[0050] Wherein, each parameter in step S100 can be determined by the following method:

[0051] The green body thickness L can be obtained by the following method: if the metal slurry ceramic plate is a plate member and the geometric shape is regular, a micrometer or a thickness gauge with a resolution of 0.01 mm is used. The thickness L is measured at five points including the four corners and the center of the plate member i , and the average is taken If correction is needed according to the process temperature T (considering thermal expansion), the linear expansion coefficient α T is used T (T-25℃)); wherein L: the nominal thickness of the green body at room temperature; L(T): the equivalent thickness at temperature T; α T : the linear expansion coefficient; T: the process temperature. If the metal slurry ceramic plate is a plate member and the edge is complex / multi-layer, the projection / caliper edge thickness is measured in combination with local slicing measurement or the average cross-sectional thickness is directly measured by three-coordinate / laser displacement scanning, and the average value is taken according to the above formula.

[0052] The effective heat exchange area A (the outer surface area of the green body visible to the furnace gas / radiation) can be obtained by the following method: plate member approximation: length a, width b, thickness L: A≈2ab+2L(a+b); wherein A: the outer surface area of the green body visible to the furnace atmosphere / radiation. If there is shielding (material rack clamping jaw, baffle), the corresponding shielding area is deducted; when multiple pieces are loaded, the sum is taken. Complex geometry: based on the CAD model to derive the surface area; or the measured outer dimensions are composed of the sum of the patch areas.

[0053] Batch consistency: establish a "specification-area" table for the first piece, and call in batches.

[0054] The green body mass m p (the total mass of the green body loaded into the furnace in this batch) can be obtained by the following method: tool: electronic balance with an accuracy of 0.01 g. Sampling conditions: room temperature, room humidity; if free water is contained, first dry at 105℃ for 30min to 60min to remove surface water before weighing (to avoid damaging the binder). Multiple pieces are loaded: each piece is weighed or average is taken by sampling Total mass N: number of pieces Average mass of a single piece.

[0055] Specific heat of chlorite C p (T) (isobaric specific heat at temperature) can be obtained by DSC calibration (ASTM E1269, same kind): instrument: differential scanning calorimeter (aluminum crucible / inert gas). With sapphire as a standard sample, the temperature rise rate is 10 K / min -1 , and the interval is 25-600℃. Calculation: output C p (T); stored as a piecewise polynomial: C p (T) ≈ a0 + a1T + a2T 2 ; or obtained by mixing method when the formula is known. Powder volume fraction φ cer , binder volume fraction φ b : C p (T) ≈ φ cer C p,cer (T) + φ b C p,b (T); where a0, a1, a2: fitting coefficients; φ cer , φ b : ceramic phase, binder volume fraction; C p,cer (T), C p,b (T): specific heat of each phase, taken from the material manual or existing DSC data.

[0056] Initial binder content m b0 can be obtained as follows: instrument: thermal gravimetric analyzer (inert atmosphere N2 / Ar). Sampling: 10-20 mg of the same batch of chlorite powder is crushed. Procedure: 25→600℃, 5-10 K / min -1 . Calculation: initial and final mass m ini , m 600 , m b0 = m ini -m 600 (or mass fraction ); solvent extraction calibration: organic matter is extracted with a solvent (such as ethanol / acetone), and after drying, the mass is re-measured to obtain for cross-validation. Where m b0 : total mass of the initial organic binder / solvent in the chlorite; m ini , m 600 : TGA starting / ending mass; ω b0 = m b0 / m ini : binder content mass fraction.

[0057] Initial porosity φ0can be obtained as follows: bulk density-true density method: bulk density V is obtained from geometric dimensions (or drainage method). True density p s : helium pycnometer (gas displacement method). Porosity: Open pore correction: if open pores are high, use Archimedes method (liquid immersion) to correct volume V. Where, φ0: total porosity in the sense of green body volume fraction; p bulk : bulk density; p s : solid true density; V: apparent volume.

[0058] Pore permeability-porosity calibration k perm (φ) can be obtained as follows: Darcy steady-state gas permeation experiment: clamp: effective cross-sectional area A, sample thickness L, upstream and downstream pressure P1, P2, volumetric flow meter Q, temperature T. Use gas: same control gas as process. Prepare green body samples with several φ levels under different forming pressure / sintering pretreatment, measure Q (ΔP) one by one. Calculate (Darcy's law): Fitting (Kozeny-Carman experience): Where: d p is the average particle size (measured by laser particle size analyzer), K c (Kozeny constant) is obtained by least squares fitting; or directly with (φ i , k i ) to do polynomial / power law fitting, table lookup calculation; k: gas Darcy permeability; L s : gas permeation sample thickness (different from green body thickness L); A s : sample flow cross-sectional area; ΔP: upstream and downstream pressure difference; viscosity μ (T) is described later.

[0059] Allowable pore pressure threshold-temperature / porosity calibration p max (T, φ) can be obtained as follows: Mechanical strength experiment is realized by equivalent pore pressure conversion. Sample: bending sample (such as 3×4×40mm) consistent with green body formulation. Equipment: high temperature three-point bending device, test temperature covers 25-600℃. Get bending strength σ t (T, φ) (grouped according to sample porosity).

[0060] Obtaining of empirical conversion coefficient κ: differential pressure foaming test: make thin plate, pressurize the back cavity, record the cavity pressure p * when the first foaming / cracking occurs, record σ t at the same time, take κ = p * / σ t average or conservative quantile; or process calibration: gradually increase the temperature ramp in the test furnace until microcracks appear, obtain the pore pressure peak p * from the corresponding model, and accordingly deduce κ. Final relationship: p max (T, φ) = κσ t(T, φ), κ take conservative values (e.g. P10 percentile); p max : maximum equivalent pore pressure allowed inside the pores (to avoid blistering / cracking); σ t : equivalent bending / tension strength of the green body; κ: force-pressure conversion factor (take conservative values).

[0061] Gas viscosity μ(T), gas density ρ v (T) as a function of temperature, can be obtained by calculation based on the process gas composition and temperature, corrected if necessary by composition sensors. Let the mixture consist of components i (e.g. N2, O2, CO2, H2O, and small amounts of volatiles) with volume fractions y i : density (ideal gas approximation):

[0062] Viscosity (Wilke mixing method): knowing the pure component viscosities μ i (T) (Sutherland / Chapman-Enskog formula), mixture viscosity:

[0063] Component acquisition: if only a "control gas" (e.g. nitrogen) is used and the ratio is fixed: set y i to a constant value; if a gas composition sensor is configured: update y i according to real-time readings. Calibration curve: discretize T according to the process temperature range, calculate and store μ(T), ρ v (T) look-up table. Wherein, μ(T): mixture viscosity; ρ v (T): mixture density; p bulk : cavity body pressure; R: gas constant; M i : component molar mass; μ i (T): pure component viscosity (obtained from Sutherland / Chapman-Enskog); y i : volume fraction; Φ ij : Wilke interaction term.

[0064] Atmosphere body pressure p bulk (sintering cavity 110 / gas line gas absolute pressure or relative pressure (as selected), used for pore pressure safety and flow estimation), can be obtained by the following way: sensor: absolute / relative pressure transmitter installed outside the first opening, between the valve body 20 and the gas fan 70, with heat insulation / condensation bend. Calibration: two-point / three-point calibration with atmospheric pressure point and known pressure point; temperature compensation according to manufacturer's instructions. Reading: controller directly reads, if necessary, smooth filtering (such as first-order low-pass, time constant 0.5-2s).

[0065] Maximum allowable temperature difference ΔT allowThe maximum allowable temperature difference between each cross-section can be obtained through an engineering-determined method (material / geometry related). Non-uniform heating (or jet cooling) is set up on both sides of the test furnace, and the temperature difference ΔT is gradually increased until microcracks appear. The critical ΔT is recorded. c ;ΔT allow =ηΔT c η∈[0.5,0.7] (safety factor); or take the material thermal expansion α. T Young's modulus E, Poisson's ratio ν, fracture strength σ f According to the approximation of thermal stress in a flat plate: Where: σ f The aforementioned σ can be used t Instead of taking the conservative approach, σ th E: Thermal stress estimation; E: Young's modulus; ν: Poisson's ratio; σ f η: Fracture strength; η: Safety factor. Upper limit of hardware heating rate slope r max (The maximum permissible heating rate of the device under specified loading and clearance conditions (K min)) -1 The temperature can be obtained in the following way (no-load / half-load step test): Set the upper limit of the surface temperature of heating tube 410. Step heating was performed at different r (neutral gap) values, and the controlled region T was recorded. p The initial slope of (t) Select the maximum sustainable slope that will not cause overshoot / overcurrent as r. max Multiply by a safety factor of 0.7 to 0.85. Verify once after batch loading, and use the smaller value.

[0066] The surface emissivity ε(T) of heating tube 410 can be obtained by the following method: thermocouple-infrared comparison calibration: in a static furnace, the surface of heating tube 410 is stabilized at several temperature points T. w (Measured using a mounted thermocouple). Simultaneously, a temperature reading of T is obtained using an adjustable emissivity infrared thermometer. IR Adjust ε to make T IR =T w Record ε(T) w Piecewise linear / polynomial fitting of ε(T) (the hemispherical emissivity of the surface of heating tube 410 at temperature T; T) w Surface temperature of heating element 410; T IR (Infrared inversion temperature) is stored in the table; if the surface treatment is changed, it needs to be re-labeled.

[0067] The Stefan-Boltzmann constant σ can be obtained as follows: Physical constant: σ = 5.670374419 × 10 - 8 W m -2 K -4It can be used directly as a constant.

[0068] The apparent factor—distance calibration curve F(r) can be obtained through geometric calculation and empirical verification. Calculation: Based on the relative positions of the heating tube 410 (approximately cylindrical / ringed) and the plate (rectangular plate), the apparent factor between surfaces is solved using a closed-loop or numerical method. When no readily available analytical solution is available, a Monte Carlo ray tracing / radiative network is used to obtain a point series of F(r) (the geometrical apparent factor between the heating tube 410 and the green body (which varies with distance r (the closest normal distance or equivalent distance from the heating tube 410 to the green body)) and fitted as follows:

[0069] Verification: Under normal atmospheric pressure and static atmosphere, with a fixed temperature T p Temperature is increased at a low slope, r is changed, and steady-state temperature is recorded. p ,T w ,Depend on The inverse solution F(r) is compared with the calculated value, and (a,b,n) is fine-tuned.

[0070] Where a, b, n: empirical fitting coefficients; q rad Radiative heat flux; T p : Characteristic temperature of the green body (represented by the measurement point represented by the surrounding thermocouple).

[0071] thermogravimetric curve m b (T) (residual organic matter mass with temperature) can be obtained as follows: see TGA test, derive m b (T) and Save as a lookup table or polynomial / spline function for use with S200 slope limits and insulation criteria.

[0072] The κ (coefficient for converting flexural strength to equivalent pore pressure) in the above parameters is calibrated once according to the aforementioned "differential pressure bubbling / process calibration", taking a conservative percentile. The rate of weight loss per unit temperature increase (reflecting adhesive expulsion strength). Kozeny constant K. c Or the regression coefficient of k(φ): obtained by multi-point regression according to the Darcy experiment described above, R 2 Freeze after reaching ≥0.95. ε(T) and F(r): Obtained according to the offline calibration → online verification process described above, forming a traceable calibration version number.

[0073] The target temperature curve in step S200 can be obtained through the following steps:

[0074] Based on the pore permeability calibration relationship of green bodies, the allowable pore pressure threshold calibration relationship, the temperature calibration curves of gas viscosity and gas density in sintering chamber 110, and the atmosphere pressure, the upper limit of the temperature change rate to avoid the pore pressure exceeding the threshold is determined at different temperatures. Combined with the maximum allowable temperature difference, the upper limit of the hardware heating slope, the surface emissivity of heating tube 410, the Stefan-Boltzmann constant, and the apparent factor-distance calibration curve, the upper limit of the temperature change rate limited by temperature difference and hardware capability at different temperatures is determined. The smaller of the two upper limits is used as the limiting condition for the corresponding temperature. The temperature-time relationship is generated and its consistency is checked using the thermogravimetric curve to determine the target temperature curve.

[0075] Specifically, step S210: Data discretization and preprocessing.

[0076] Temperature range upper and lower limits T min ,T max (e.g., 0℃ and 600℃, in this embodiment the upper limit of the binder removal temperature of the metal slurry ceramic plate during the sintering process is 600℃), time step Δt (e.g., 1~5s), temperature step ΔT (e.g., 1K).

[0077] Procedure: Generate temperature grid T k =T min +kΔT. For the thermogravimetric curve m b (T) Perform numerical differentiation. To suppress noise, first use Savitzky-Golay smoothing (window width w should be an odd number, w = 11~21 is recommended, polynomial order 2), then perform central difference:

[0078] Determining the window / order: Calculate the differential noise standard deviation σ for the stationary segment with T < 100℃. m′ Adjust w until σ m′ It no longer decreases significantly (empirically, increasing w until the noise improvement is less than 10% is sufficient to freeze the process).

[0079] Step S220: Weight loss peak threshold and insulation termination condition. Substitute m from step S100. b (T), m b0 .

[0080] (1) Weightlessness rate threshold (used for whether to keep warm): θ=zσ m′ z = 3.

[0081] Where: σ m′ This represents the differential noise estimated in the previous step; z = 3 represents the 3σ discrimination (obtained directly from the statistics in step S210), as shown below:

[0082] Objective: To define the "weightlessness rate threshold": θ = zσ m′The z-axis uses a data-driven determination method that ensures that "non-decomposition noise" is almost never misjudged as a "weight loss peak" that requires insulation.

[0083] 1. (Take a "noise" sample).

[0084] Select a temperature range [T] during the stable phase before glue removal begins. lo ,T hi (For example, the range of 25–80℃ or the interval before the first significant drop in the TGA curve). For the entire m... b (T) First perform Savitzky-Golay smoothing (window w should be odd, 11-21 recommended, order 2), then use central difference to calculate Only extract the derivative sample set from the stationary segment:

[0085] 2. Estimate (“noise scale” σ) m′ ).

[0086] Conventional estimation Typically μ m′ ≈0.

[0087] Robust estimation (for a small number of outliers): Calculate first Again σ m′ ≈1.4826×MAD.

[0088] 3. (Change "allowed false alarm rate" α to z).

[0089] like Approximately normal (Shapiro–Wilk or KS test can be performed, p>0.05): Given the target "false alarm probability" α (the probability of treating pure noise as a peak; in engineering, α = 0.0027, i.e., 0.27%), then: z = Φ -1 (1-α / 2);

[0090] When α = 0.0027, z ≈ Φ -1 (0.99865)≈3, which is the “3σ criterion”.

[0091] like Non-normal distribution: Use empirical quantiles directly to set the threshold.

[0092] Then, we can deduce z back to z = θ / σ m′ This is just for record-keeping. This ensures that "in the stationary phase, only the probability of α being falsely triggered".

[0093] Conclusion: Once the statistics in step S210 are completed, σ is obtained. m′Once the expected false alarm rate α is selected (0.27% is recommended), z can be directly obtained using the above formula (z = 3 for normal distribution; use the empirical 99.73% quantile for non-normal distribution).

[0094] (2) Permissible residual adhesive ratio (one of the conditions for termination of insulation): ε r (e.g., 1–2%).

[0095] Method of obtaining: Perform a small-scale isothermal replenishment test (TGA isothermal 30–60 min) using samples with the same formulation to verify that when the residual adhesive is ≤ ε r Subsequent sintering density / electrical properties are qualified; the maximum acceptable residual material that meets the quality requirements is taken as ε. r .

[0096] (3) Maximum heat preservation time τ max (Termination Condition Two):

[0097] Method of obtaining: at a representative temperature T h Perform isothermal TGA at the center of the main weightlessness peak temperature region and fit the first-order kinetics m. b (t)=m b,∞ +(m b,0 -m b,∞ )e -t / τ We obtain the time constant τ. Let τ... max =3τ (3 time constants, residual <5%)

[0098] The specific method is as follows:

[0099] When entering the insulation At that time, an upper limit τ is given to "how long should the temperature be kept warm before it should be increased again". max This ensures that the residual adhesive is sufficiently low (<5%) without unnecessarily delaying the process.

[0100] (1) Select a representative temperature T h From the already smoothed Find the dominant weightless peak (the peak with the largest area or height), and take its position as: If there are multiple similar peaks, the center temperature of the peak with the largest mass loss area can be taken.

[0101] (2) Perform isothermal TGA experiments. Conditions: same atmosphere as the process; rise to T at a relatively fast slope. h (To avoid premature decomposition), then record m at a constant temperature. b (t) approaches a plateau. Initial / final value: m b,0 =m b (t=0)(just reached T) h (time), m b,∞ (Platform average over a long period of time).

[0102] (3) Fit the first-order dynamics to obtain the time constant τ. Fit using nonlinear least squares (such as the LM algorithm): m b (t)=m b,∞ +(m b,0 -m b,∞ )e -t / τ Initial value suggestion: τ (0) Take the time to reach 63% decay; Fit quality requirement: coefficient of determination R0 2 ≥0.98 indicates that the residuals are not systematic. If a single first-order reaction does not fit well (e.g., R0.98), the result is not systematic. 2 <0.98 or the residual shows systematic bias), use two items in parallel: m b (t)=m b,∞ +A1e -t / τ1 +A2e -t / τ2 Choose between a 1-term or 2-term model using AIC / BIC; then define: τ = max(τ1, τ2) (taking the time constant of the slowest process);

[0103] (4) Give the upper limit of heat preservation time: τ max =3τ: Reason: e -3 ≈0.0498, meaning that by 3τ, the remaining fraction of undecomposed material is ≤5%. Expressed in terms of residual glue content: m b (t=3τ)-m b,∞ =(m b,0 -m b,∞ )e -3 ≤0.05(m b,0 -m b,∞ Therefore, in the control logic, if the insulation reaches 3τ but another termination condition (residual adhesive allowable ratio ε) is not met, the control logic will terminate the process. r It can also be determined that continuing to raise the temperature will not significantly impair the adequacy of glue removal.

[0104] (5) Consistency and safety values ​​across multiple batches: Repeat the above isothermal TGA (≥3 times) and take the statistical upper bound of τ (e.g., P90 quantile) as the batch parameter; if the process needs to be more conservative, τ can be taken as the upper bound of the batch parameter. max =3τ P90 ×γ, where γ∈[1.0,1.2] is the safety factor. The sample set of derivatives in the stationary segment; μ m′ ,σ m′ : Mean and standard deviation; MAD: Median absolute deviation. α: False alarm probability; Φ -1 (·): Quantile function of the standard normal distribution; The empirical quantile. T h The main weightlessness peak represents temperature; m b,0 ,m b,∞τ: Residual gum mass at the start and end of isothermal decomposition; τ: Time constant of isothermal decomposition; max : Upper limit of insulation control. A1, A2, τ1, τ2: Amplitude and time constant of two first-level models; AIC / BIC: Information criteria (used for model optimization). Following the above process: z is given by the distribution statistics of your own data with one click; τ max It is directly calculated from a single (or a small number) isothermal TGA fitting.

[0105] Step S230: Gas property curves and cavity pressure.

[0106] Select μ(T),ρ in step S100 v (T) Calibration method, gas composition y i (Fixed or derived from a gas composition sensor), gas pressure p bulk .

[0107] If the composition is fixed: calculate and store μ(T) at each temperature node according to the mixing rules and formulas in step S100 (Wilke mixing method, ideal gas law). k ),ρ v (T k The specific steps are as follows:

[0108] Obtain the known parameters from step S100: component volume fraction: y i (e.g., i∈{N2,O2,CO2,H2O,…}), ∑ i y i =1; Molar mass of component: M i (kg·mol -1 Pure component viscosity model parameters (choose one, Sutherland recommended); Sutherland: μ i,ref (Pa·s), T i,ref (K), S i (K); or Chapman–Enskog / LJ model coefficients (also obtained in step S100); nominal body pressure: p bulk (Pa); Universal gas constant: R u = 8.314462618 J·mol -1 ·K -1 Temperature grid: settings (K is the number of nodes, in K; for example, 300–900K, with a step size of 5K).

[0109] Calculate node by node, for each T k :

[0110] Viscosity of pure components (using Sutherland as an example)

[0111] Wilke Mixed Viscosity:

[0112]

[0113] Average molar mass of the mixture:

[0114] Density (ideal gas):

[0115] Table saving / retrieval implementation: Generating three lists: (T k ,μ(T k ),ρ v (T k )).

[0116] Saved as CSV / EEPROM / controller parameter block; at runtime, μ(T), ρ at any temperature can be obtained using one-dimensional linear interpolation or piecewise spline interpolation. v (T).

[0117] Unit consistency check: p in Pa, T in K, μ in Pa·s, ρ in kg·m -3 .

[0118] If a gas composition sensor is subsequently configured, only an update to y is required before operation. i And recalculate the table; S300 can also periodically (e.g., every 10–60 seconds) recalculate the table according to real-time y. i refresh.

[0119] If a component sensor is present: record the current y value before operation. i Calculate the curve; allow updates based on the sensor during step S300.

[0120] p bulk The pressure sensor reads the value online; step S200 uses its nominal set value, and step S300 uses the real-time value.

[0121] S240 permeability vs. allowable pore pressure curve:

[0122] Using k in S100 perm (φ), p max (T,φ).

[0123] Take the porosity φ0 of this batch of chloroplasts, and find k from the table / substitute the values. perm (φ0), as follows:

[0124] In the aforementioned step S100, a permeability-porosity calibration (one of two equivalent forms) has been established:

[0125] Form 1: Analytical model (Kozeny–Carman regression).

[0126]

[0127] Where, d p The equivalent particle size (m) is given by the particle size analyzer; K c The constant is the Kozeny constant (dimensionless) obtained from the regression.

[0128] Substituting the actual measured φ0 of the batch directly into the equation, we get k. perm (φ0)(unit m) 2 ).

[0129] Form 2: Calibration table (φ–k point column).

[0130] There are N sets of data

[0131] Interpolation: To ensure monotonicity and positive values, it is recommended to perform piecewise linear interpolation in the logk space.

[0132] Find φ j ≤φ0≤φ j+1 ,make:

[0133] k perm (φ0) = exp(log k(φ0));

[0134] Out of bounds: When φ0 exceeds the table range, only one-sided linear extrapolation is allowed and an alarm is given; a more reliable approach is to first extend the calibration.

[0135] For each temperature node T k p is obtained from the calibration relation max (T k ,φ0), specifically as follows:

[0136] Two calibration steps have been completed in step S100:

[0137] (1) Obtain the two-dimensional relationship σ of high temperature bending strength t (T,φ);

[0138] (2) The coefficient κ was obtained by calibrating the pressure differential bubbling / process, and the following was established: p max (T,φ)=κσ t (T,φ);

[0139] Implementation methods (two commonly used and equivalent methods):

[0140] Method 1: Two-dimensional polynomial regression (completed offline in one step).

[0141] Using experimental data {T m ,φ n ,σ t,mn Fitting a low-order polynomial (recommended up to quadratic): σt (T,φ)≈a 00 +a 10 T+a 01 φ+a 20 T 2 +a 11 Tφ+a 02 φ 2 ;

[0142] The regression yields the coefficients {a} ij}(least squares, R) 2 ≥0.98; if insufficient, add a cubic term or use a spline.

[0143] At runtime: for each node T k , first calculate σ t (T k ,φ0), then p max (T k ,φ0)=κσ t (T k ,φ0);

[0144] Method 2: Two-dimensional spline interpolation (better shape preservation).

[0145] σ t The data is placed on a regular / irregular grid to construct tensor product splines (regular grid) or thin plate splines / RBF (scatter).

[0146] At runtime: Input (T) k ,φ0), the spline directly outputs σ t Then multiply by κ. κ is the ratio obtained by back-calculating the "critical crack slope" from the pressure differential bubbling test or process. Take the conservative quantile (e.g., P10) as κ and fix it as the library value of the formula / geometry.

[0147] Calibration coefficient acquisition: k perm The regression coefficients (e.g., the Kozeny constant K) of (·) c ) Darcy's experiment regression from step S100; p max The κ in (·) comes from the differential pressure bubbling / process calibration in step S100, and a conservative quantile (such as P10) is taken.

[0148] The upper limit of the heating slope for step S250 where the orifice pressure does not exceed the limit.

[0149] Obtain A,L,m from step S100 b0 ,μ(T),ρ v (T),k perm (φ0),p max (T,φ0),p bulk ,as well as Required parameter: Volatile matter mass coefficient Y v .

[0150] Y v How to obtain:

[0151] Conservative selection: If the TGA residue after 600℃ is an inorganic phase and no solid carbon retention occurs, then Y is selected. v =1.

[0152] When carbon residue is present: Perform one TGA test each under inert (N2) and oxidizing (air) conditions, and record the residue mass difference Δm. char Considered as carbon retention, χ char =Δm char / m b0 Then Y v =1-χ char .

[0153] MS / FTIR: directly using the mass fraction of the volatile phase as Y v .

[0154] Finally, calculate (for each T) k ):

[0155] Step S260: The upper limit of the heating slope where the temperature difference does not exceed the limit.

[0156] Get h min A,m p C p (T),ΔT allow Among them, h min The method for obtaining it is as follows:

[0157] Experimental method: Under no-load and representative loading conditions, with the valve body at minimum disturbance state 20, a small power step was performed to identify the near-wall convection coefficient, and the minimum value was taken as h. min The remaining parameters can be obtained from step S100. Calculate (for each T) k C p Take the temperature value or the piecewise constant):

[0158] S270 overall slope limit and hardware limit.

[0159] Get

[0160] calculate:

[0161] Unit conversion: If r max With K min -1 Given, first convert to K s -1 Then take the smallest value.

[0162] in, The upper limit of the heating slope for when the pore pressure does not exceed the limit is given by the following formula:

[0163] Each quantity in this formula is obtained from the steps mentioned above:

[0164] ρ v (T k (Density of the mixed gas): Calculated from composition and temperature; When the composition is fixed, it is based on the preset y i Generate (T,ρ) v Look up the table; if there is a component sensor, update and recalculate according to the reading composition. μ(T) k (Volume of mixed gas): Wilke mixing method; μ is first obtained using Sutherland / Chapman–Enskog. i (T), then The table (T,μ) is generated offline. perm (φ0) (chloroplast permeability): Calibrated by Darcy's air permeability test;

[0165] Form 1: Kozeny – Carman Returns Substitute φ0 for this batch;

[0166] Form 2: (φ i ,k i The point sequence is obtained by piecewise linear interpolation of logk to obtain k(φ0).

[0167] A (Effective heat exchange area of ​​the green body): Geometric calculation / measurement; for rectangular plates, A≈2ab+2L(a+b), for complex shapes, use CAD / measurement and summation (deducting the obstruction of the material rack). L (Green body thickness): Thickness gauge / micrometer, take the average value from multiple points; correct for thermal expansion if necessary.

[0168] p max (T k ,φ0)(Allowable pore pressure threshold): Strength calibration × conversion factor; First establish σ t The 2D polynomial / spline of (T,φ), multiplied by κ (obtained from pressure differential bubbling / process calibration): p max =κσ t At runtime, for (T) k ,φ0) direct evaluation.

[0169] p bulk (Cavity pressure): Actual pressure sensor measurement; installed on the outside of the first opening, in the air passage between valve body 20 and fan, the real-time value or nominal set value can be used for control.

[0170] (Weight loss rate): Differential of the TGA curve.

[0171] For m b (T) First perform Savitzky-Golay smoothing, then take the derivative of the central difference to obtain each T. k The value of .

[0172] Y v (Volatile mass coefficient): Corrected for TGA carbon residue.

[0173] Take 1 for no residual carbon; the difference in residue between the two TGAs (inert / oxidized) gives the carbon retention ratio χ. char Take Y v =1-χ char .

[0174] 2) Upper limit of the heating slope when the temperature difference does not exceed the limit formula:

[0175] The source and acquisition of each quantity in this formula: h (conservative lower limit of convective heat transfer coefficient): Under the condition of valve body small opening of 20° and no fan disturbance, a small power step is made, and h is identified according to the first-order thermal inertia. The smallest value is taken as h. min A: Same as above. p (Total mass of green bodies): Weighed using an electronic balance; summation of multiple items. C p (T k (Specific heat of chloroplasts): DSC calibration or mixed method; during operation, T k Look up the table / substitute into the fitted formula. ΔT allow (Maximum permissible temperature difference): Thermal shock test or thermal stress estimation.

[0176] Experiment: Measuring the critical temperature difference ΔT c Take ΔT allow =ηΔT c (η = 0.5~0.7); Estimation:

[0177] Upper limit of hardware temperature rise slope r max Definition: Under the given equipment and loading conditions, the maximum sustainable rate of temperature rise that allows for long-term stable operation without overheating or overflow. Acquisition: No-load / half-load step test (S100-12): Set the neutral clearance and maximum safety threshold T. w Scan command slope r, record controlled region T p The initial slope and steady-state tracking quality of (t). Take the maximum r that will not cause power saturation, overcurrent, overtemperature, or significant overshoot, and multiply it by a safety factor of 0.7–0.85 to obtain r. max If the unit is K min -1 In S200-7, first convert it to K s -1:

[0178] Among them, T k Discrete temperature nodes; T is represented by K. ρ v μ: Density / viscosity of the gas mixture; calculated from composition and temperature or looked up in a table. k perm : Darcy permeability of chloroplasts (m 2 A,L,m p C p : Effective heat exchange area, thickness, total mass, and specific heat of the green body. max : Maximum permissible equivalent pore pressure in the pore channel; p bulk : Cavity pressure. |dm b / dT|: Rate of weight loss per unit temperature; Y v : Volatile matter mass coefficient. h min : Conservative lower limit of near-wall convective heat transfer coefficient; ΔT allow : Maximum allowable temperature difference. max : Upper limit of sustainable heating rate for hardware / system. κ: Conversion factor from strength to pore pressure (conservative value calibrated by foaming / process).

[0179] Step S280: Identification of insulation zone and insulation strategy.

[0180] Obtain from the preceding steps Threshold θ, ε r ,τ max Identification: If Marked as "Candidate Insulation Point". Upon reaching this point, the system enters the constant temperature zone. Until any termination condition is met:

[0181] m b (T)≤ε r m b0 or t hold ≥τ max ;

[0182] Note: m b (T) Directly retrieve the TGA curve from step S100; t hold The accumulated insulation time is an internal timer within the controller, starting from the moment the insulation process begins and accumulating only during the insulation phase. It requires no external instruments and is a "state variable" of the control algorithm, accumulating in cyclic steps within the insulation phase. τ max : The time constant derived from the isothermal TGA performed at the main peak temperature (taken as 3τ).

[0183] Step S290 generates the time trajectory T ref (t).

[0184] Initialization: t←0, T←T minLoop (each step Δt): If T falls within the candidate insulation zone and the termination condition is not met, then T... ref (t+Δt)=T, t←t+Δt. Otherwise, proceed according to the comprehensive slope (take the node where the current temperature is located). ): Update T←T ref (t), until T≥T max Further, smoothing (optional): for the obtained T ref (t) Perform a first-order lag filter, with a time constant τ f =5~10Δt, to ensure that the slope of the set curve does not fluctuate too quickly, which is conducive to the execution of step S300.

[0185] Specifically, as described below: all parameters have been obtained or set in the preceding steps. Temperature range upper and lower limits T min ,T max Time step Δt; Temperature grid {T} k} and node slope table Thermal insulation criteria: Threshold θ, residual adhesive ratio ε r Maximum heat preservation time τ max TGA curve m b (T).

[0186] Step S291: Preprocessing and interpolator construction.

[0187] for |dm b / dT|(T),m b (T) Construct a one-dimensional piecewise linear interpolator (or spline): Interp m′ (T),

[0188] Define candidate functions for insulation (considering only temperature):

[0189] Note: To suppress jitter, a hysteresis of 1–2K can be used for H(T): the threshold is θ. ↑ =θ, exit threshold is θ ↓ =0.8θ.

[0190] Step S292: Time step consistency check.

[0191] make Require

[0192] This is to avoid skipping multiple temperature nodes in a single step (1.25 is a safety factor). If this condition is not met, Δt will be automatically reduced.

[0193] Step S293: The main loop generates T ref (t).

[0194] Initialization: t←0, T←T min , t hold ←0. Loop: Repeat when T<T max : Determine whether to keep warm:

[0195] If H(T) = 1 and the heat preservation termination condition is not met (see below), then: T ref (t + Δt) = T, t hold ←t hold +Δt;

[0196] Otherwise, enter the temperature increase promotion.

[0197] Temperature increase promotion: Obtain the current target slope: Calculate: Reset the heat preservation duration count t hold ←0. Iterative update: t←t + Δt, T←T ref (t); Upper limit clipping and monotonicity protection: If T>T max , then set the T of this step ref (t) to T max and end the loop; If numerical jitter causes T ref (t + Δt) < T, force T ref (t + Δt) = T (ensure monotonic non-decrease). Heat preservation termination condition (either of the two): Residual glue criterion: Duration upper limit: t hold ≥τ max .

[0198] Step S294: Curve smoothing.

[0199] Perform first-order lag filtering on the obtained discrete sequence T ref [n]:

[0200]

[0201] where: τ f = 5 - 10Δt. If adopted, this smoothed sequence is used as the final T ref (t).

[0202] Above: T min , T max Temperature zone upper and lower limits; Δt control period; Comprehensive target slope; |dm b / dT|(T) weight loss rate; θ weight loss peak threshold; ε r Residual glue allowable ratio; τ max Heat preservation duration upper limit; thold Cumulative heat preservation time; m b0 Initial glue content; T ref (t) Target temperature curve.

[0203] Step S2100: Output and consistency verification.

[0204] Output: Store T ref (t); Synchronously generate node table For use by the controller to limit the slope. Verification: Rapid simulation of pore pressure and temperature margin: Estimation of generated flow rate. in Estimate discharge flow rate verify (Equivalent to satisfying the slope limit of S200-5), and |T w -T p | Corresponding temperature difference constraint (from step S260). If not met, increase the insulation or decrease ΔT. allow Regenerate.

[0205] Specifically:

[0206] Get the generated T ref (t); Node table And the physical properties and constraints in S100 / S200: m b0 、|dm b / dT|(T),Y v ρ v (T), μ(T), k perm (φ0), A,L,p max (T,φ0), p bulk m p C p (T), ΔT allow ε,σ,F(r) (If it is necessary to estimate T) w ).

[0207] Step 2110: Output storage.

[0208] Time track files / tables: storage (t i ,T ref (t i (CSV / parameter block). Limit slope node table: stores... Used for online slope limiting during control execution. Insulation section marking: for each (t) i ) or (T i Store the Boolean bit isHold[i] to facilitate the execution of the state machine by the S300.

[0209] Step 2120: Rapid pore pressure and temperature margin verification (time-by-time).

[0210] For each time point t i (or temperature T) i =T ref (t i ))implement:

[0211] Generate flow rate estimates:

[0212]

[0213] Discharge rate estimation (Darcy):

[0214]

[0215] Pore ​​pressure safety ratio:

[0216]

[0217] Require (It is better to have a 10-20% safety margin).

[0218] Temperature difference / thermal inertia check (rapid):

[0219] Define the allowable slope of thermal inertia:

[0220]

[0221] examine:

[0222]

[0223] Furthermore, the surface temperature of the optional heating element 410 is accessible.

[0224] During the planning interval r plan Estimate the required T w (Ignore / simplify convection):

[0225]

[0226] Require (Heating element 410 limit), otherwise it is judged as unreachable.

[0227] Step S2130: Automatic rollback and regeneration strategy when failure occurs.

[0228] If any of the following conditions are triggered:

[0229] or or (Optional check);

[0230] Then, rollback will be executed according to priority, and step S290 will be re-executed:

[0231] Expand the insulation zone: Multiply θ by 0.9 (makes it easier to enter the insulation zone), or... Insert a constant temperature platform (3-5K wide) directly near the peak value.

[0232] Reduce allowable temperature difference: ΔT allow ←0.9ΔT allow Globally slowing slope: (For all T or only for the out-of-range).

[0233] (Optional) Increase the planning interval r plan To increase the effective radiative coupling of F(r) and reduce the required T w .

[0234] Repeat step S2120 for verification until all passes.

[0235] Step S2140: Final freezing and distribution.

[0236] Frozen version number (including: θ, ε) r ,τ max ,ΔT allow ,r max ,r plan ,κ,K c Versions of key parameters such as ε(T), F(r).

[0237] Issued to: T ref (t) Timetable; Node slope limit table; insulation section marker isHold(t).

[0238] The meanings of the parameters involved in step S200 are as follows:

[0239] The instantaneous slope of the actual curve (obtained by difference); Y v Volatile matter mass coefficient; ρ v μ gas density / viscosity; k perm (φ0) Green body permeability; A,L Effective heat exchange area / thickness; p max ,p bulk Permissible pore pressure threshold / cavity pressure; Pore ​​pressure safety ratio; h min Lower limit of near-wall convection coefficient; m p C p Green body mass / specific heat; ΔT allow Allowable temperature difference; Temperature difference safety ratio; ε,σ emissivity / Stivan-Boltzmann constant; F(r) plan Apparent factor; ΔH enthalpy of exfoliation reaction; The maximum allowable temperature for heating element 410. It should be noted that the enthalpy of the debinding reaction ΔH can be accurately measured using DSC / DSC-TGA / DSC-EGA, as detailed below:

[0240] Differential scanning calorimetry (DSC / modulated DSC) combined with TGA / gas analysis yielded the results. The objective was to separate the thermal effects "related to binder decomposition / combustion" from irrelevant thermal effects such as "solvent evaporation, matrix phase change, and powder reaction," thus obtaining ΔH normalized to the binder mass.

[0241] 1. Sample and baseline

[0242] 1.1 Three groups of samples (with the same heating rate β, usually 2–10℃ / min, to ensure peak shape stability).

[0243] Pure binder sample (m_b^A≈5–15mg) is derived from the proportionate mixing of the formulation components and solvent removal; unbound preform / powder sample (m_p^B≈5–15mg) is the same powder or preform after sufficient binder removal; binder-containing preform sample (m_g^C≈10–30mg) has binder mass m_b^C determined by the total weight loss of TGA or known formulation; all samples are prepared with empty crucible baselines (crucibles of the same material with identical lids).

[0244] 1.2 Atmosphere: Matched to the actual process (inert N2 / Ar or oxidizing Air / oxygen-controlled O2%), flow rate 40–60 mL / min; oxidizing conditions result in combustion exothermic effects, while inert conditions result in pyrolysis / carbonization heat effects.

[0245] 1.3 Calibration: Perform temperature and heat flow calibration using indium / zinc; perform a blank baseline for subsequent baseline subtraction.

[0246] 2. Synchronize TGA / EGA (MS or FTIR, not required but strongly recommended).

[0247] 2.1 TGA provides the mass loss curve m(T) to determine the burnout temperature range of the binder and m_b;

[0248] 2.2 MS / FTIR is used to identify the CO2 / H2O / organic pyrolysis product release regions, aligned with the heat flow peaks, to avoid miscalculating powder phase transition or sintering exothermics as debinding heat.

[0249] 3. Data processing and integration (normalized according to adhesive quality).

[0250] 3.1 Pure binder ΔH:

[0251]

[0252] T1–T2 are the main peaks of the depolymerization process (confirmed by TGA / EGA).

[0253] 3.2 Matrix correction (to avoid incorporating intrinsic thermal effects of the powder):

[0254]

[0255] It is usually close to 0, but if the powder has phase change / adsorption exothermic effect, it needs to be deducted.

[0256] 3.3 Equivalent ΔH of the preform containing rubber:

[0257]

[0258] here

[0259] 3.4 Result selection: If the difference between ΔH_neat and ΔH_green is <10–15%, take ΔH = average value; if the difference is larger, take ΔH_green first (it includes the actual coupling effect such as pore diffusion / local oxygen supply, which is more in line with the engineering "apparent enthalpy").

[0260] 3.5 Modulated DSC (MDSC, optional): Use "inverted / non-inverted" to separate the reversible heat capacity term from the kinetic exothermic / endothermic term, further purifying the integration region (eliminating the influence of Cp·β).

[0261] 4. Uncertainty and Safety Factor

[0262] Repeat n≥3 times and give the mean ±95% CI; take the upper limit of |ΔH| ×1.2 in the process feedforward (take a larger value for exothermic and a smaller value for endothermic) to prevent overshoot caused by underestimating exothermic.

[0263] Furthermore, it should be noted that the glue removal process variable ξ(t)∈[0,1], its rate

[0264] Thermal balance feedforward additions: Where A safe ∈[1.1,1.3] represents the safe amount of heat release (>1 for exothermic and <1 for endothermic).

[0265] Will As a feedforward compensation for the inner loop power setting, it is superimposed on PI; if ΔH is undetermined, it can be initially set to 0 and the actual power difference curve can be observed to deduce ΔH for secondary calibration.

[0266] In summary: θ: the differential noise σ of S210 m′ The 3σ threshold is given (obtained directly from statistics). ε r The performance of isothermal TGA and subsequent sintering was compared in a small-scale test to determine the optimal residual binder ratio; the maximum residual binder ratio that meets the specified parameters was taken. max The isothermal TGA fitting time constant τ is taken as 3τ.v TGA (Inert vs. Oxidative) method for obtaining carbon residue ratio χ char Y v =1-χ char When there is no residual carbon, Y v =1. h min The minimum value is determined by either step identification (experiment) or the Nusselt formula (calculation). perm (φ): Darcy steady-state permeability experiment regression (step S100). p max (T,φ): High-temperature bending strength calibration and κ (bubbling / process calibration) (step S100). ε(T), F(r): Thermocouple-infrared comparison (emissivity) and steady-state radiation verification (apparent factor) (step S100). T min ,T max : Upper and lower limits of the glue discharge temperature zone; Δt, ΔT: Time / temperature step size. m b (T): Residual organic matter mass as a function of temperature; Rate of weightlessness. σ m′ ε: Standard deviation of numerical differential noise; θ: Threshold for weightlessness peak discrimination. r Residual glue ratio; τ max Maximum heat preservation time at a single point. μ(T),ρ v (T): Viscosity and density of the gas mixture; p bulk : Cavity pressure. k perm (φ0): Gas permeability of the chloroplast at porosity φ0; p max (T,φ0): Permissible pore pressure threshold curve. A,L,m p C p (T): Effective heat exchange area, thickness, mass, and specific heat of the green body. Y v : The mass coefficient of volatile matter generated per unit of excreted adhesive. Two types of upper limits for slope; Overall target slope. T ref (t): Target temperature curve; t hold Cumulative heat preservation time.

[0267] Step S300: Control the operation of the regulating mechanism 50 and the heating mechanism 40 according to the target temperature curve (can be executed directly).

[0268] Step S310: Input and basic settings (once before power-on / batch).

[0269] Target temperature profile: (Output from S200); Node slope limit table: Sensor grouping: Thermocouples are grouped by plane. Let the first... The sensor set for each circumferential partition is as follows .

[0270] Execution quantity: Heating mechanism 40: Surface temperature / power setting T for each or each group of heating tubes 410 w,z (or power P) z Adjustment mechanism 50: Heating tube 410 of each zone—distance r between the material rack z (Given by the stroke of the moving part 550 and the angular displacement of the drive plate 530).

[0271] Limits and Calibration: Distance from Boundary: r z,min ≤r z ≤r z,max Maximum displacement / angular velocity Heating element 410 temperature / power limit: or Allowable temperature difference: ΔT allow (From step S100); Apparent factor curve F z (r z ), emissivity ε(T), m p C p (T), A, |dm b / dT|(T), ΔH. Control period: Δt ctrl = 0.5~1s (1s recommended).

[0272] Step S320: Sampling and zone temperature calculation (per Δt) ctrl ).

[0273] Collect the original temperature of each thermocouple First-order low-pass filter:

[0274] Regional uniformity index: use: Used to track T ref ;ΔT z Used for geometric compensation (adjusting r) z ).

[0275] Step S330: Control the state machine (press T) ref (t) paragraph).

[0276] Heating phase: The goal is to "track the slope under the premise of uniform temperature". Insulation section: The goal is to "converge and maintain the temperature." Cooling phase (if included in the process): Same as heating phase, except... (The core of this step remains unchanged.)

[0277] Step S340: Dual closed-loop control structure (power / temperature inner loop combined with geometric outer loop).

[0278] Step S341: Error definition (for each partition z).

[0279] set up: (All partitions share the same objective; if necessary, a small weight bias can be added to the outer layer to achieve temperature equalization priority).

[0280] Tracking error: Temperature uniformity error: in (Full circumferential average).

[0281] Step S342: Inner ring: Heating setting T w,z / Power P z .

[0282] Feedforward (based on thermal balance, suppressing hysteresis):

[0283] For each partition, according to the current r z Calculate the feedforward heating temperature:

[0284]

[0285] Among them, A z The effective heat exchange area for each zone (can be A / Z or geometrically allocated).

[0286] Feedback (PI control): Temperature setting: Or power type:

[0287] Tuning Recommendation: Heating Stage K p Take a smaller value, allowing for a slight under-tracking; increase K in the insulation section. p ,K i Reduce steady-state error.

[0288] Anti-integral saturation: If the sum of squares (sat) reaches its peak or bottom, then the integration is paused (anti-windup).

[0289] Step S343: Outer ring: Geometric compensation (adjustment mechanism 50 setting r) z ).

[0290] Objective: To reduce the temperature difference ΔT between temperature zones. z This involves "moving" heat from hotter zones to colder zones (by changing the visual factor).

[0291]

[0292] Dead zone δ T =1~2℃ (to avoid vibration); outer loop period Δt geom = 3-5s (slower than the inner loop).

[0293] Directionality: If the zone is cold Let r z ↓(Closer to the material rack, increase F) z ); if it is too hot, then r z ↑.

[0294] Execution order: first r z The instruction (see S300-4) states that after the position is in place or the timeout period has elapsed, the latest r will be used as the basis for further action. z Calculate inner loop feedforward and update T w,z .

[0295] Step S344: Adjust the timing of the mechanism 50 (one "insert\adjust\exit" cycle). The moving part 550 is pushed forward, causing the drive shaft 542 to be in the slot → the drive plate 530 to move radially due to angular displacement → decoupling is disengaged.

[0296] For partition z that needs adjustment:

[0297] Insertion phase: The lifting rod 551 advances along the first direction to the "approach position" (stroke s). pre ); micro-advance until the clamp 552 contacts the annular protrusion 5421 of the drive shaft 542 (based on force / current rise or proximity switch criteria); continue advancing Δs to insert the end of the drive shaft 542 into the drive groove 531 (based on position switch / encoder criteria).

[0298] Adjustment phase: Issue 530° angular displacement Δψ to the drive board z (Converted according to the calibrated ψ→Δr relationship) ); angular velocity limiting Stop when the corner position is in place (encoder) or when the timeout occurs.

[0299] Exit Phase: The lifting rod 551 retracts to the safe position (ensuring that the drive shaft 542 exits the slot and is limited by the guide plate 520); the "mechanical coupling" interlock of this zone is cleared, and the independent operation of the temperature control inner loop is restored. Arrival / Timeout Criterion: Any phase exceeding the set timeout t... stage,max If the machine stops, rolls back, and issues an alarm (e.g., 3–5 seconds), it will immediately stop, roll back, and issue an alarm.

[0300] Step S350: Paragraph Strategy. Warming-up Paragraph Inner loop: Prioritizing the coldest area, it allows for slightly underpowered hot areas to prevent overshoot. This can be... Set as Outer loop: Perform small-step geometric compensation every 3–5 seconds until max. z ΔT z ≤ΔT 均温阈 (e.g., 3℃). Insulation section Inner ring: Increase K p ,Ki Rapidly eliminates steady-state errors; outer loop: more aggressive temperature equalization, target maximum. z ΔT z ≤1~2℃. Cooling section: If the fan is disabled: the strategy is the same as for heating, but the slope is negative; if the fan is allowed: use the fan / valve body 20 as an external convection source, adjust according to the differential pressure closed loop, and maintain the geometry at a neutral or slightly large r. z (To prevent hot spot backflow).

[0301] Step S360: Limits, Interlocks, and Safety Cutting (checked every cycle). Temperature Difference Limit: If any cross-section / zone ΔT z >ΔT allow Then: Freeze T immediately w,z The uplink increment is prioritized for execution r. z ↑ (Opening the gap) and local power reduction; slope limiting: actual slope Mandatory: Heating element 410 upper limit: If First r z ↓ Check again to see if the limit is still exceeded; if it is, reduce the overall slope and trigger an "insufficient power" alarm. Mechanical limit: r z Touch r z,min / max After marking the outer ring, only reverse adjustment is allowed; multiple touches trigger an "insufficient geometric travel" alarm. Sensor malfunction: If any zone thermocouple is disconnected / over-range, then "conservative control" (limiting T) applies to that zone. w,z Rise, r z ↑), and request maintenance at the same time. Valve body 20 / fan interlock (optional): If the cavity pressure is abnormal or the exhaust gas rises abnormally (external conditions), suspend the geometric approach operation and reduce the slope.

[0302] Step S370: Fault rollback and resynchronization. Position asynchrony: If the insertion / exit criterion fails, execute a "return to origin" procedure once (lifting rod 551 returns to zero, drive board 530 returns to zero), and try again after recovery; if it fails twice in a row, manual intervention is required. Temperature control saturation: If T remains saturated for N consecutive cycles (e.g., 10 seconds)... w,z In the upper limit and e z <0 (temperature is still low), automatically adjust the global slope. And record it. Excessive temperature deviation: If |e z |>ΔT allow If the temperature remains above 5 seconds, "over-tolerance protection" is triggered: the freeze slope = 0 (temporary insulation), only geometric temperature equalization is performed, and heating resumes after recovery. Where: T ref (t): Target temperature curve; Upper limit of target slope (node ​​table interpolation). The filtering temperature of the i-th thermocouple; Zone average / minimum / maximum temperature; ΔTz Temperature difference between zones. z : Partition tracking error; Temperature average error (relative to global average); ΔT allow Allowable temperature difference. (T) w,z / P z : Surface temperature / power setting for zoned heating element 410; r z : Spacing between zones; F z (r): Apparent factor curve. K p ,K i ,K r Inner-loop PI and outer-loop geometric gain; δ T Outer ring dead zone; Institutional rate limit. A z : Effective heat transfer area of ​​the partition; ε(T), σ: Emissivity and S–B constant; |dm b / dT|(T), ΔH: TGA weightlessness derivative and reaction enthalpy. "Insert, Adjust and Exit": a three-stage action sequence of moving part 550 advancing, driving plate 530 angular positioning, and moving part 550 exiting.

[0303] In summary, the S300 employs an inner-loop power tracking T... ref With the geometric mean temperature of the outer ring r z The system employs a dual closed-loop design: the inner loop uses feedforward + PI acceleration, while the outer loop uses slow-cycle, small-step adjustment to suppress circumferential temperature difference; all actions are performed under interlocked protection against temperature difference / slope / mechanical factors and upper temperature limits. Following this procedure, engineers can directly configure parameters, conduct online commissioning, and ensure stable operation of the sintering furnace.

Claims

1. A fully automated high-temperature hot sink sintering furnace based on hydrogen-powered SOFC for sintering metal slurry ceramic plates, characterized in that, include: The furnace body includes a sintering cavity and a first opening and a second opening communicating with the sintering cavity; A valve body that is located at the first opening and is controlled to be opened and closed; A loading platform for placing and controlling the movement of a rack that supports ceramic plates, driving the rack into or out of the sintering chamber through the second opening, wherein the direction of movement of the loading platform is defined as a first direction. The heating mechanism is installed in the sintering chamber. The heating mechanism includes a plurality of heating tubes with controlled temperature rise. Each heating tube is configured to surround the material rack after the material rack enters the sintering chamber, and the extension direction of the heating tube is parallel to the first direction. The adjustment mechanism includes an adjustment end that is controlled to move and is drivenly connected to the heating tube to drive the heating tube to move toward or away from the material rack; Several temperature sensors are disposed around the furnace body and have detection ends that extend into the sintering chamber. Each detection end is arranged around the material rack after the material rack enters the sintering chamber to obtain the ambient temperature detection value. The controller is connected to the heating mechanism, the regulating mechanism, and the temperature sensor, and is configured to control the operation of the regulating mechanism and the valve body based on the temperature detection value, so as to adjust the relative position of the heating tube and the material rack and the on / off state of the valve body.

2. The fully automatic high-temperature hot sink sintering furnace according to claim 1, characterized in that, The number of adjustment ends is several, and each one corresponds to a heating tube.

3. The fully automatic high-temperature hot sink sintering furnace according to claim 2, characterized in that, The material rack includes several material placement plates arranged perpendicular to the first direction and arranged sequentially along the first direction; each of the temperature sensors is divided into several groups and arranged parallel to each other. All the temperature sensors in each group are located in the same plane perpendicular to the first direction. The temperature sensors in each group are configured to be distributed around the material rack after the material rack enters the sintering chamber, and the spacing between two adjacent temperature sensors in each group is equal.

4. The fully automatic high-temperature hot sink sintering furnace according to claim 1, 2, or 3, characterized in that: The first opening and the second opening are respectively located on opposite sides of the inner wall of the sintering cavity, the center lines of the first opening and the second opening are collinear, and the center line of the first opening is parallel to the first direction; The temperature sensor is a thermocouple sensor, and the extension direction of the thermocouple sensor is perpendicular to the first direction.

5. The fully automatic high-temperature hot sink sintering furnace according to claim 4, characterized in that, The adjustment mechanism also includes: A central tube is disposed on the inner wall of the sintering cavity on the side where the first opening is located. The axis of the central tube is coaxial with the center line of the first opening. One end of the central tube is connected to the first opening, and the other end of the central tube is connected to the sintering cavity. A guide plate is fixedly sleeved outside the central tube. Several sets of first guide grooves are equally spaced in a circular array with the center of the guide plate as the center. The extension direction of the first guide grooves is configured to point towards the central tube or away from the central tube. A drive plate is rotatably sleeved outside the central tube and rotates under control. The drive plate has several sets of drive grooves arranged in a circular array at equal intervals with its own center as the center. The number of drive grooves is the same as the number of the first guide grooves and they correspond one-to-one. Furthermore, the axis of the drive plate is collinear with the axis of the guide plate. A plurality of adjustment frames are provided, each of which corresponds one-to-one with each group of the first guide grooves and each group of the drive grooves. Each adjustment frame includes at least one frame body and a drive shaft disposed at one end of the frame body. The number of all the frame bodies is the same as the number of the heating tubes and each of them carries the corresponding heating tubes. The end of the drive shaft away from the frame body passes through the first guide groove and is aligned with the drive groove, and the end of the drive shaft away from the frame body is the adjustment end. Several movable components are movably connected to each of the drive shafts and controlled to move along the first direction, so as to drive the adjustment frame to move synchronously along the first direction when moving along the first direction; When the adjustment mechanism is in operation, the moving member drives the adjustment frame to move toward the direction of the first opening so that the drive shaft is inserted into the drive groove. Then, when the drive plate is rotated in a controlled manner, the inner wall of the drive groove abuts against the drive shaft, so as to force the drive shaft to move along the extension direction of the first guide groove.

6. The fully automated high-temperature hot-sink sintering furnace based on hydrogen-powered SOFC as described in claim 5, characterized in that: The movable component includes: A lifting rod that moves in a controlled manner along a first direction, wherein the axis of the lifting rod is parallel to the first direction; A clamping member is disposed at one end of the lifting rod and located between the drive plate and the guide plate. The clamping member has a through second guide groove, and the extension direction of the second guide groove is consistent with that of the first guide groove. The drive shaft has an annular protrusion on its circumference, located between the clamping member and the drive plate; When the adjustment mechanism is in operation, the clamping member moves along the first direction with the lifting rod and abuts against the annular protrusion, thereby driving the adjustment frame to move along the first direction and inserting the drive shaft into the drive groove.

7. A fully automated high-temperature hot-sink sintering furnace based on hydrogen-powered SOFC as described in claim 5 or 6, characterized in that: The adjustment frame also includes a limiting end disposed on the side of the frame body near the second opening; The sintering furnace also includes: A limiting member is provided on the inner wall of the sintering chamber, which has a second opening side and corresponds to each of the adjustment frames. The limiting member is provided with a third guide groove, and the extension direction of the third guide groove is consistent with the extension direction of the first guide groove. When the adjusting frame moves along the extension direction of the first guide groove, the limiting end is inserted into the third guide groove, so that the side of the adjusting frame near the second opening can only move along the extension direction of the first guide groove.

8. The fully automated high-temperature hot-sink sintering furnace based on hydrogen-powered SOFC as described in claim 7, characterized in that, It also includes a ventilation fan, which is installed on the air passage outside the first opening and located outside the valve body. The ventilation fan is connected to the controller and is controlled to operate.

9. A control method for the fully automatic high-temperature hot sink sintering furnace as described in claim 8, characterized in that: The metal paste ceramic plate includes a green body, which refers to a green body formed by combining ceramic powder of the ceramic plate under the action of organic binder and / or solvent contained in the metal paste and without undergoing sintering densification treatment. The sintering furnace also includes: A gas pressure sensor is disposed on the gas path outside the first opening and located between the valve body and the ventilation fan to obtain the gas pressure in the sintering chamber. A gas composition sensor is disposed on the gas path outside the first opening to obtain the gas composition in the sintering chamber; The control method includes: The thickness of the green body, the effective heat exchange area of ​​the green body, the mass of the green body, the specific heat of the green body, the initial gum content of the green body, the initial porosity of the green body, the calibration relationship of the pore permeability of the green body, the calibration relationship of the allowable pore pressure threshold of the green body, the calibration curve of the gas viscosity and the gas density in the sintering cavity, the atmosphere pressure in the sintering cavity, the maximum allowable temperature difference, the upper limit of the hardware heating slope, the surface emissivity of the heating tube, the Stefan-Boltzmann constant, the apparent factor-distance calibration curve, and the thermogravimetric curve are obtained. The target temperature curve is determined based on the thickness of the green body, the effective heat exchange area of ​​the green body, the mass of the green body, the specific heat of the green body, the initial gum content of the green body, the initial porosity of the green body, the calibration relationship of the pore permeability of the green body, the calibration relationship of the allowable pore pressure threshold of the green body, the temperature calibration curve of the gas viscosity and the temperature calibration curve of the gas density in the sintering cavity, the atmosphere pressure in the sintering cavity, the maximum allowable temperature difference, the upper limit of the hardware heating slope, the surface emissivity of the heating tube, the Stefan-Boltzmann constant, the apparent factor-distance calibration curve, and the thermogravimetric curve. The regulating mechanism and the heating mechanism are controlled according to the target temperature curve and each of the temperature sensors.

10. The control method according to claim 9, characterized in that, The steps for determining the target temperature permission include: Based on the pore permeability calibration relationship of green bodies, the allowable pore pressure threshold calibration relationship, the temperature calibration curves of gas viscosity and gas density in the sintering chamber, and the atmosphere pressure, the upper limit of the temperature change rate to avoid exceeding the threshold is determined at different temperatures. Combined with the maximum allowable temperature difference, the upper limit of the hardware heating slope, the surface emissivity of the heating tube, the Stefan-Boltzmann constant, and the apparent factor-distance calibration curve, the upper limit of the temperature change rate limited by temperature difference and hardware capability is determined at different temperatures. The smaller of the two upper limits is used as the limiting condition for the corresponding temperature. The temperature-time relationship is generated and its consistency is checked using the thermogravimetric curve to determine the target temperature curve.

Images