Continuous Furnace
A continuous furnace with precise temperature control and an artificial neural network system addresses the inefficiencies in carbon fiber oxidation, enhancing production speed and safety by maintaining controlled oxidation.
Patent Information
- Application Number
- JP2024529572
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2021-11-29
- Filing Date
- 2022-11-28
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The carbon fiber production process, particularly the oxidation step from polyacrylonitrile (PAN) precursor fibers, is time-consuming and energy-intensive due to its exothermic nature, leading to potential combustion and uncontrollable emissions of toxic gases.
A continuous furnace with adjustable radiant heaters, gas flow perpendicular to the processing direction, and a control system using optical detection and a self-learning artificial neural network to maintain a precise temperature difference between the fiber and gas temperatures, ensuring controlled oxidation.
This approach reduces energy consumption and reaction time, minimizing the risk of combustion and emissions while enabling faster and more controlled oxidation of precursor fibers.
Smart Images

Figure 2024542477000001_ABST
Abstract
Description
[Technical field]
[0001] This application relates to a furnace and method for oxidizing polyacrylonitrile-based precursor fibers in a carbon fiber manufacturing process.
[0002] Carbon fiber has become an important material in the past 150 years. Its first industrial use is believed to have been in the late 19th century as incandescent wire in carbon filament light bulbs. However, in the late 20th century, the mechanical properties of high performance carbon fiber have led to its being a tough, lightweight material. Typically, carbon fiber is used in composite materials embedded in a thermoplastic or thermosetting resin matrix, often known as carbon fiber reinforced plastics, for example to fabricate airplane fuselages or wings. As an example, the state-of-the-art passenger aircraft, the Boeing 787 and the Airbus A350, are both made to a significant extent from carbon fiber reinforced plastics. Also, sporting equipment such as skis, ski poles, rackets for various sports such as hockey and tennis, and even entire hulls of boats are made from carbon fiber reinforced plastics.
[0003] Furthermore, carbon fibers are used as reinforcing materials for materials other than resins, such as concrete.
[0004] Generally, carbon fibers are produced from organic fibers containing carbon in a thermal process. For carbon filament incandescent lamps, plant-based cellulose fibers, for example from bamboo or cotton, are carbonized, but the most common carbon fiber precursor these days is polyacrylonitrile, a typically petroleum-based polymer. Other possible precursors are cellulose fibers, for example from the viscose or lyocell process, or rayon-based fibers obtained from pitch.
[0005] From polyacrylonitrile (PAN), fibers are typically spun from solution by precipitating the polymer and dried. The dried fibers form the starting point for a very complicated manufacturing process that consumes both a great deal of time and energy.
[0006] The method of the present invention offers the possibility of reducing energy consumption and thus improving the carbon dioxide footprint of carbon fibers. The method of producing carbon fibers from PAN precursor fibers generally consists of two steps: oxidation and carbonization. In the oxidation step, PAN undergoes a cyclization reaction to form chains of annealed nitrogen-containing hexagonal rings, and then in an oxidative dehydrogenation reaction to form chains of annealed pyridine rings, resulting in the formation of oxidized PAN fibers. The oxidized PAN fibers can be used as flame-retardant infusible materials themselves and can also be processed into carbon fibers. In the oxidation step from the fusible PAN, oxidized infusible PAN is obtained. Moreover, since a lot of chemical energy is released from the precursor fibers in the oxidation step, the subsequent carbonization step can be carried out without significant heat release and is therefore easy to control. While the processing of carbon fibers from oxidized PAN fibers can be carried out relatively quickly in a carbonization furnace at temperatures above 300°C up to 3000°C under inert gas, the oxidation step is a time-determining step since it is highly exothermic and therefore needs to be carried out under well-controlled conditions to ensure the prevention of runaway reactions that could lead to combustion and thus destruction of the fibers, explosions in the oxidation furnace, and uncontrolled release of highly toxic gases such as carbon monoxide and hydrogen cyanide.
[0007] Typically, the oxidation step is carried out in a furnace with several stages, each having a constant temperature, which increases from stage to stage during the initial cyclization and subsequent oxidative dehydrogenation. In the furnace stages, the temperature must be kept at a very constant level to ensure that the cyclization and oxidative dehydrogenation reactions occur, but keep the reaction rate at a level where only a limited amount of heat is released without causing the reaction to runaway. For these reasons, the reaction rate must be kept at a very low level, and the oxidation step is a time-determining step in carbon fiber production, taking 45 to 60 minutes to obtain oxidized, infusible PAN fibers from fusible PAN fibers.
[0008] Cellulose fibers as precursors are not meltable, unlike PAN fibers, but they are also flammable, so require thorough control of the reaction temperature. The same is true for pitch as a precursor for carbon fibers, although the oxidation of rayon- and pitch-based precursor fibers is less exothermic. For all these precursors, the reaction is partially or totally exothermic. In the exothermic part of the process, the temperature of the precursor fibers during the reaction is higher than the temperature in the furnace where the reaction takes place.
[0009] In this application, the terms "precursor fiber" and "precursor yarn" are used interchangeably. In general, both terms refer to any type of yarn or fiber that can be used to produce those carbon fibers by heat treatment. This explicitly includes PAN-based precursor fibers, pitch-based precursor fibers, and rayon-based precursor fibers. Within the scope of this application, rayon-based precursor fibers may contain lignin.
[0010] During the passage through the oven, at least one precursor fiber may undergo a chemical reaction, meaning that its chemical composition changes between when it enters and when it leaves the oven. The PAN precursor fiber or yarn, or any other precursor fiber or yarn, entering the oven according to the present application may exit the oven in an oxidized state. During the passage through the oven, the fiber or yarn may change its chemical composition, via intermediates known to those skilled in the art, from meltable PAN to oxidized non-meltable PAN, from combustible rayon-based precursor fiber to rayon-based oxidized fiber, or from pitch to oxidized pitch fiber. For the sake of simplicity of explanation only, the PAN precursor fiber or yarn, or any other precursor fiber or yarn, passing through the oven according to the present application will be referred to as "precursor yarn" or "precursor fiber" throughout this application, regardless of the extent to which the chemical composition has changed relative to the precursor fiber or yarn. However, after passing through the oven, the yarn or fiber will be understood to be an oxidized precursor yarn or fiber.
[0011] The objective of the present application is to alleviate the aforementioned drawbacks by reducing the time and energy consumption in carbon fiber production.
[0012] It has now surprisingly been found that these problems are solved by a continuous oven having a process direction and at least one oven module, the oven module having two openings arranged such that the at least one precursor fiber can pass through the oven module in the process direction, the oven module further comprising at least one blower suitable for forming a gas flow perpendicular to the process direction, at least one adjustable radiant heater, preferably a microwave heater, radiating perpendicular to the process direction for heating the at least one precursor fiber to a first temperature, at least one second heater, preferably an electrical resistance heater, for heating the gas in the oven to a second temperature, at least one optical detection means, preferably an infrared camera or an infrared spectrometer, and at least one control unit, preferably a computer, characterized in that the blower, the radiant heater, the second heater, the optical detection means, and the control unit serve as regulating means for the heat released by the at least one precursor fiber in order to keep the temperature difference between the first temperature and the second temperature below 10°C.
[0013] A furnace according to the present application is any device that includes at least one opening and is capable of heating an article to a temperature that is typically above 100° C., with virtually no upper limit. The heat source may be radiation, such as microwave or infrared, an electric resistance heater, or any type of combustion material, such as gas, oil, coal or wood. Similar devices are known to those skilled in the art under the terms "kiln" or "furnace". In this application, the term "furnace" is always used, but it is clarified that whenever the term "furnace" appears, the term "kiln" or "furnace" may also be used.
[0014] The transfer of heat from the heat source to the furnace can be performed by any technique known in the art, for example by placing devices such as electric resistance heaters in the furnace cavity, heating the furnace walls from inside and / or outside the furnace with electric resistance heaters, or by burning materials such as gas, oil, coal, or wood. Additionally, the furnace according to the invention may be heated using an energy carrier liquid such as thermal oil using lost heat from other processes, such as a carbonization furnace where the oxidized precursor fibers are processed into carbon fibers. Furnaces using this heating technique are known to those skilled in the art as muffle furnaces. Heat can also be transferred from the heat source to the furnace using a transport medium such as air, or any other gas or gas mixture, or a liquid such as thermal oil. The air, gas, or gas mixture may be moved by natural convection only or by forced convection by any suitable means known to those skilled in the art, such as at least one ventilator, at least one pump, and / or at least one turbine.
[0015] The furnace module according to the present application is a part of a furnace and can function as a furnace by itself, but it may also be combined with an unlimited number of other furnace modules to form a composite furnace. It is understood that in the present application, the term "furnace" can always mean a furnace including only one furnace module, or a furnace including two, three, four, five, or any other number of furnace modules. Thus, in the present application, the term "furnace" can always mean a furnace including one single furnace module, or a furnace including more than one furnace module. In an embodiment, a furnace may consist of one furnace module, which means that the furnace module is a furnace. The term "composite furnace" is used for furnaces including more than one module only when necessary. Thus, a furnace according to the present application always includes one furnace module, but the amount of furnace modules included in a furnace according to the present application is not limited.
[0016] The furnace modules included in one of the furnaces according to the present application may be of the same or different shapes.
[0017] To be suitable for connection with other furnace modules, the furnace module according to the present application may have at least one means for connection to other furnace modules, for example at least one opening having the same size as the opening of the other furnace module, and the furnace modules may be combined such that said openings connect the furnaces, for example as part of a connecting tunnel. In an embodiment, the furnace module may comprise more than one connection means, such that one furnace module may be connected to either two or more furnace modules, thus forming a furnace comprising a chain or even a network of furnace modules.
[0018] In one embodiment, the means for connecting may be an opening in the furnace module arranged so that at least one precursor fiber can pass through the furnace in the process direction. The openings can be interpreted as the inlet and outlet of the furnace module, respectively. If the furnace consists of only one furnace module, the inlet and outlet of the furnace module are the inlet and outlet of the entire furnace, respectively. If the furnace has more than one furnace module, the inlet of the furnace is the inlet of one furnace module, the outlet of the furnace is the outlet of the other furnace module, and the inlets and outlets of the other furnace modules are connected to each other.
[0019] In one embodiment, several furnace modules are connected to one furnace stage, and several furnace stages may be combined into one composite furnace.
[0020] When connected together, the connecting means of the furnace modules may be sealed to prevent heat from escaping the furnace.
[0021] In one embodiment, the furnace modules may have connection means for electrical connections, which may be plugs and / or sockets for one or more wires and / or cables forming power line connections between the furnace modules and / or data connections between the furnace modules.
[0022] In one embodiment, the furnace module may have the shape of a tube or tunnel segment, or a dome, with at least one opening for connection to other furnace modules.
[0023] Any furnace module according to the present application can be a furnace in itself, so that the furnace module includes all the components of the furnace according to the present application, such as at least one radiant heater, at least one second heater, at least one optical detection means, and at least one control unit, which in any furnace module act as regulating means for the heat emitted by at least one precursor fiber to keep the difference between the first temperature and the second temperature constant. The at least one radiant heater, at least one second heater, at least one optical detection means, and at least one control unit of any furnace module can operate separately from the heaters, optical detection means, and control units of other furnace modules in the same furnace, or the at least one heater, at least one optical detection means, and at least one control unit of a furnace module can be connected to the heaters, optical detection means, and / or control units of at least one, some, or all of the other furnace modules. In one embodiment, the composite furnace according to the present application further comprises a central control unit for the entire furnace, which may be independent of the control units of the furnace modules or may be connected to one, some or all of the control units of one, some or all of the furnace modules.
[0024] In one embodiment, the modules that make up the furnace are similar or identical so that the modules can be easily replaced in the event of a defect.
[0025] The furnace according to the present application is a continuous furnace, which means that the at least one precursor fiber in the furnace moves constantly or with interruptions while being heated. This means that the furnace has an elongated shape, which may be in the form of a tunnel, a tube, a slot or a duct through which the at least one precursor fiber passes. The direction through which the at least one precursor fiber passes is called the process direction. The process direction can be any direction. It can be horizontal, for example in a tunnel-shaped furnace, or vertical, for example in a duct-shaped furnace. The process direction can also be inclined or follow a curved path.
[0026] The conditions in the furnace, such as the second temperature, the gas circulation, or the partial pressure of a gas, such as nitrogen, oxygen, or carbon dioxide, may be constant or may vary along the process direction. In one embodiment, the conditions are constant in one furnace module and may vary between different modules. "Constant" according to the present application means that the conditions, such as temperature, gas flow, or gas partial pressure, do not vary by more than 5%.
[0027] The chemical reaction of the at least one precursor fiber during passage through the oven is exothermic, and thus the temperature of the at least one precursor fiber is greater than the second temperature, in which case the chemical reaction produces energy, while the heat of the oven only provides the activation energy for the reaction.
[0028] The second heater according to the present application can heat the gas flow in the furnace to a second temperature. In one embodiment, the second heater is an electric resistance heater and includes a heat exchanger for transferring heat to the gas flow in the furnace module. According to this embodiment, the furnace module according to the present application and the furnace having said furnace module can be considered as a hot air stove or a hot air stove module, respectively.
[0029] In one embodiment, the second heater is located outside the furnace module and heats the walls of the furnace module in the form of a muffle furnace. In such a case, the second heater may be an electric resistance heater or a combustion material such as wood, coal, oil, or gas. The furnace module according to this embodiment may be operated in the form of a muffle furnace. In such a case, the gas flow in the furnace is not heated directly, but by the walls of the furnace. And the main purpose of the gas flow is to distribute the heat evenly in the furnace.
[0030] In this application, the term "second temperature" is used interchangeably with the term "furnace temperature." Because the reaction is exothermic, the second temperature is the temperature of the air flow in the furnace that is lower than or equal to the first temperature.
[0031] The second temperature may vary within the furnace. In one embodiment, every furnace module has its own constant second temperature. In one embodiment, the second temperature varies throughout the furnace modules.
[0032] The number of precursor fibers passing through the furnace is not particularly limited and depends only on both the size of the furnace and the size of the precursor fibers.
[0033] The precursor fibers or bundles of precursor fibers may be passed through the furnace in parallel, in one embodiment, hundreds, or even thousands, of precursor fibers can be passed through the furnace in parallel.
[0034] A blower according to the present application is any device capable of generating an air or gas flow. In one embodiment, the blower is a fan, such as an axial fan, a centrifugal fan, or a cross-flow fan. Furthermore, the blower may be a fan utilizing the Coanda effect. The gas flow generated by the blower is directed. In one embodiment, the blower may include means for heating or cooling the air or gas flow generated by the blower.
[0035] The blower according to the present application generates a gas flow substantially perpendicular to the process direction.
[0036] As used herein, the term "substantially perpendicular to the process direction" means that the angle is between 80° and 100° with respect to the process direction.
[0037] A radiative heater according to the present application is any device that emits electromagnetic radiation suitable for heating an object. In one embodiment, the radiative heater may be a microwave heater. A microwave heater is a device such as a cavity magnetron that emits electromagnetic radiation with wavelengths between 30 cm and 1 mm that can excite molecules into rotationally excited states, thereby heating an object.
[0038] In one embodiment, the heater is an infrared source. Infrared radiation is electromagnetic radiation with wavelengths between 1 mm and 780 nm that can excite molecules into vibrationally excited states and thereby heat objects. Infrared sources can be, for example, Nernst lamps, glow bars, heat lamps such as quartz heat lamps, or carbon heaters.
[0039] In one embodiment, the heater is a laser that emits either infrared or visible light.
[0040] The heater is adapted and positioned to emit radiation substantially perpendicular to the process direction.
[0041] The heater is capable of heating the at least one precursor fiber to a first temperature.
[0042] As used herein, the term "first temperature" is used synonymously with the term "temperature of at least one precursor fiber."
[0043] The first temperature may vary as the at least one precursor fiber passes through the oven. If the at least one precursor fiber undergoes a chemical reaction during its passage through the oven, the first temperature at a location within the oven may depend on the chemical composition of the at least one precursor fiber at the location.
[0044] Using the radiant heater, the second heater, the blower, the optical detection means, and the control unit, it is possible to keep the difference between the first temperature and the second temperature substantially constant, meaning that the temperature difference between the first temperature and the second temperature is less than 20° C. In one embodiment, the difference between the two temperatures is less than or equal to 10° C., 7° C., 5° C., 3° C., or 2° C.
[0045] In one embodiment, the second temperature is always 20° C. or 10° C. lower than the decomposition temperature of the oxidized precursor yarn.
[0046] To keep the temperature difference substantially constant, only the first temperature, only the second temperature, or both temperatures may be varied. To keep the temperature difference substantially constant, the first and second temperatures may both be kept constant, or may be varied in a uniform temperature sequence in which the first and second temperatures vary relative to one another with a substantially constant temperature difference. Using such a temperature sequence, the first and second temperatures may be constantly increased, for example during a passage through a furnace.
[0047] An optical detection means according to the present application is a device capable of receiving, measuring and analyzing electromagnetic radiation, both in terms of its intensity and spectral distribution.
[0048] The optical detection means is sensitive in the infrared region of the electromagnetic spectrum (wavelengths 1 mm to 780 nm), the visible region of the electromagnetic spectrum (780 nm to 300 nm), or the ultraviolet region of the electromagnetic spectrum (300 nm to 100 nm), or a combination thereof. The optical detection means may have a radiation source that emits some type of radiation. The optical detection means can detect, for example, infrared lamps or Nernst lamps, glow bars, heat lamps such as quartz heat lamps or carbon heaters, light bulbs, light emitting diodes, fluorescent lamps, or any combination thereof, that emit visible or ultraviolet radiation. In contrast to radiative heaters, whose main purpose is to heat at least one precursor fiber, the radiation source included in the optical detection means should be able to emit radiation with a broad wavelength distribution in order to function as a light source for the spectrometer.
[0049] The optical detection means may be a pyrometer, a camera or a spectrometer. A pyrometer can be used to measure the temperature of a distant object by detecting its infrared radiation.
[0050] A camera can be used to capture an image of an object and measure properties such as wavelength and intensity of radiation emitted from the at least one precursor fiber in two dimensions, rather than just for a single spot, thereby providing, for example, a two-dimensional image of the color and / or temperature of the at least one precursor fiber.
[0051] A spectrometer is a device that can resolve electromagnetic radiation into the wavelengths contained therein and thus measure the spectral distribution of electromagnetic radiation. This allows the temperature of the at least one precursor fiber as well as its chemical composition to be determined. The spectrometer, which may be an infrared spectrometer, a visible spectrometer, and a UV spectrometer, can be used to monitor the chemical reaction undergone by the at least one precursor fiber in the furnace.
[0052] A control unit according to the present application may be any electronic device capable of receiving and evaluating measured data and regulating other devices, in one embodiment the control unit is a computer.
[0053] At least one control unit is connected to the at least one radiant heater, the at least one blower, and the at least one optical detection means, and evaluates data received therefrom and adjusts both the radiant heater and the blower based on the obtained data, which may be data regarding the first temperature and / or the second temperature, data regarding the chemical composition of the at least one precursor fiber, and thus data regarding the state of the chemical reaction undergone by the at least one precursor fiber, or any combination thereof.
[0054] The oven may have at least one guide means for the yarn. The guide means for the yarn according to the present application may be at least one pulley, at least one loop, at least one comb, at least one godet, or any combination thereof. The oven may have a plurality of guide means for the yarn so that several yarns, for example hundreds or thousands of yarns, can pass through the oven in parallel.
[0055] In one embodiment, the means for guiding the yarn is arranged at the boundary between the modules of the furnace.
[0056] The guide means for the yarn may be equipped with a force sensor capable of measuring the drawing force of the yarn being passed over the guide means for the yarn. Measurement of the drawing force may provide another source of information regarding the chemical reaction the yarn is undergoing as it passes through the oven. In the chemical reaction, the yarn may shrink or expand, and the amount of shrinkage or expansion allows the rate of the chemical reaction to be determined. Data regarding the shrinkage or expansion may complement data regarding the temperature and / or chemical composition of the yarn obtained by the optical detection device, and may therefore also be used to adjust the blower and / or heater.
[0057] The furnace according to the present application may comprise a paddle suitable for guiding the gas flow within the furnace, said paddle may be mounted such that it is movable and its movement is controllable by the control unit.
[0058] In one embodiment, the at least one blower is a turbo blower.
[0059] In one embodiment, at least one blower has a nozzle oriented to create a gas flow substantially perpendicular to the process direction. The nozzle may be adjustable in both width and direction, and can be used to vary the gas flow in the furnace. The adjustable nozzle may be controlled by a control unit.
[0060] In one embodiment, the furnace according to the present application may have at least one nozzle, which is connected to an inert gas reservoir, such as a cylinder or tank containing nitrogen, carbon dioxide, argon, or any other inert gas. The nozzle may then be used to selectively reduce the oxygen partial pressure in the at least one precursor fiber, and at the same time, to remove heat from the at least one precursor fiber to prevent the at least one precursor fiber from undergoing uncontrolled oxidation reactions, such as combustion. Thus, the effect of the nozzle releasing the inert gas can be likened to that of a gas-based fire extinguisher.
[0061] The present application further relates to a method of oxidizing PAN or other precursor yarns, comprising a self-learning artificial neural network.
[0062] An artificial neural network is composed of artificial neurons, which are software structures that can take input values, multiply them with weighting coefficients, and generate network input values from these weighted inputs, which are fed into activation functions that generate neuron outputs, which can be numbers or logical values ("true" or "false").
[0063] A self-learning artificial neural network can be trained using a training data set that includes both input values and desired ("correct") output values. Algorithms within the neurons of a neural network known to those skilled in the art can adapt weighting coefficients such that, given a training data set, the artificial neural network can generate the desired output values and thereafter learn to evaluate similar input values.
[0064] In this application, the term "self-learning artificial neural network" refers to any structure of artificial neurons that can be trained to generate reasonable outputs from given inputs. It is also explicitly mentioned that an isolated artificial neuron is also considered as a self-learning artificial neural network as long as it has the ability to adapt its input weights.
[0065] The number of neurons included in the self-learning artificial neural network according to the invention is practically unlimited. A typical self-learning artificial neural network includes artificial neurons in one or more so-called layers with different functions. For example, one layer serves as an input layer. Typically, the input layer receives input data, for example from a sensor or a computer, which generates input data for the artificial neural network. One layer of the artificial neural network generates an output. Any artificial neuron of the output layer can generate an output in the form of either a logical value or a number. It is noted that the output layer and the input layer of the artificial neural network may be the same layer. In such a case, there is a one-layer artificial neural network.
[0066] The artificial neural network may further include layers of neurons between the input layer and the output layer, e.g., connected to all or a portion of the input neurons, to all or a portion of the output neurons, and / or to all or a portion of further neurons that are not part of the input or output layers. A layer of artificial neurons that does not receive inputs from outside the artificial neural network or generate outputs outside the artificial neural network is called a "hidden layer" and the neurons it contains are called "hidden neurons".
[0067] If the connections between the neurons of an artificial neural network are only between neurons of adjacent layers, for example between the neurons of the input layer and the first hidden layer, between the neurons of the first hidden layer and the second hidden layer, and between the neurons of the second hidden layer and the output layer, the artificial neural network is called a feedforward artificial neural network. In a feedforward artificial neural network, information can only be transmitted from the input layer towards the output layer. An important type of feedforward artificial neural network is known to those skilled in the art as a perceptron.
[0068] Artificial neural networks can also be constructed so that an artificial neuron can receive inputs from itself, from neurons in the same layer, or from neurons in a layer closer to the output layer than the artificial neuron itself.
[0069] Said form of artificial neural network, in which information is not transmitted directly towards the output layer but in so-called feedback loops, is known to those skilled in the art as a recurrent artificial neural network. Due to the presence of feedback loops, a recurrent artificial neural network can have a memory.
[0070] Any kind of artificial neural network may be used in the method according to the present application. In one embodiment, the artificial neural network used is a perceptron. In one embodiment, the artificial neural network used is a recurrent artificial neural network. In one embodiment, the artificial neural network used is an adaptive linear neural network (ADALINE).
[0071] The key function of the artificial neural network used in the method according to the present application is to control the oxidation process of at least one precursor fiber.
[0072] Due to its highly exothermic nature, the oxidation process of the precursor yarn requires strict temperature control of the yarn undergoing the oxidation reaction to prevent the yarn from burning or even exploding due to heat accumulation in the fiber bundle. In current processing technology, the temperature control of the yarn is performed by controlling the temperature of the oven through which the precursor yarn passes. The oven temperature is therefore kept at the lower end of the temperature window in which the desired reaction occurs to avoid the precursor yarn ignition due to the highly exothermic oxidation process. It would not be possible to control the process at an effective level, for example by one person constantly monitoring the precursor yarn, because hundreds or even thousands of fibers are being processed simultaneously in the oxidation oven, making it impossible for one person, or a reasonable number of people, to monitor and manually control the oxidation process.
[0073] Therefore, there is currently no alternative to keep the yarn temperature at the lower end of the temperature window where the desired reactions occur. Typically, furnaces used in such processes include several stages, each of which is kept at a different constant temperature. The precursor yarn entering a stage begins to react when the heat of the furnace stage acts as an activation energy for the oxidation reaction, which causes the yarn to rise to a temperature above the temperature of the furnace stage. The temperature of the furnace stage is selected so that the oxidation reaction does not become runaway as it occurs, and the reaction rate and the temperature of the fiber decrease during the passage through the furnace module, resulting in a slower reaction rate and therefore a longer reaction time.
[0074] The use of the oven according to the invention allows the control of the reaction using an artificial neural network or several artificial neural networks. In the method according to the invention, the artificial neural network is provided with data from at least one optical detection means and / or a sensor on the guide means for the yarn. The data may be data on the temperature of the yarn, the temperature of the oven and / or data on the current chemical composition of the precursor yarn during oxidation or data on the contraction or expansion of the precursor yarn.
[0075] In one embodiment, said data may be recorded for each precursor yarn separately and fed to the artificial neural network together with global data, such as data on the temperature in the furnace or furnace module, or the speed and direction of gas flow in the furnace or furnace module, etc. Thus, the artificial neural network used in the method according to the present application may include at least as many input neurons as there are precursor yarns passed through the furnace.
[0076] The artificial neural network is then trained to be capable of calculating from the input data how the yarn temperature and / or the oven temperature may be varied to keep the difference between the first and second temperatures substantially constant. Thus, the artificial neural network should include at least as many output neurons as there are precursor yarns passing through the oven or oven module.
[0077] Depending on the output data of the neural network, the temperature of any individual precursor yarn can be increased or decreased using blowers and / or heaters to keep the difference between the first and second temperatures substantially constant. The use of artificial neural networks can provide powerful enhancements in the control of the oxidation process, and thus can be performed with much faster responsiveness than prior art.
[0078] The present application further relates to a method for oxidizing polyacrylonitrile precursor yarn, the method including the steps of providing a polyacrylonitrile precursor yarn, providing an oven according to the present application, drawing and opening the PAN precursor yarn, passing the PAN precursor yarn through the oven, controlling the amount of heat released by the polyacrylonitrile precursor yarn during oxidation, adjusting the amount of heat released by the polyacrylonitrile precursor yarn during oxidation using a heater and a blower, and winding up the oxidized polyacrylonitrile precursor yarn.
[0079] The present application further relates to a method for oxidizing a precursor yarn, the method including the steps of providing a precursor yarn, providing an oven according to the present application, opening and optionally stretching the precursor yarn, passing the precursor yarn through the oven, controlling the amount of heat released by the precursor yarn during oxidation and / or carbonization, using a heater and a blower to adjust the amount of heat released by the precursor yarn during oxidation, and winding up the oxidized and / or carbonized precursor yarn.
[0080] In one embodiment, a blower can blow an inert gas, such as nitrogen or carbon dioxide, into the furnace to regulate the amount of heat released by the precursor yarn during oxidation.
[0081] In one embodiment, the precursor yarns undergoing oxidation are selectively heated or cooled depending on their temperature.
[0082] In one embodiment, the tuning method is based on a self-learning artificial neural network, preferably a recurrent self-learning artificial neural network. [Brief description of the drawings]
[0083] [Figure 1] 1 is a schematic diagram of a furnace module according to the present application; [Diagram 2] FIG. 1 illustrates the gas flow and precursor or oxidized precursor yarn temperature during passage through a furnace module according to the present application. [Diagram 3]1 is a schematic diagram of one embodiment of a furnace according to the present application;
[0084] Description of the drawings 1 is a schematic diagram of a furnace module according to the present application having an inlet for precursor yarn 1 and an outlet for oxidized precursor yarn 2. In the furnace, a gas flow 3 is generated by a blower 7 and infrared radiation 4 emitted by a radiant heater 8 is directed towards the yarn in the furnace module. The infrared radiation is detected by a spectrometer 5 connected to a control unit 6 which controls the blower and the radiant heater. The blower may be equipped with a second heater (not shown).
[0085] FIG. 2 is a diagram illustrating the temperature of gas stream 21 and precursor or oxidized precursor yarn 20 during passage through a furnace module according to the present application.
[0086] 3 is a schematic diagram of an embodiment of an oven according to the present application, comprising a first stage 31 and a second stage 32. The first stage comprises three oven modules 33 according to the present application, and the second stage comprises two oven modules 33 according to the present application. Between each stage, a yarn transport unit 34 is installed. The gas flow in each oven module is maintained at a constant temperature level 35, while the temperature level of the PAN precursor yarn (not shown) in the oven module is maintained at a constant temperature level 36, which is higher than the constant temperature level 35, by using the oven module's radiant heater, blower and control unit. Between these stages, the temperature of the precursor yarn is the temperature outside the oven 37 while passing through the yarn transport unit.
Claims
1. A continuous furnace having a process direction and at least one furnace module, the furnace module having two openings arranged such that at least one precursor fiber can pass through the furnace module in the process direction, the furnace module further comprising at least one blower suitable for forming a gas flow perpendicular to the process direction, at least one adjustable radiant heater, preferably a microwave heater, radiating perpendicular to the process direction for heating the at least one precursor fiber to a first temperature, at least one second heater, preferably an electric resistance heater, for heating the gas in the furnace to a second temperature, at least one optical detection means, preferably an infrared camera or an infrared spectrometer, and at least one control unit, preferably a computer, the blower, the radiant heater, the second heater, the optical detection means, and the control unit function as regulating means for the heat released by the at least one precursor fiber to keep a temperature difference between the first temperature and the second temperature less than 10° C., and the first temperature is higher than the second temperature.
2. The furnace of claim 1 , wherein the furnace comprises several furnace modules.
3. 3. The furnace according to claim 1 or 2, wherein the furnace has guide means for the yarn, such as pulleys, loops, combs or godets, at the interface between at least two modules.
4. 4. The furnace of claim 3, wherein the drawing force of the guide means for the yarn is measurable and can further be used as a regulating means for the heat released by the at least one precursor fiber.
5. The furnace of claim 1 , wherein the at least one precursor fiber is at least one thread of polyacrylonitrile precursor yarn.
6. The furnace according to any one of claims 1 to 5, wherein the radiant heater is an infrared laser.
7. The furnace according to any one of claims 1 to 6, wherein the furnace comprises a paddle suitable for guiding the gas flow within the furnace.
8. 8. The furnace according to claim 1, wherein at least one blower is a turbo blower.
9. 9. The furnace of claim 1, wherein the at least one blower has a nozzle directed into the furnace.
10. 1. A method for oxidizing a precursor yarn, comprising at least one self-learning artificial neural network.
11. 11. The method of claim 10, wherein the at least one self-learning artificial neural network is a recurrent neural network.
12. 1. A method for oxidizing polyacrylonitrile precursor yarn, comprising: a. providing a polyacrylonitrile precursor yarn; b. Providing a furnace according to any one of claims 1 to 9; c) A step of drawing and opening the polyacrylonitrile precursor yarn; d. passing the polyacrylonitrile precursor yarn through the oven; e. controlling the amount of heat released by the polyacrylonitrile precursor yarn during oxidation; f. using the heater and the blower to adjust the amount of heat released by the polyacrylonitrile precursor yarn during oxidation; g. Winding the oxidized polyacrylonitrile precursor yarn; A method comprising:
13. 13. The method of claim 12, wherein the blower for adjusting the amount of heat released by the polyacrylonitrile precursor yarn during oxidation can blow an inert gas, preferably nitrogen or carbon dioxide, into the furnace.
14. 14. The method of claim 12 or 13, wherein the polyacrylonitrile precursor yarns undergoing oxidation are selectively heated or cooled depending on their temperature.
15. 15. The method according to any one of claims 12 to 14, wherein said adjustment of said method is based on at least one algorithm or at least one self-learning artificial neural network, preferably a recurrent self-learning artificial neural network.
