RU2197575C2 - Method of forming nonwoven material from foam pulp with the usage of illegible algorithm controller - Google Patents

Abstract

FIELD: manufacture of nonwoven materials. SUBSTANCE: method involves mixing cellulose, synthetic or glass fibers, water, air, recirculating foam and surfactant in mixing reservoir for preparing foam pulp; discharging foam pulp into web former; controlling forming process; forming fibrous web in web former at rate equal to pulp feeding rate by removing liquid and foam from pulp and collecting at least part of removed liquid and foam in pan located under netted conveyor; providing additional action upon fibrous web formed in web former for obtaining ready nonwoven web; performing at least one of said stages with the use of illegible algorithm controller for regulating level of removed liquid and foam in pan under netted conveyor, level of pulp in mixing reservoir, pressure in web former collector, foam density, flow coefficient (foam flow rate divided into netted conveyor rate), as well as for controlled feeding of surfactant, total surface density of produced fibrous web and, in particular, in case of manufacture of fibrous web from fiber glass, level of binder in reservoir. Certain input parameters are used for each illegible algorithm controller. For example, at least some of following parameters: density and rate of foam flow repeatedly fed into mixing reservoir from pan under netted conveyor, pH value of foam in mixing reservoir, level of foam in reservoir and amount of pulp added into mixing reservoir, are used as input parameters for controlling level of foam in reservoir. Neuronic control network may be used for quality controlling throughout the whole system for manufacture of nonwoven material and information on laboratory tests is introduced into neuronic control network for increasing its level of effectiveness. EFFECT: increased efficiency in formation of fibrous web and wider range of fibers and fillers used for manufacture of nonwoven materials. 29 cl, 25 dwg, 6 tbl

Description

Background to the invention and a brief description of the invention
With the effective implementation of the method of forming a fibrous canvas from a fibrous mass in the production of non-woven material, for example, described in US patents 3716449 and 3871952 (the essence of which is incorporated into this application by reference), a number of advantages are achieved in comparison with the method of forming a fibrous canvas from a water-fibrous mass . However, in the past it was difficult to make the process of forming a fibrous web from a foam pulp industrially efficient using many different types of fibers. Although there are a number of industrial plants for forming fibrous canvas from polypropylene fibers or glass fibers in the production of nonwoven materials, difficulties may arise in the management of such plants and effective industrial production has not been achieved in the molding of fibrous canvas from cellulose or synthetic fibers (except for the polypropylene processing plants mentioned higher).

 Until now, the processes of molding fibrous canvas from foam in the manufacture of nonwoven materials have been controlled manually or using proportional, integral, and differential controllers (PID controllers). The processes can be controlled manually, but this requires long periods of training, knowledge of the processes in perfection and a great concentration of attention of the operating personnel in order to be ready to perform all the required adjustment operations in the correct order and precisely in size. In the established mode of operation, when there are no process disturbances, manual control or control using PID controllers can be considered acceptable, since quality products are usually provided that meet customer requirements. However, some consumers put forward higher demands on product quality (probably due to higher demands on finished products consumers), which easily lead to a significant increase in the number of rejects, i.e. to release products that do not meet the requirements of the consumer, which must be disposed of. In addition, all process disturbances, such as start-up of a line, changes in the quality of raw materials, etc., lead to additional problems and require even more competent service personnel in order to make quick and smooth adjustments possible when starting a line or changing quality raw material products.

 When comparing a manually controlled process using PID controllers with a process using the first experimental versions made in accordance with the present invention, it was quickly found that the time required to compensate for a line start is halved, the time required to compensate changes in the quality of raw materials is reduced by at least half, in some special cases, time can be reduced to almost zero, the number of scraps was reduced by at least half, the spread is controlled process parameters has been reduced by half and the physical parameters of the spread batt was reduced by half. Since the above results were obtained on the line with the beta index made in accordance with the present invention, it is hoped that a better understanding of the invention and fine tuning of fuzzy control algorithms and equipment will lead to even better results.

 In accordance with the present invention, it is possible to efficiently control the process of forming a fibrous web from a fibrous mass with the efficient use of virtually any fibers and fillers in the production of a wide variety of raw materials of non-woven materials that differ in appearance and surface density, and take advantage of the process of forming a fibrous web from foam fiber mass. The main distinguishing features of the present invention, due to which it is possible to effectively control, are the use of controllers with a fuzzy algorithm in a number of different stages performed during the formation of a fibrous canvas. It is also preferable to use a neural control network to collect test data to determine quality indicators (produced off-line) and process indicators to determine milestones for long-term regulation and forecasting. Multiparameter control can also be used to measure the thickness of the fibrous web and to control the dilution in / or / when supplied to individual distribution pipes to set control points for various controllers with a fuzzy algorithm. Fuzzy algorithm controllers, a neural control network, and multi-parameter control tools used in accordance with the invention are all common standard products, such as those supplied by Honeywell-Alcont.

 In accordance with one aspect of the present invention, a system is provided for the production of nonwoven materials from cellulosic, synthetic or glass fibers. This system contains the following components: mixing tank for mixing cellulosic, synthetic or glass fibers, water, air, reusable (recyclable) foam and surfactant to form a foam-fiber suspension (mass); a web for forming a fibrous web with a fibrous web moving speed corresponding to the forming speed by removing liquid and foam from the mass and collecting at least a portion of the removed fluid and foam in a pallet under the mesh conveyor; a pump for injecting the foam mass from the mixing tank into the web-forming agent; means for subsequent exposure to the fibrous canvas molded on the canvas, to obtain a finished non-woven material; and a plurality of controllers with a fuzzy algorithm, including at least one controller with a fuzzy algorithm for automatically controlling the density of the foam in the mixing tank and at least one controller with a fuzzy algorithm for automatically controlling the weight level in the mixing tank.

 The input parameters of the controller with a fuzzy algorithm for automatically controlling the level in the mixing tank are the density and speed of the foam flow, which is re-fed into the mixing tank from the pallet under the mesh conveyor, the pH of the foam in the mixing tank, the level of foam in the pallet under the mesh conveyor and the amount of fiber added into the mixing tank. It is also preferable to use controllers with a fuzzy algorithm to control at least the level in the pallet under the mesh conveyor, the pressure in the collector of the weaving agent and the outflow coefficient, as well as to control the supply of surfactants and the total surface density of the produced non-woven material. A binder is also added in the manufacture of a nonwoven material containing at least 10% glass or aramid fibers, the binder being fed to the binder reservoir. In this case, the system further comprises a controller with a fuzzy algorithm for controlling the level in the reservoir for the binder.

 Typically, the web forming agent comprises a movable mesh conveyor and a headbox. One of the controllers with a fuzzy algorithm preferably comprises a controller with a fuzzy algorithm for automatically controlling the ratio of the air / foam mixture supplied to the scraper, including controlling the speed of the scraper conveyor and pressure in the headbox; moreover, at least some value of the surface density of the formed fibrous web, pressure in the headbox, the level of foam in the pan under the mesh conveyor, the density of the recirculated foam and the amount or speed of removal of foam from the headbox are used as input parameters of the controller with a fuzzy algorithm.

 The means for subsequently treating the fibrous web from the foam may include means for washing the fibrous web and for removing liquids from the fibrous web during or in connection with the washing (typically any conventional washer and / or vacuum fibrous web can be used). In this case, one of the controllers with a fuzzy algorithm automatically controls the means for washing and removing liquids, and at least some of them are used as input parameters of the controller with a fuzzy algorithm: the speed of the canvas, the surface density of the fibrous canvas, the temperature of the washing liquid, the speed suction foam and pressure in the flushing agent.

 The means for the subsequent processing of the molded fibrous web may include a conventional dryer, in which case one of the controllers with a fuzzy algorithm automatically controls the dryer, and at least some of the following parameters are used as input parameters of the controller with a fuzzy algorithm: setting drying temperature , the speed of transportation of the fibrous canvas, the amount of energy supplied to the dryer, the humidity level in the dryer and the pressure drop over and under the fibrous web at various points along the length of the dryer.

 The system may further comprise a neural control network for at least partially interacting with controllers with a fuzzy algorithm to control the process of forming the fibrous web and / or to carry out quality control over essentially the entire line for the manufacture of non-woven material.

 According to another aspect of the present invention, there is provided a method for manufacturing a non-woven material from cellulosic, synthetic or glass fibers, comprising the following steps: (a) mixing cellulosic, synthetic or glass fibers, water, air, recyclable foam and surfactant in a mixing tank for preparing foam mass; (b) injecting the foam into the scrim; (c) control of the operation of the binder; (d) in a scrim, forming a fibrous web at a web speed by removing liquid and foam from the mass in the scrim, and collecting at least a portion of the removed fluid and foam in a pan under the mesh conveyor; (e) an additional effect on the canvas molded in the canvas, to obtain a finished non-woven material; (f) the implementation of at least stage (a) using a controller with a fuzzy algorithm.

 Stage (a) can be carried out partly by adjusting the mass level in the mixing tank, and stage (f) can be done partially by automatically adjusting the level in the mixing tank using a controller with a fuzzy algorithm in which at least some of the input parameters are used of the following parameters: density and flow rate of the foam re-fed into the mixing tank from the pallet under the mesh conveyor, the pH of the foam in the mixing tank, the level of foam in the pan under the mesh conveyor and the amount of fiber added to the mixing tank. Stage (a) can also be carried out by automatically controlling the amount of surfactant added and by recirculating a certain amount of water removed from the fibrous web during molding and separated from the air; and then stage (f) is carried out partially by automatically controlling the amount of surfactant added using a controller with a fuzzy algorithm, in which at least some of the following parameters are used as input parameters: surfactant flow rate, pressure in the collector of the blowing agent, foam level in the tray under the mesh conveyor, the flow rate of the added fiber and the flow rate of recycled water.

 Step (c) can be carried out, at least in part, by automatically adjusting the air / foam ratio when fed to the scrim, including controlling the speed of the conveyor belt in the scrim and pressure in the headbox; and then stage (f) is carried out in part by using a controller with a fuzzy algorithm, in which at least some of the following parameters are used as input parameters: surface density of the formed fibrous web, pressure in the headbox, the level of foam in the pallet under the mesh conveyor, the density of the recirculated foam and the amount or rate of removal of foam from the headbox. Stage (e) is performed for washing the canvas and removing liquid from the fibrous canvas during or in connection with the washing; and then step (f) is performed in part to automatically control step (e) by using a fuzzy controller in which at least some of the following parameters are used as input: fibrous web spinning speed, washer pressure, surface density fibrous canvas, the temperature of the wash water, the suction rate of the foam.

 The method may also further comprise the step of using a neural control network to implement quality control over substantially the entire nonwoven fabrication line.

 According to another aspect of the present invention, there is provided a method of manufacturing a non-woven material from cellulosic, synthetic or glass fibers, which comprises the following steps: (a) mixing cellulosic, synthetic or glass fibers, water, air, recyclable foam and surfactant in a mixing tank for preparing foam mass; (b) injecting the foam into the scrim; (c) control of the operation of the binder; (d) in a web former, forming a fibrous web at a speed of moving the fibrous web by removing liquid and foam from the mass in the web and forming at least a portion of the removed fluid and foam in a pan under the mesh conveyor; (e) additional exposure to the fibrous canvas molded in the canvas, to obtain the finished non-woven material; (f) performing at least steps (a) to (e) using a controller with a fuzzy algorithm; and step (g) using a neural control network to implement quality control over essentially the entire nonwoven fabrication line.

 Step (e) can be performed to dry the fibrous web, with most of the fibers added in step (a) being glass fibers to which a binder is added. In this case, step (f) is performed in part to control the drying process of the fibrous web and the addition of a binder using controllers with a fuzzy algorithm.

 Stage (a) can also be performed in part to accurately control the pH in the mixing tank, using a variety of pH meters to measure acidity; and then step (f) is partially performed using a controller with a fuzzy algorithm to control and coordinate pH meters.

 In accordance with another aspect of the present invention, a method for manufacturing a nonwoven material from cellulosic, synthetic or glass fibers, which contains the following stages: (a) mixing cellulosic, synthetic or glass fibers, water, air, recycled foam and surfactant in a mixing tank for preparation foam fiber mass; (b) injecting the foam into the scrim; (c) control of the operation of the binder; (d) forming a fibrous web at a web former at a speed of moving the fibrous web by removing liquid and foam from the mass in the web and forming at least a portion of the removed fluid and foam in a pan under the mesh conveyor; (e) additional exposure to the fibrous canvas molded in the canvas, to obtain the finished non-woven material; and stage (f), the use of controllers with a fuzzy algorithm to control at least the level in the pallet under the mesh conveyor, the level in the mixing tank, the pressure in the collector of the scum, foam density and the expiration coefficient.

 Stage (f) can be additionally performed to control the supply of surfactants and the total surface density of the produced non-woven material. A binder can also be added during the manufacture of a nonwoven material containing at least 10% glass or aramid fibers, the binder being added to the binder tank, and step (f) can be performed to control the level in the binder tank.

 The main objective of the present invention is the creation of an effective control system for the process of forming fibrous canvas from foam in the production of non-woven material.

 This and other objects of the invention will become more apparent upon reading the detailed description of the invention and the appended claims.

Brief Description of the Illustrations
In FIG. 1 is a general block diagram of an exemplary system for implementing a process for forming a fibrous web from a foam pulp made in accordance with the present invention;
in FIG. 2 is a detailed schematic view, partially in cross section, showing the supply of foam material from a mixing tank to a pump supplying a manifold and a headbox;
figure 3 is a detailed schematic isometric view, partially with tears, showing the possibility of adding foam itself to the pipe between the collector and the headbox;
figure 4 is a side view, partially in section of the part shown as an example, the canvas with an inclined mesh conveyor, which can be used to carry out the process of forming a fibrous canvas from the foam mass;
5 is a schematic diagram illustrating the effect of adding foam to pipes directed from the manifold into the headbox;
Fig.6 is a schematic representation of the surface density of the fibrous canvas in the headbox shown in Fig.4 and 5, with the addition of foam and without it;
Fig. 7 is a schematic end view, partially in section, of an exemplary vertical web-forming machine that can be used to carry out the process of forming a fibrous web from a foam-fiber mass instead of the web-forming machine with an inclined mesh conveyor shown in Fig. 4;
in FIG. 8 is an end view with parts of components with tears for clarity of the illustration, in which pipes are shown in cross section of the device for feeding other materials, located in the center of the device shown in FIG. 7;
in FIG. 9 is a schematic end view, partially in section, of one of the vacuum chambers used in conjunction with the pressure boxes / holders presented in FIGS. 4 or 7;
figure 10 is a side view of the canvas, presented in figure 7, together with other components of the system for implementing the process of forming a fibrous canvas from the foam mass;
figure 11 is a schematic view showing an embodiment of the components of the line shown in figure 10, with a device for returning foam from the vacuum chambers to the pallet under the mesh conveyor;
in FIG. 12 is a schematic side view showing, by way of example, a flow chart of processing a fibrous web formed by the apparatus of FIG. 1, after molding, including washing the canvas and applying a layer of material using a simple installation for coating;
Fig.13-16 is a block diagram of various input parameters and controlled functions of the controllers with a fuzzy algorithm of the system shown in Fig.1;
in FIG. 17 is a block diagram of connections between a fuzzy logic control system, a neural control network, and a multi-parameter control system that can be used in accordance with the present invention;
in FIG. 18 is a more detailed block diagram of the control system shown in FIG. 17, showing various systems and parameters that can be controlled, and input parameters input to control systems made in accordance with the present invention;
on Fig is a block diagram showing the use of a control system with fuzzy logic to determine the difference between the desired density and the measured density of the foam used in the process of forming a fibrous canvas from the foam in accordance with the invention;
in FIG. 20 is another block diagram showing a foam density monitoring system using a controller with a fuzzy algorithm;
in FIG. 21 is a block diagram of the conversion of measurement parameters into a group of fuzzy sets "phase";
in FIG. 22 is a graphical representation illustrating an example conversion of measured foam density values into a group of fuzzy sets;
on Fig is a diagram illustrating the principle of operation of the "rule base" used when converting measurement parameters into a group of fuzzy sets;
in Fig.24 is a block diagram similar to that shown in Fig.21, only illustrating the principle of "dephasing";
on Fig is a schematic representation of an example algorithm "dephasing".

Detailed description of illustrations
An exemplary system for making mats or fibrous webs from cellulosic and synthetic fibers in accordance with the method of forming a fibrous canvas from a foam according to the invention is shown schematically in FIG. 1 and indicated by pos. 10. The system includes a mixing tank or pulper 11, equipped with an inlet 12 for supplying fiber, an inlet 13 for supplying a surfactant and an inlet 14 for supplying other additives, for example chemical compounds for controlling pH, for example potassium carbonate or acids, stabilizers etc. The nature of the fibers, surfactants, and particulate additives does not play a significant role and can be varied widely depending on the exact characteristics of the product being produced (including its surface density). It is desirable to use surfactants that can be washed quite easily, since surfactants lead to a decrease in the surface tension of the finished fibrous web if they remain in the fibrous web, in particular this refers to the Weyerhaeuser products mentioned below, for which this is an undesirable property.

 The mixing tank 11 itself is a completely conventional tank of the same type as the tank that is used as a mixer in conventional paper making in the process of deposition of the fibrous layer from the water-fiber mass. The only difference is that the side walls of the mixing tank 11 protrude up to three times higher compared to the tank intended for pulp, since the foam has a density of approximately one third of the density of water. The speed and configuration of the mixer of a conventional mechanical mixer in the mixing tank 11 can be varied depending on the specific properties of the product being produced, but these parameters are not significant, and a wide range of different components and variable parameters can be used. Braking agents can also be arranged on the walls. At the bottom of the mixing tank 11, a vortex generator is arranged through which foam is drained, but the vortex is not visible during start-up, since the mixing tank 11 is filled with foam and fiber.

 A large number of pH meters 15 are also preferably installed in the mixing tank 11 for measuring pH at a number of different points. pH has an effect on surface tension, and therefore acidity must be precisely determined. pH meters 15 are checked daily.

 At the initial start-up, water with fiber is added via pipeline 12, surfactants - via pipeline 13, and other additives - through pipeline 14; however, as soon as the process is started, additional water is not required, and mainly foam is maintained in the mixing tank 11, not only foaming.

 The foam mass is discharged through the bottom of the mixing tank 11, through the vortex generator, through the pipe 16 under the influence of the pump 17. The pump 17, like all other pumps in the system 10, is preferably a degassing centrifugal pump. The foam leaving the pump 17 enters through a pipe 18 to subsequent devices.

Figure 1 shows a selectable standby, or buffer, tank 19, shown by dashed lines. A reserve, or buffer, tank 19 is optional, but may be desirable to provide relatively even distribution of the fiber in the foam when there are some fluctuations when the mixture is fed into the mixing tank 11. This means that the buffer tank 19 (which is sufficient small, usually its volume is only about 5 m ³ ) plays a role more or less similar to the “foam trap” for equalizing the distribution of the fiber. Since the total travel time from the mixing tank 11 to the headbox is usually only 45 seconds when performing the process in accordance with the invention, the buffer tank 19, if used, provides a margin of time for balancing vibrations in the fiber distribution.

 When the buffer tank 19 is used, the foam is pumped 17 through a pipe 20 to the top of the buffer tank 19, and withdrawn from the bottom of the tank through a pipe 21 under the influence of a pump, preferably a centrifugal pump 22, and then fed to a pipe 18. This means that when used buffer tank 19, pump 17 is not directly connected to the pipe 18, but only through the buffer tank 19.

 The pipe 18 is directed into the pallet 23 under the mesh conveyor. The pallet 23 under the mesh conveyor itself is a regular bath, again the same as in a conventional line for the process of deposition of fibrous canvas from the water-fiber mass, but with higher side walls. It is important to make the pallet 23 under the mesh conveyor so that there are no dead corners and, therefore, the pallet 23 should not be too large. A conventional device 24, with which it is possible to direct the foam-fiber mixture into the pipe 18 and to the pump 25 (which is operatively connected near the bottom of the pallet 23 under the mesh conveyor), will be described below with reference to Figure 2. In any case, the foam-fiber mixture introduced by the device 24 is transported by the pump 25 through the pipe 18, and an additional amount of foam from the pallet 23 under the mesh conveyor is fed into the pipe 26. Since a sufficiently large amount of foam from the pallet 23 is sucked out by the pump 25, usually the consistency of the mixture in pipe 26 is significantly less than in pipe 18. The consistency of the mixture in pipe 18 is usually 2-5% solid (fiber), while in pipe 26 it is usually 0.5-2.5%.

 In the pallet 23 under the mesh conveyor there is no significant separation of the foam into layers with different densities. Although there is a minimal increase in density towards the bottom of the pallet, this degree of increase in density is usually small and does not adversely affect the operation of the line.

 Through the pipeline 26, the foam-fiber mixture is fed into the manifold 21, which contains foaming nozzles 28 associated with it. Preferably, the nozzles 28 — which are conventional foaming nozzles (which strongly mix the foam) described in patents 3,716,449 and 3,871,952 — are mounted on the manifold 27, with a large number of nozzles 28 mounted on the manifold 27. A pipe 29 extends from each nozzle 28, which faces the headbox 30 of the web-forming agent through which a conventional mesh conveyor (perforated element) (or conveyors) is moved.

 The headbox 30 contains a plurality of vacuum chambers (usually three to five) 31, through which the foam is sucked off from the opposite side of the mesh conveyor from the side of the foam-fiber mixture inlet, and the last separation chamber 32, which is located at the outlet of the fiber canvas forming zone 33 from the headbox 30. The number of vacuum chambers 31 located in the zone of the molding table for controlling drainage is increased in the manufacture of denser products or when operating at a higher speed. The molded fiber web 33, which typically has a solids consistency of about 40-60% (e.g., about 50%), is preferably washed, as shown schematically as washing stage 34 in FIG. 1. Flushing stage 34 is designed to remove surfactants. Due to the high consistency of canvas 33, a minimum amount of drying equipment is required.

The canvas 33 after the washer is passed through one or more optional installations 35 for coating and sent to a conventional drying machine 36. In a conventional drying machine 36, when synthetic fibers with a shell-core structure (for example, Cellbond fiber) form part of the canvas 33 drying machine 34 is used in a mode in which the temperature of the canvas is brought to a value higher than the melting temperature of the sheath material (usually polypropylene), while the core material (usually polyethylene terephthal t (PET)) is not melted. For example, when using Cellbond fiber in canvas 33, the temperature in the dryer is usually about 130 ^{° C.} or slightly higher, and this temperature is equal to or slightly higher than the melting temperature of the fiber sheath, but well below the melting temperature of the fiber core, which is about 250 ^{° C.} Thus, the process of bonding the canvas is carried out due to the material of the fiber sheath, but the integrity of the product (provided by the fiber core) is not adversely affected.

 Although not always necessary, the method according to the invention provides for the addition of clean foam to or directly next to headbox 30 to achieve a number of favorable purposes. As shown in FIG. 1, with a pump, preferably a centrifugal pump, 41 suction the foam from the pallet 23 under the mesh conveyor through a pipe 40. The foam is fed through a pipe 40 to a feeder 42, from which foam is then distributed through a large number of different pipelines 43 directed to the pressure head box 30. Foam can be poured, as shown by line 44, directly under the cover of head box 30 (in the case of a head box with an inclined mesh conveyor) and / or through pipes 45 to pipelines 29 (or nozzles 28) for inflowing foam-fiber mixture and into the headbox 30. In detail, a foam infuser will be described with reference to FIGS. 3-6.

 Vacuum chambers 31 suck out the foam leaving the headbox 30 and direct them through pipelines 46 to the pallet 23 of the mesh conveyor. Usually, pumps are not required for these purposes.

 A significant amount of foam from the pallet 23 under the mesh conveyor is re-sent to the mixing tank 11. The foam is sent through a pipe 47 by a pump, preferably a centrifugal pump, 48 and then passed through a pipe 47 through a conventional density measuring device 49 installed in the pipe (densitometer) for input ( as shown schematically, item 50) back to the mixing tank 11. In addition to measuring the density of the foam in the pipe 47 using a densitometer 49 (as shown schematically in FIG. 1), one or more meters 49A devices (for example, densitometers) 49A can be installed directly in the mixing tank 11.

 In addition to recycling the foam, water is also recycled. The foam removed from the last vacuum chamber 32 is sent via line 51 to a conventional separator 53, for example, a cyclone separator. A separator 53, for example, of a vortex action, separates air and water from the foam supplied to the separator 53, to obtain water with a very small air content in it. The separated water is sent through a pipe 54 from below from the separator 53 to the tank 55 for water. The air separated by the separator 53, is directed through the pipeline 56 through the fan 57 from the top of the separator 53 and is released into the atmosphere or used in the combustion process, or processed in any other way.

 In the water tank 55, a fluid level 58 is maintained by draining the fluid to a certain sewer or for treatment, as schematically shown in pos. 60 in FIG. 1. Water is also withdrawn below level 58 from the water tank 55 through line 61 and, under the influence of a pump, preferably a centrifugal pump, 62 is fed through line 61 through a conventional flow meter 63 (by which pump 62 is controlled). Finally, recycled water is supplied, as shown schematically in pos. 64 in FIG. 1, from above into the mixing tank 11.

 Typical exemplary flow rates are: 4000 l / min of pulp in a pipe 18, 40,000 l / min of pulp in a pipe 26, 3500 l / min of foam in a pipe 47 and 500 l / min of foam in a pipe 51.

 System 10 also contains a number of new control components. The first controller 71 with a fuzzy algorithm controls the level of foam in the tank 11. The second controller 72 with a fuzzy algorithm controls the addition of surfactants through the pipe 13. The third controller 73 with a fuzzy algorithm controls the formation of the canvas in the low zone of the headbox 30. The fourth controller 74 with a fuzzy the algorithm controls the washing machine 34. The fifth controller 75 with a fuzzy algorithm controls the pH meters 15 and, possibly, controls the flow of other additives through line 14 into mixing tank 11. The control using fuzzy controllers are also used for adjusting the surfactant content and the forming web. A multi-parameter control system and a neural control network (see FIG. 18) are also preferably used as means imposed on other control systems. A multi-parameter control system is also used to control the expiration coefficient when forming a fibrous web. Parameters can be changed depending on their influence on the required process control and final results.

 To improve the control quality of the various components of the system, scales 76 are usually installed in the fiber supply section 12 in order to accurately determine the amount of fiber added per unit time. Valve 77, as well as scales 78 can be installed in the pipe 13 to control the supply of surfactants. Valve 79 may be installed in the pipe 14.

 System 10 should be considered unique among systems for the production of non-woven materials from foam, since there are practically no valves deliberately contacting the foam at any point during processing, with the possible exception of valves installed in pipeline 46, which will be described with reference to 11.

 In addition, during the entire process of FIG. 10, the foam is maintained in a state of relatively high shear modulus. Since the higher the shear modulus, the lower the viscosity, it is desirable to maintain the foam under the condition of a high shear modulus. The foam-fiber mixture behaves like a pseudo-plastic, demonstrating non-Newtonian behavior.

 The use of the process of deposition of canvas from a foam-fiber mixture allows to achieve a number of advantages in comparison with the process of deposition from a water-fiber mixture, especially in the manufacture of materials with a high degree of absorption. In addition to the reduced need for drying, due to the high consistency of the fibrous canvas 33, the process using foam allows for the uniform distribution of almost any kind of fiber or particles (without excessive "settling" of high-density particles. While low-density particles "settle" in which at least - they don’t settle in water at all) in the mass (and, ultimately, in fibrous canvas), due to the fact that the fibers or particles have a specific gravity in the range 0.15–13.0. The process of deposition from the foam also allows the formation of fibrous canvases with a surface density that can be changed over a wide range, and to obtain a material with increased uniformity and greater bulk in comparison with the material made using the molding of fibrous canvas from a water-fiber mixture, and having a very high uniformity. A plurality of pressure boxes can be installed in series, or two layers can be produced simultaneously inside a pressure box with two mesh conveyors, and / or simple plants 35 can be used to apply additional layers (as a coating) with great simplicity of the process.

 A detailed description of the components of the system shown in FIG. 1, if they are in any way different from fully known equipment, is given below with reference to FIGS. 2-16.

 Figure 2 shows the process of feeding the foam-fiber mixture and foam into the pump 25 connected to the pallet 23 of the mesh conveyor. The device 24 is known by a process previously carried out by Wiggins Teape in which the direction of movement of the foam mixture supplied through line 18 is sharply changed, as shown in FIG. 2, by using a curved pipe 83 so that the foam mixture exits directly from its open end 84 pump inlet channel 25. Foam from the pallet 23 under the mesh conveyor also enters the inlet channel 85, as shown by arrow 86. The operation of the pump 48 is controlled by a fuzzy control system; adjust the level in the pallet 23 under the mesh conveyor.

 In cases where the fiber to be used in the preparation of the foam is sufficiently long (i.e., of the order of several inches), then instead of directing the pipe 18 into the suction channel 85 of the pump 25 (see Fig. 2), the pipe 18 cut into the pipe 26 downstream of the product from the pump 25. In this case, the pump 17 should, of course, create a higher pressure than would be necessary in any other case to ensure sufficient pressure to allow the passage of flow from the pipe 18 to the pipe 26, not shaking on the pressure in the pipeline 26 created by the pump 25.

 In FIG. Figure 3 shows the details of one form of a new foam accessory, which is a distinctive feature of the Alstrom process. Figure 3 shows how the foam itself, from the pipeline 45, is fed into the foam-fiber mixture in the pipe 29 immediately in front of the headbox 30. In other words, the foam is purely added to the foam-fiber mixture coming from the manifold 27 through nozzles 28. When they use pipelines 45 for injection, it is not necessary to inject foam through all the tubes 29, the quantity of them is enough to achieve the desired result.

 Figure 4 shows an example canvas with an inclined mesh conveyor and its headbox 30I, in which two different types of foam injection are used (one of which is shown in Figure 3 and plus another shape). In the headbox 30I of FIG. 4, a conventional inclined mesh web 90 is formed in the direction of the arrow and the foam injection means 45 is dispersed in the headbox 30I, dispensing from the tubes 29, generally as shown in Figure 4. Foam is also let into the headbox 30I through tubes 44 so that the foam flows in a general direction in the direction indicated by arrow 92 in FIG. 4. It is the foam flow in the direction of arrow 92 that moves along the bottom of the lid 93 of headbox 30I. A wall 94 may be provided in headbox 30I to provide an initial foam flow in the direction of arrow 92 from each of the plurality of tubes 44.

 The foam supplied through the tubes 44 is designed to provide less fiber displacement in the headbox 30I, thereby preventing the possibility of a shift between the fibers and the cover 93 of the headbox 30I to cause the fibers to rotate and orient in one direction, i.e. in the direction of movement of the mesh conveyor 90. According to the basic principles of flow dynamics, if the foam-fiber mixture moves along the cover 93, then at the boundary layer there are disturbances in the orientation of the fibers, which is undesirable. By means of the foam supplied in the stream in the direction of arrow 92, this problem associated with the boundary layer is removed. In addition, due to the flow of foam supplied through the tubes 44 in the direction of arrow 92, the bottom side of the lid 93 is kept clean, which is also desirable.

Foam supply to tubes 45 (typically at an angle of about 30-90 ^° ), as shown in FIG. 3 and 4 are produced for a different purpose. Figure 5 is a schematic top view (which shows only three tubes 29, while there are usually a lot of them) of the pressure box 30 (for example, 30I), which shows the difference that occurs when clean foam is injected. In the absence of injection of substantially fiber-free foam in direction 45, the foam-fiber mixture supplied through tubes 29 is distributed generally as shown by lines 91 in FIG. 4 and 5. However, when the foam is injected in direction 45, the profile of the canvas changes, since there is a large dispersion of the foam-fiber mixture, as shown schematically by lines 96 in FIG. 5. The effect of this phenomenon on the profile of the formed fibrous web is shown in the schematic drawing in FIG. 6. The normal profile of the fibrous web (when there is no foam injection) depicted by line 91A includes a large bulge 97. However, when the foam is injected, as shown by line 96a, the bulge 98 is much less. This means that the fibrous layer is more uniform. Profile control is carried out by diluting the foam in the collector in the main stream immediately before or immediately after the tubes 29 (the first case is shown at 45 in FIG. 4).

 If desired, the tubes 29 can direct foam from nozzles 28 into the blast chamber in pressure boxes 30I, 30V. However, it makes no real sense to use an explosive chamber in pressure boxes during the Alstrom process. If it is used, then it is intended simply to ensure security.

 In FIG. 7 shows an alternative configuration of the headbox that can be used in the system 10. The entire web former, as well as the headbox 30V, is similar to the usual vertical web forgerboard with two mesh conveyors and a headbox and contains mesh conveyors 90, 90A. In the exemplary embodiment shown in FIG. 7, a capacitor 100 is shown located at the outlet portion of the web former, and shafts 101, 101A for guiding the mesh conveyors 90, 90A. In one embodiment, the mesh conveyor 90A can also be guided by a capacitor 100, as shown by dashed lines, although during normal operation, the mesh conveyor 90A bends around the upper shaft 101 along with the fibrous web 33 after exiting the headbox. Perforated tables are cheaper than cylindrical capacitors, and their use is preferable, although capacitors can be used, for example, such as the designated pos. 100 in FIG. 7.

 Headbox 30V includes bottom 102 and side walls 103, 104. Between the side walls 103, 104 and the central device 110 are volumes 105, 106 for the foam-fiber mixture. Although the same foam-fiber mixture can be supplied to volumes 105, 106, they are usually supplied with completely different mixtures, from which two different layers are formed in the fibrous canvas 33. One foam-fiber mixture is supplied from the collector 27 through nozzles 28, for example, through tubes 29 through the bottom wall 102 of headbox 30V, as shown by inlets 107, while another foam-fiber mixture is supplied from manifold 21A, through nozzles 28A through inlet 107A in bottom wall 102 of headbox 30V. Alternatively or additionally, the foam-fiber mixtures can be introduced through pipes 29 ′ and 29′A through the inlets 108, 108A in the side walls 103, 104, respectively. In any case, the flows of foam-fiber mixtures are let up through chambers 105, 106, where they are in contact with the mesh conveyors 90, 90A and where they create a vacuum using conventional vacuum chambers 31, 31A.

 In headbox 30V, device 110 is also shown (see FIG. 8). The device 110 is used only for separating volumes 105, 106, as well as for introducing additional substances into the suspension so that the substances do not directly contact the mesh conveyors 90, 90A. This is significant when using certain substances, for example, substances with superabsorbent ability (SAP), since they damage the mesh conveyors 90, 90A in case of contact with them. By supplying through the use of device 110, the injected substances (e.g., SAP) are supplied immediately before the actual formation of the fibrous web, and they are not given the opportunity to contact the mesh conveyors 90, 90A or otherwise interact with the equipment used in the process.

 The internal structure of device 110 (see FIG. 8) contains a plurality of tubes 113, through which additives, such as SAP, from a source 111 with a solids consistency of about 10-20% are fed upward so that they exit through the expanding triangular ends 114 of the tubes 113 Between the tubes 113 with their expanded ends 114, plates 115 may be mounted to hold the tubes 113 in their position. Plates 116 (see Figs. 7 and 8) are installed on opposite sides of the tubes 113 to determine the direction of movement of the foam-fiber mixture in chambers 105, 106. SAP or another substance is discharged, as shown in pos. 117 in FIG. 8, at a point located at least behind the first vacuum chamber 31, 31A and substantially in the middle part of the foam-fiber mixture in a place where it is almost impossible for the substance released at point 117 to directly contact with mesh conveyors 90, 90A.

 The tubes 113 preferably have a circular cross section, while the expanded ends 114 have flat walls and a substantially rectangular opening through which the substance 117 is discharged. The expanded ends 114 occupy substantially the entire upper surface of the wall 116, as shown in FIG.

 Products made using headbox 30V typically contain two or more different layers that form as a unit in the fibrous canvas 33, and if the substance 117 is introduced, it is introduced so that it is essentially between the layers and partially penetrates into each layer .

 Figure 9 shows in section in a direction perpendicular to the longitudinal direction of the machine, taken as an example, the design of the vacuum chamber 31 of the scraper or headbox 30, which is currently (and for many years) used in the manufacture of non-woven materials from fiberglass and which is likely to be used in the manufacture of fibrous canvases 33 in accordance with the method of Alstrom. As shown in FIG. 9, a forming mesh conveyor 90 extends above the upper edge of the vacuum chamber 31 having side walls 118. Holes or tubes 119 are cut into the side walls 118 to allow air to enter the vacuum chamber 31 under the mesh conveyor 90 in addition to foam 120, sucked from a foam-fiber mixture, located on the opposite side of the walls 118 of the mesh conveyor 90. Air freely enters the tubes 119 due to the vacuum created in the vacuum chamber 31 in the usual way. However, the tubes 119 are provided with valves, the passage opening of which is controlled automatically, or at least manually. The foam then passes through a pipe 46 into a pallet 23 under the mesh conveyor. However, since air is sucked in through the tubes 119, it is desirable to remove the excess suction air (but in such a way that the air-liquid ratio in the foam does not substantially change in comparison with the state in the foam-fiber mixture). Therefore, the pipe 121 is connected to the pipe 46 and the air is sucked through the pipe 121 by the fan 122.

 In FIG. 10 schematically shows a vertical web-forming machine comprising a headbox 30V shown in FIG. 1, shown together with other components of the web-forming machine, including various shafts that are used to guide and / or drive the mesh conveyors 90, 90A, as well as a washer 34 and a dryer 36. A distinctive feature of the device shown in FIG. 10, which is of significant importance, is the presence of tubes 124, which are directed to the manifolds 125, in turn, connected to the pipes 46. The tubes 124 are connected to the vacuum chambers 31, 31A. FIG. 11 schematically shows how a plurality of tubes 124 are connected to a manifold 125 and how manifolds 125 are coupled to a pallet 23 under a mesh conveyor.

 11 illustrates one way by which the level of 128 foam in the collectors 125 is controlled. A remote-controlled (for example, using a solenoid) valve 127 is installed in each pipe 46, directed from the collector 125 to the pallet 23 under the mesh conveyor, and this valve is controlled using the controller 129. If the valve 127 is closed or partially closed, the level of the foam may rise in the manifold 125, as shown in FIG. 11. This makes it possible to control the level of foam in the manifold 125. When the valves 127 are fully open, the foam flows freely through pipes 46 into the pallet 23 under the mesh conveyor, leaving below the level of the foam in the pallet itself.

 In all embodiments of the system 10, it is preferable not to use pumps in the pipe 46 to suck the foam; it is better that the foam simply flows freely under its own weight into the pallet 23 under the mesh conveyor.

 12 schematically shows a washer and a coating system that can be introduced into the system 10. The washer fluid is supplied to the washer 34 from the upper side of the canvas 33, and the vacuum chamber 130 is located below and creates a vacuum in it with a fan 131 for removing the wash liquid after it passes through the fibrous web and substantially removes the surfactant from the fibrous web 33. The washer 34 may be of any conventional design, such as is currently used to remove the binder it (using chemicals instead of water) in the production of non-woven materials from fiberglass according to the method of the company "Alstrom".

 The method according to the invention makes it easy to apply additional layers to the fibrous canvas 33 without the need for additional pressure boxes. Although other pressure boxes can be used for these purposes, it is much simpler to use one or more coating plants 35 behind the washer 34 along the course of the product for applying various materials, for example, in the form of a layer 132, applied using a simple coating system 35 . The simple installation 35 for coating is a completely conventional device by which a layer 132 of any other material of the required thickness (which may include another mixture of fibers) is applied over a fibrous canvas 33. Behind the installation 35 for application along the product, i.e. after layer 132 has been applied, a dewatering unit 133 is placed which is brought into contact with layer 132 to remove water from it.

 As usual, the perforated or mesh web 134 guided by the shafts 135 is moved in the same direction as the fiber web 33 under the vacuum chamber 136. In the vacuum chamber 136, excess liquid is removed from the layer 132, while the fibrous web 33 support from below with conventional shafts 137, conveyor belt, etc. It is important that the vacuum chamber 136 is located on the opposite side of the layer 132 from the fibrous canvas 33 in order to properly remove excess fluid. Between the conveyor belt 134 and the shafts 137 (or another conveyor belt), a spin is produced to remove water from the layer 132.

 Following the dehydration unit 133, it is desirable to use a blower 139 as part of a conventional dryer 36 to blow air passing through the pipe 140 through the layer 132 and the fibrous web 33 from above, the pipe 140 being connected to a suction source to facilitate the movement of air from the blower 139 The drying machine 36 may also have other features, which is usually the case.

 Any number of coating plants 35, 35 'can be installed in line with either the dewatering unit 133 combined with each coating system 35, 35' or with a number of coating units located before the dewatering unit 133, depending on the specific number of layers that are applied to the fibrous canvas 33.

 In FIG. 13-16 show various input parameters that are input to controllers 71-74 with a fuzzy algorithm in order to accurately control the system 10, and FIG. 17 shows the relationship of the control system with fuzzy logic with other control systems. The precise control of the system 10 is the main factor that enables the method according to the invention, while others fail in the industrial production of fibrous canvases from cellulose and synthetic fibers and in the production of improved fiber canvases from glass and aramid fibers.

 As shown in FIG. 13, the level of the foam in the mixing tank 11 is controlled by the controller 71 with a fuzzy algorithm. The input parameters entered into the controller 71 with a fuzzy algorithm include the following parameters: the density of the foam (determined either by a density meter 49 installed in the line or densitometer 49A installed in the mixing tank 11, but not both at once), pH acidity, measured by pH meters 15, flow rate of recirculated foam in line 47, determined by the speed of the centrifugal pump 48 (measured by ordinary dstvami), the level 128 of foam in the wire pit 23 and fiber stream conveyor through conduit 12 to mixing tank 11 or other flow parameters. The fiber flow in the pipe 12 is accurately determined using a balance 76, which determines the amount of fiber per unit time supplied to the mixing tank 11.

 On Fig shows the input parameters input to the second controller 72 with a fuzzy algorithm that is used to control the valve 77 and / or unloading the balance 78, or another mechanism that controls the addition of surfactants to the mixing tank 11. Among the input parameters input to controller 72 with a fuzzy algorithm, includes the following parameters: surfactant flow rate, determined, for example, using weights 78, manifold pressure 27 (which is usually 1.0-1.8 bar depending on the product being produced), 128 foam level in pallet 2 3 below the grid conveyor, the pH determined by the pH meters 15, the fiber flow determined using the weights 76, and the flow rate of recycled water in the pipe 61, determined by the flow meter 63.

 On Fig shows the input parameters entered into the third controller 73 with a fuzzy algorithm, which is used to control the ratio of air / foam when forming a fibrous canvas in the head box 30 (for example, control the speed of the conveyor belt or pressure in the head box of the canvas). The input parameters entered into the controller 73 with a fuzzy algorithm include the following parameters: pressure in the headbox 30, foam level 128 in the pallet 23 under the mesh conveyor, the amount of foam removed from the headbox through the vacuum chambers 31, the density of the foam, determined densitometers 49 or 49A, the surface density of the fibrous web 33 (after forming the fibrous web or after the dryer 36) and the level of vacuum in each vacuum chamber 31 (or 31A). The pressure in the headbox 30 is controlled by adjusting the speed of the pump 25.

 On Fig shows the input parameters entered into the fourth controller 74 with a fuzzy algorithm that is used to control the washing machine 34, namely the flow rate of the washing fluid and the suction. The input parameters entered into the controller 74 with a fuzzy algorithm include: the surface density of the fibrous canvas 33, the speed of the fan 131 suction, the pressure in the washer 34, the temperature of the washing liquid and the speed of forming the fibrous canvas (speed of the conveyor belts 90, 90A).

 For a short supply period (of about 45 seconds) from the mixing tank 11 to the headbox 30, the foam-fiber mixture is preferably kept in a high level of mixing / shear modulus. The shear modulus is mainly controlled by controlling the level of the foam in the mixing tank 11, where the foam is mixed with a conventional rotary mixer, pressure drops over the foam nozzles 28, position of the headbox 30, control of the primary drainage, for example, by regulating the vacuum in the slots in the vertical and inclined headboxes 30V, 30I and centrifugal pump speeds 17, 25 and 48. In addition to valves 127 in FIG. 11 (if used), valves are not used throughout the mixing tank 11, especially valves that would dnamerenno contact with the foam. Measure the current on the engine of the recirculation pump 48 and the speed of the recirculation pump 48, if there is a pressure drop on the nozzles 28. If the current on the engine of the recirculation pump 25 changes, while the density of the foam (measured by the density meter 49) remains the same, then then the size distribution of the bubbles has changed. This means that it is necessary to change the supply of surfactants (either by adding more surfactants, or by reducing the added amount of surfactants) through line 13 in order to restore the desired size distribution of the bubbles.

 The multi-parameter controller sets the computer-specific settings to all the controllers 72-75 with a fuzzy algorithm, as shown in FIG. 17, and the neural control network 145 shown in FIG. 17 receives data from the quality determinants and process parameters 149, provides long-term control and forecasting as well as installation parameters.

 In FIG. 17 shows a schematic conventional control system using a neural network 145 operatively coupled to provide and receive data and control signals to and from a conventional multi-parameter control system 146 and fuzzy logic control systems 147, 148. Quality parameters 149 obtained in laboratory tests (which are usually performed outside the production line, for example, determination of foam stability), are introduced into the neural control network so that settings can be obtained for long-term regulation and prediction. An example of one of these measurements is the determination of the stability of the foam, described here below.

 The foam should remain stable and substantially uniform throughout the duration of the process. The stability of the foam is determined by a simple test, usually conducted outside the production line. A liter vessel with graduation applied to the vessel wall is filled with foam to the top and the foam protruding above the upper edge of the vessel is removed. As soon as the vessel is filled with foam, a stopwatch is started. The net weight of the foam in the vessel is determined by weighing (g) and this value is divided by 2. The stopwatch continues to count the time until enough water has settled out of the foam and its level (mm) reaches the value corresponding to the weight on the scale on the wall of the vessel foam divided by 2. (When performing this test, it is assumed that the entire weight of the foam is the weight of water, i.e. that the weight of air is zero.) For example, one liter of foam can weigh 320 g. 320 divided by 2 is 160 When the water level in the vessel reaches a height of 160 mm, the stopwatch is stopped. The optimum stability of the foam is considered to be when it takes approximately 7 minutes for half of the water to settle. If the test time goes beyond the limits of 4-10 minutes, this means that the foam does not have acceptable stability.

 The process of deposition of fibrous canvas from the foam-fiber mixture according to the invention is carried out using the parameters shown in Table I. Although a number of these parameters, such as pH and pressure in the collector, depend on the manufactured products, the values given are the initial values proposed for the manufacture of two materials, which are property of Weyerhaeuser and which are known as the “Single Layer Composition” (USC) and the “Mesh Container Material” (RSC). These Weyerhaeuser proprietary materials are made from a blend of synthetic and cellulosic fibers. Other parameters can be used in the manufacture of materials from glass fibers.

In a complex system according to the invention (for example, a glass fiber non-woven material production line), parameters that can be controlled using a fuzzy logic control system (and / or a multi-parameter control system and / or a neural control network) include:
- Total surface density [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: fiber mass flow, fiber mass moisture, binder suction, binder flow, suction before binder feed, content binder, binder viscosity, binder pH, binder temperature, line speed].

 - The level of the binder in the tank [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: binder feed, binder formula, dry binder content, suction before the binder feed, mesh conveyor speed, binder pH and air content in the binder]. The binder can also be introduced and controlled at the washing stage and at the chemical injection stage.

 - The level in the pallet (23) under the mesh conveyor [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: a level control pump, suction in vacuum chambers (web-binder), pump speed (rpm), energy consumed by a running pump, manifold pressure, headbox pressure, flow rates in suction pipelines and foam density].

 - Level in the mixing tank (11) [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: pressure in the collector, density of the foam in the mixer, pH of the foam, return of foam from the pallet under the mesh conveyor, or foam flow rate in the short circulation circuit, surfactant supply, recycled water flow, foam density in the short circulation circuit, mass flow, level in the buffer tank, level in the pan under the mesh conveyor, stirring energy, foam temperature ].

 - The pressure in the manifold (27) [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: pump speed (25), manifold exhaust valve, pressure in the vacuum chamber of the baler , foam density, foam stability, surfactant supply, level in the mixing tank, level in the pan under the mesh conveyor, pH of the foam, mass flow, recycled water flow, slurry density in the mixer, pressure in the spraying system of the mesh conveyor, water suction control at washing the mesh conveyor, vacuum in the vacuum chamber of the dryer, vacuum at the outlet of the scraper, overflow from the scraper and the temperature of the foam].

 - The density of the foam [having as input parameters at least some (eg, two) of them, and preferably all of the following parameters; surfactant supply, levels in all tanks, temperature, pH, circulating water flow, mass flow, water suction control when washing the mesh conveyor, manifold pressure, system speed and pump and mixing tank energy].

 - The expiration coefficient [having at least some (for example, two) of them as input parameters, and preferably all of the following parameters: pressure in the manifold, foam density, pressure in the headbox, vacuum in all parts of the web-forming machine, mass flow, level in the pallet under the mesh conveyor, the speed of the mesh conveyor, temperature and overflow in the canvas generator].

 - Surfactant supply [having, as input parameters, at least some (for example, two) of them, and preferably all of the following parameters: foam density, foam temperature, mass flow of fiber and foam stability].

 Controllers with a fuzzy algorithm, a neural control network and a multi-parameter control system used in accordance with the invention are conventional commercially available controls supplied, for example, by Honeywell-Alcont.

 Using a multi-parameter control system, the thickness (profile) of the fibrous web is usually measured and dilution is controlled in / or when fed to separate distribution tubes and the set points are set in control systems with fuzzy logic for variable parameters. The neural control network receives data from the means for determining quality indicators and the values of process indicators and provides setting points for long-term management and forecasting. All variable parameters can be changed depending on which factor has a greater influence on the implementation of proper regulation and is most important for the production of finished products.

 FIG. 18 shows schematically various interconnections between components of control systems according to the invention when using neural network 145 (to which values of indicators determined in laboratory conditions are supplied, as shown, for example, at 149 in FIG. 17). Using the neural network 145, three different segments are controlled: the formation of a fibrous canvas, indicated schematically by pos. 150 in FIG. 18; binder, pos. 151 (usually used only when most of the fiber from which the fibrous web is formed is fiberglass, or aramid fiber, or a similar fiber); and a drying system 152. Three basic subsystems are connected to neural network 145: control optimization system 153, horizontal multi-parameter predictive control (NMRS) 154 (conventional multi-parameter controller) and statistical process control system (SPC) 155. The control of the shaper is indicated schematically by pos. 156 with all of the various input parameters and self-monitoring systems associated with it (shown schematically below at 156 in FIG. 18). Similarly, control systems 157 and 158 of the adhesive machine and the dryer, respectively, are shown.

 This means that at the first level of control of the process of forming a fibrous web from a foam-fiber mixture in accordance with the invention, a neural network model 145 is used, which is active in the quality control of essentially the entire production process. Any of versions 1-3: Model-SS, PROP model (with proportional algorithm), or model (ENZO) with evolutionary algorithm, or their combination can be used as the kernel of the operating system, learning algorithm, prediction code, simulation code and optimization code neural network model 145. In addition, newer versions of the above systems can be used, as well as completely new operating system kernels, learning algorithms, prediction codes, simulation codes and optimization codes for the neural network model.

 INPUT PARAMETERS of a neural network model are qualitative process parameters, for example, surface density, fiber weight, binder content, thickness, porosity, tear strength, tensile strength, fiber orientation, tensile strength at high temperature, oil permeability, opacity, wet tensile strength , foam stability, etc., obtained from tests outside the production line (for example, as shown at 149 in FIG. 17), or measurements taken on the line itself.

 Model 145 OUTPUT PARAMETERS are control parameters or process parameter setting values. These include, but are not limited to, the following parameters: fiber feed 145, manifold pressure (27), level in the mixer (11), level in the buffer tank (19), pressure in the circulation pipes (e.g., pumps 25, 48 and / or 62) and vacuum in the vacuum chambers (31) of the web-forming agent.

During testing using a combination of the evolutionary algorithm (ENZO) and the PROP (proportional) algorithm, the surface density of the fibrous canvas with a deviation of less than 1.4 g / m ^{2 is} obtained with a measurement reliability of 95%.

 The thickness of the finished fibrous canvas 33 in the longitudinal and transverse directions at the output of the production line can be controlled using either a controller 154 with a horizontal multi-parameter predictive control system (NMRS) based on measurements made online, or using a predictive multi-parameter controller. These control systems can also be used to control any other part of the process, the control problems of which are too complicated to solve by conventional control methods (PID controllers).

 The control system NMRS 154 is preferably used to control the thickness of the fibrous canvas 33 in the longitudinal direction. A device for controlling the thickness of a fibrous web in the longitudinal direction contains a prognostic multi-parameter algorithm based on simulation. The NMRS controller 154 is a matrix-type control algorithm with many inputs and outputs and is used to predict the steady state of a process by using a specific process model. The NMRS controller 154 also takes into account the boundary states of the drives and the optimizing functions of the adjustable parameters.

 A multi-parameter control device (NMRS 154) takes into account the interaction of parameters that should be controlled (measured parameters that need to be maintained at the set values, for example, surface density, speed and humidity) and process parameters (drive parameters, for example, mass flow rate). Table II shows the model of the control matrix of the NMRS control system for regulating the thickness of the fibrous canvas in the longitudinal direction.

 The control system 154 controls a plurality of output parameters simultaneously and maintains monitored parameters at a desired level. The NMRS controller 154 also takes into account disturbing parameters. Such disturbing parameters are taken into account, for example, when starting up the machine (the entire line) or when it is necessary to change the production (for example, changing the surface density of the produced material). These are measurable parameters that influence the monitored parameters, but they are not controlled by the control system 154. The disturbing parameters can also be used to control perturbation. The control system 154 predicts how disturbing parameters affect the monitored parameters. Forecasts are then used to introduce the necessary corrections into the outputs of the control system 154.

 One of the advantages of control system 154 is the ability to predict the steady state of a process. These prognostic indicators serve for more complete information of the attendants about the future situation. Indicators of the final control situation are also displayed to inform maintenance personnel. The HMPC controller 154 can also predict when the control systems will bring the drive to their operating limits and can adjust the control strategy to suit the situation. The forecasting ability allows you to set complex tasks for the control system, which is why prioritization functions are used to optimize the state of the line. For example, the control system can automatically lower the setting value of the surface density of the fibrous canvas, if you want to increase the binder content for the increased thickness, and the surface density is already at the upper limit.

 In one embodiment of a process for forming a fibrous web from a foam-fiber mixture according to the invention, a 3x3 matrix is used to control the thickness of the web in the longitudinal direction. However, depending on the number of parameters, other types of matrices can also be used, for example, a 10x10 matrix (10 input parameters and 10 output parameters). The controlled parameters can be the amount of fiberglass (or other fibers), the amount of binder (if used) and speed. Output parameters may include fiber feed, binder suction, and speed. During experimental launches of the line, it was possible to reduce the scatter in the values of surface density, fiberglass content and binder content by 50% in the longitudinal direction due to the use of the control system according to the invention in the production of fiberglass canvas from fiberglass.

 The goal of optimization management (153 in FIG. 18) is to reduce the cost of the process to a minimum, increase output to a maximum, or eliminate bottlenecks in production. An example of optimization is interactive optimization of material flow, supply of chemical compounds, energy consumption, solving problems associated with product quality and production volume in each case. The best possible mode of operation of the production line was selected by optimizing the process in accordance with both the set goal and the existing process limitations.

 You can also use the optional Statistical Process Control (SPC) method 155.

 Process control is performed by using fuzzy logic neural networks, PID controllers, or a combination thereof. This means that fuzzy logic, neural networks, PID controllers, or a combination thereof can be used to control the forming section 156, the bonding section 157 and the drying section 158 in the production line to carry out the process of forming a fibrous web from a foam-fiber mixture according to the invention. In the molding section 156, the process can be controlled using fuzzy logic, a neural network, PID controllers, or a combination thereof. In the molding section 156, the following parameters can be controlled: the density of the foam in the pan 23 under the mesh conveyor, the density of the foam in the mixing tank 11, the orientation of the fiber, the level in the mixer 11, the level in the pan 23 under the mesh conveyor, the pressure in the headbox 27, the height of the outlet slit the head box 30 (the thickness of the fibrous canvas at the outlet of the head box 30), the weight of the glass fiber (the mass flow of fiberglass or other fiber is 12, 76 in FIG. 1), pH, level in the buffer tank 19, the surfactant stream 13, the thickness of the fibrous canvas pr molding (suction or outflow in the vacuum chambers 31, 32), the speed of the mesh conveyor 90, the flow density of the supplied pulp (mass flow, kg / min), the pressure in the vacuum chamber 31, the pressure in the manifold 27, the total vacuum in the headbox (flow, rarefaction levels, pressure in the vacuum chamber 31, etc.), rarefaction in the zone of release of the fibrous canvas (after molding the fibrous canvas), suction during drying (after deep vacuuming), the change in pressure drop in the molding zone (ΔР between the upper and lower sides of the mesh conveyor EPA), the thickness of the web 33, and porosity of the fibrous web 33.

 With a further explanation of some of the parameters listed above, the orientation of the fibers can be considered as an indicator characterizing the strength ratio, i.e. the ratio of the strength of the fiber web in the longitudinal direction to the strength in the transverse direction. This ratio is controlled by the speed of the mesh conveyor 90, the flow in the headbox 30, the pressure, the density of the foam and the profile of the curve of the vacuum created in the vacuum chambers 31. The pressure created by the pump, the settling time of the foam, the density of the foam, pH and other factors can also play a certain role. Ultimately, the expiration coefficient, which is the ratio of the speed of the foam and the speed of the mesh conveyor, is calculated. The speed of the mesh conveyor is usually kept constant during the execution of a particular process. Each vacuum chamber 31, 32 is monitored.

 The vacuum at the outlet of the headbox, which is created after molding the fibrous web and the vacuum chambers 31 with normal vacuum, is a pressure that is greater than the pressure in the vacuum chambers. Through this vacuum, the foam is sucked off, as shown at 32 in FIG. 1, and a separator 53 is used to separate the air.

 Suction during drying is usually carried out after suction, for example using vacuum pumps with a liquid ring, for example, Nash pumps, or other high vacuum pumps. This process usually involves the removal of water, which is performed immediately before drying (see key 36 in FIG. 1). The binder bonding section 157 in the production line can be controlled using a fuzzy logic control system, a neural network, a PID control system, or a combination of these systems. You can control at least the following parameters: the binder content in the fibrous canvas 33 (the binder suction during molding after the means for controlling the addition of the binder determine the addition of the binder in zone 157), the pH of the binder, the surface density and the binder suction during molding, level binder in the tank for its circulation, binder temperature and suction rate.

 Drying 158 in a production line can be controlled by using a fuzzy algorithm controller, neural network, PID control system, or a combination of these systems. At least the following parameters can be controlled: the drying temperature at various points along the dryer, the speed of the fiber web 33, the energy consumption supplied to the dryer 36, the humidity in the dryer and the pressure drop (above and below the fiber web 33 at various points along the dryer).

 An example of using fuzzy logic in accordance with the present invention to control the density of the foam is illustrated in Figs. 19 and 20.

 In FIG. 19, a controller 160 for controlling the density of the foam is shown schematically connected to a fuzzy logic control system 161 and a processor 162. A foam density setpoint is an input parameter to the controller 160, while other parameters are input parameters to the controller 161 with a fuzzy algorithm and processor 162, which finally gives the value of the measured foam density. The setting value minus the measured value determines the difference between them. This value is used as shown in the schematic diagram of the foam density control in FIG.

 As shown in FIG. 20, various input parameters 163 are introduced to prepare the task for solving by the methods of fuzzy logic 164 “phaseization”, while laying the basis of the rules 165, providing a solution to the problem by the methods of fuzzy logic at the output 166 to control measurements on the production line. Operations 164-166 together form a controller 167 with a fuzzy algorithm. It is desirable that the input parameters entered at stage 163 contain the following parameters: the measured value of the density of the foam, for example, a density meter 49, the overflow indicator in the buffer tank 19, the pH of the foam (determined, for example, by pH meters 15), the fiber feed (for example, in paragraphs 12, 76 in FIG. 1), the temperature of the foam, the density of the foam in the mixing tank 11, the difference between the setting value of the density of the foam and the value determined when measuring the density of the foam, the flow of recycled water, the viscosity of the foam, the quality of the fiber, the settling time of the foam ( opre divided according to the method described above), water quality (for example, its pressure and hardness), suction after cleaning the mesh conveyor (suction of water from the mesh conveyor), surfactant chemistry (Z-potential), which may depend, inter alia, on the type of surfactant , its pH, the type of fibers used and the hardness of the water, and the feed rate of the surfactant (see pos.13, 77, 78 in Fig.1).

 The density of the foam in the mixer 11 (static pressure + level) or density in a short circulation pipe (measured by density meter 49) is used as the actual value for the controller (164-166) with a fuzzy algorithm. In this example, the density in the short circulation pipe (measured by density meter 49) was used as the actual value. The density of the foam-fiber mixture in the mixer is used as the initial value, and the change in the difference in the density of the foam in comparison with the density of the foam is used as one of the input values 163 of the control system.

 The input and disturbing parameters of a fuzzy logic control system can be improved using a statistical control system (SPC) or a neural network, or a PID system (proportional, integral and differential controllers). In this case, either the value of the measured parameter value is taken as a constant value using the PID control system, or the input value is improved using a neural network at the input.

 At the first stage, the input parameters of a fuzzy logic control system are prepared (164) to be “phased” for solving by fuzzy logic methods. Preparation of a task for solving by fuzzy logic methods can be performed in three or five stages. When preparing a problem for solving by fuzzy logic methods, the numerical value of the parameter is converted to belonging to fuzzy sets, i.e. dimensionless comparable value. In the example, the conversion was performed in such a way that it was easy to switch from a three-stage solution to a five-stage solution to the problem, and vice versa. In FIG. Figure 21 illustrates the principle of converting a measured parameter in a process using fuzzy logic methods into five fuzzy groups. The preparation of the problem for solving by the methods of fuzzy logic (164) is carried out using functions of the level of membership in a fuzzy set, by which the membership of each fuzzy set is determined as a function of the numerical value of the parameter.

 In FIG. 22 schematically illustrates the “phasing” of foam density increment values during a process, for example, functions of the level of membership in fuzzy foam density measurement sets.

VE is the difference in the density of the foam, i.e. VE = SET-MES (setting value minus measured value, see FIG. 19). Levels of membership in fuzzy sets are as follows:
BPO is a very positive value;
POS is a positive value;
ZER - zero value;
NEG is a negative value;
BNE is a very negative value.

The tuning values for regulation are as follows:
FBPO - tuning value for the VRO group;
FPOS - tuning value for the POS group;
FZER - tuning value for the ZER group;
FNEG - tuning value for the NEG group;
FBNE is the tuning value for the BNE group.

 On Fig on the abscissa axis presents the differences in the values of the density of the foam VE (g / l) (VE = SET-MES; values from -30 to +30 g / l are shown), and on the ordinate axis weights (from 0,0 up to 1.0) for functions of levels of membership in fuzzy sets. Against the zero point corresponding to the difference in the foam density values VE = 0 on the abscissa axis, the FZER function is located, represented by two oblique lines starting from the point FZER = 1.0. The lines cross the abscissa axis (FZER = 0) at those points where FNEG = 1 (to the left of VE = 0) and FPOS = 0 (to the right of VE = 0). In the same way, the FPOS function is represented by two oblique lines starting at FPOS = 1. The line on the left intersects the horizontal line (FPOS = 0) at FZER = 0, and the line on the right crosses FBPO = 0. Further, the FBPO function is represented by two lines starting from the point FBPO = 1. The line on the left is oblique and crosses the abscissa axis (FBPO = 0) at the point FPOS = 1, and the line on the right is horizontal at the level of FBPO = 1. A similar description applies to the FNEG and FBNE functions. In other words, FIG. 22 contains five different functions and their corresponding graphical representations. In accordance with Fig. 22, the interpretation of the difference in the foam density values VE = 20 g / l is determined so that a vertical line (dashed line) is drawn at the point on the abscissa axis, where VE = 20 g / l, and through the points where the vertical line intersects the graphical representations of the FBPO and FPOS functions; horizontal (dashed) lines are drawn to the ordinate axis. The intersection points of the horizontal lines and the ordinate axis show that the difference in the density of the foam VE = 20 g / l can be interpreted as very positive at the level of BPO = 0.2 and as positive at the level of POS = 0.8. The value of the remaining membership levels is 0, since the vertical dashed line at the point VE = 20 g / l intersects only the displays of the FBPO and FPOS functions.

 In short, the blocks in FIG. 22 are triangular functions of the level of membership in fuzzy sets, the vertex and angles of which are determined by tuning values, as shown in FIG. 22. Thus, only two levels can simultaneously receive a nonzero value (the sum of two levels must also be equal to one). The setpoints can be either setpoints or maintenance-controlled values. Dynamic values can also be hidden inside the tuning values to perform adaptive functions.

The function of fuzzy logic is determined by creating a rule base (165 in Fig. 20), i.e. function logic. On Fig schematically illustrates the principle of the rule base. The control cycle and other tuning parameters are determined by the responses to step-by-step functional tests and by analyzing the measured parameters. The speed of the control function is determined by its cycle, measurement filtering and control gain. Control gain is achieved on the basis of both input parameters (“phaseization parameters”) when preparing a task for solving by fuzzy methods
logic, and output parameters ("disinfection parameters") when obtaining a solution to a problem using fuzzy logic methods. These factors are determined by the responses to step-by-step functional tests.

 Each of the rules is imposed taking into account the weight coefficient, which is the same as the input membership level mentioned in the conditions section of the rule. For example, the rule "if VE (the difference in the values of the density of the foam) corresponds to NEG" zero ", a VDE (the change in the difference of the values of the density of the foam) corresponds to the" negative "value, and DPY (the change in the overflow in the buffer tank) corresponds to the POS" positive "value, then 01 (dispersion agent control signal) corresponding to ZER “zero” is imposed using the measured membership value corresponding to zero as a weight value from among the values. All other rules are imposed accordingly using x own weight values.The combined influence of the rules is calculated according to the algorithm in accordance with the mutual influence.The output parameters are five values FDN, DN, ZER, UP and FUP, fluctuating between values 0 and 1, defining control output values as membership levels: quickly reduce ( FDN), decrease (DN), zero (ZER), increase (UP) and quickly increase (FUP) in output value groups.

The concept of fuzzy logic is to determine the behavior of control actions as desired reactions in various situations, and not as functions that operate with direct numerical values. In the following example, the input control parameters are the difference in the density of the foam and its rate of change, the rate of change of overflow in the buffer tank. The apostrophe in the designation of the parameter determines the value measured in the following control cycle:
The difference in the density of the foam is determined by the expression
VE = MES-SET,
where VE is the difference in the density of the foam (g / l);
SET - setting value (g / l);
MES is the measured value (g / l).

The change in the difference in the density of the foam is determined by the expression
DVE = VE-VE ',
where DVE is the change in the difference in the density of the foam [(g / l) / h];
VE - the difference in the density of the foam [(g / l) / h];
VE '- the difference in the density of the foam [(g / l) / h], measured in the next control cycle.

The change in overflow in the buffer tank is determined by the following expression:
DPY = PY-PY ',
where DPY is the change in overflow in the buffer tank [(l / s) / min];
PY - overflow in the buffer tank [(l / s) / min];
PY 'is the overflow in the buffer tank, measured in the next control cycle [(l / s) / min].

 Table III shows a number of examples of fuzzy logic control rules, which also present the preparation of the task of determining ("phaseization") the modulation of the overflow in the buffer tank for solving by fuzzy logic methods, but it was not used at the first stage of logic execution in accordance with the rules.

 In FIG. 23 illustrates Table III of the rules used to implement the regulation, the Table of rules reveals the principle of operation of the rule base. FIG. 23 contributes to a better understanding of the logic of the control functions. FIG. 23, the time (h) is shown along the X axis, and the foam density difference (g / l) is greater or less than the target value along the Y axis. On the X axis, the value of the foam density difference (g / l) is 0 g / l, and when moving upward, the foam density difference (g / l) is positive, i.e. above the target value, and when moving down the value of the foam density difference is negative. For example, when, in accordance with rule A and FIG. 23 (at point A), the value of the foam density difference (VE) is negative (NEG, i.e., below the target value) and the rate of change of the foam density difference (DVE) and the rate of change overflow in the buffer tank (DPY) is not considered at all, as a result, the dispensing agent supply chain has a control value corresponding to 01, i.e. lowering (DN), which means that the flow of dispersing agent is reduced, and thus the density of the foam begins to increase in the direction of the target value. According to rule B of “Table III and point B” in FIG. 23, when the foam density difference (VE) value is positive (POS, i.e., above the target value), the rate of change of the foam density difference (DVE) value is positive (POS, which means that the change is towards the heavier foam) and the rate of change of the overflow in the buffer tank (DPY) is positive (POS, which means that more fresh water is constantly entering the system, leading to an increase in foam density), as a result of the dispensing dosing chain the agent receives a fast increase control signal (01) (FUP), which means that the dispersant supply must be increased significantly. The control cycle and other tuning parameters are selected based on responses to step-by-step functional tests and by analyzing the measured data. Filtering of measurement indicators related to the preparation of a task for solving by fuzzy logic methods is determined on the basis of measurement data. The performance of the control function is determined by its cycle, filtering measurements, and control gain. Control gain is achieved on the basis of parameters both when preparing a task for solving by fuzzy logic methods, and when obtaining a solution to a problem by fuzzy logic methods (parameters of "phaseization" and parameters of "dephasing"). These factors are determined by the responses to step-by-step functional tests.

 Table IV shows the rule base presented in Table III in a slightly different form and it consists of some rows and columns taken as an example from a total of 25 rules. The rule base in Table IV does not include overflow changes in the buffer tank, but it is modified in the first phase. The parameters of the first phase are the value of the foam density difference (VE) and the change in the foam density difference (DVE). These conditions determine the control quantity.

 Table IV can be presented in such a way that it can be simpler and easier to interpret. The simplified form of the rule base presented in Table IV is shown in Table V.

Table V illustrates in rows the value of the foam density difference (VE), and in the columns, the rate of change of the foam density difference (DVE). The direction of control (01) is presented in the cells of the intersection of rows and columns. For example, if the measured value of the foam density difference (VE) is presented as a positive value (POS) at the level of 0.2 and a very positive value (BPO) at the level of 0.8, and the measured value of the rate of change of the foam density difference (DVE) is presented as negative value (NEG) at the level of 0.9 and very negative value (BNE) at the level of 0.1, then the following output parameters will be issued as controllers concerning the dispersant supply, (01) in accordance with the rule base of Table V
ZER = 0.2 UP = 0.8
Dn = 0.1 Zer = 0.1
When determining the output values, it should be taken into account that the output values will always be lower values, for example, at the point where the rate of change of the foam density difference (DVE) is NEG = 0.9 and the foam density difference (VE) is POS = 0.2, the control value will be Zer = 0.2.

After that, the final output values of the rule base are determined to obtain a solution to the problem by fuzzy logic methods. Based on the rule base, the following output values are obtained:
(Fast increase) FUP = 0
(Increase) UP = 0.8
(Zero) ZER = 0.2
(Decrease) DN = 0.1
(Fast decrease) FDN = 0
In the example, as a result of the action of the logic, we obtain that Zer will have two values: Zer = 0.2 and Zer = 0.1, of which the larger value will always be effective, i.e. Zer = 0.2.

 In addition, the overflow rate in the buffer tank (DPY) will be taken into account. Examples of a rule table converted into two phases are presented in Table VI.

 In accordance with rule 22 01 = FUP (foam density - control signal for a rapid increase), and DPY = POS (positive change in the buffer tank overflow), then 02 = FUP (foam density - control signal for a quick increase). The group of control output values (02), which should be obtained when solving the problem using fuzzy logic methods (“dephased”), consists of a rule base in this case, including control (01) and changing the overflow speed in the buffer tank (DPY). If necessary, other rules can also be added to the basic rules of Tables III-VI, including foam temperature, pH, fiber feed, or foam viscosity. The “conditions” in Table VI may instead of “AND” be other normal logical operations, for example, “IF,” or “OR,” or “AND / OR.”

 In the last control phase, using fuzzy logic, precise digital values are assigned for monitoring, which can be used as input parameters of the drive. This is achieved in accordance with the principle of solving the problem by fuzzy logic methods (166 in Fig. 20), which is illustrated in more detail in Fig.24.

The algorithm for solving the problem by fuzzy logic methods may be an algorithm similar to that illustrated in FIG. 25. A weighted column is presented for each fuzzy group of output value. The column height is 1, and its location along the X axis can be adjusted. When calculating the instantaneous output value, the height of the weighted value of each column is presented on a scale according to the corresponding membership levels of the input values, which are represented by the bold lower parts of the columns. The numerical output value is the projection of the total center of gravity of the weight values of the columns shown on a scale on the X axis. For example, in FIG. 25:
Output Parameters:
01 or 02 is the numerical output of the dispersant agent. The membership levels of the input values are the same as previously defined. It means:
VRO - "very positive value";
POS - "positive value";
ZER - "zero value";
NEG - "negative value";
BNE is a "very negative value".

Trim values:
FBPO - tuning value for the set "very positive value";
FPOS - tuning value for the set of "positive value";
FZER - tuning value for the set "zero value";
FNEG - tuning value for the set of "negative value";
FBNE is the tuning value for the set "very negative value".

 Perturbation control, which determines the output parameters of the control action, is used to control the control circuit. Perturbation control changes the output values in increments of 0.1% of the control unit's control range, or with the desired step of decreasing or increasing. Using fuzzy logic, the direction and rate of change of the perturbing parameter are determined and the control effect on the density of the foam is regulated by the base of fuzzy logic rules. The control diagram shown in FIG. 20 illustrates the operating principle of a disturbance control in fuzzy logic control. When controlling the density of the foam, the following disturbing parameters are used as control factors for the perturbation: foam density, measured density difference, overflow in the buffer tank, pH of the foam, fiber feed, foam temperature, change in the difference in the density of the foam-fiber mixture in the mixer and the density of the foam, the reverse flow water, foam viscosity, system activity, fiber quality, foam deposition half-cycle, water quality (pressure, stiffness), mesh conveyor suction, surface chemistry (Z-potential) and the amount of dispersing agent.

 In the method of forming a fibrous web from a foam-fiber mixture according to the invention, the control of the web-forming agent (156) is preferably performed using a neural network 170, 145 to control the orientation of the fibers. The principle of operation of the neural control network 170, 145 is to achieve a more stable orientation of the fibers, which can also be easily duplicated. The neural network 170, 145 is used to control the suction when forming in the longitudinal direction of the material (vacuum level), which is used to control the orientation of the fibers in the fibrous canvas 33. When measuring the upper vacuum chamber of the binder is also used as standards. The level of suction in the longitudinal direction of the material is controlled using the height of the slit at the outlet of the headbox. Regulation of the suction level in the longitudinal direction of the material is carried out either using the controller 167 with a fuzzy algorithm (see Fig. 20) or the PID controller 171.

 It turned out to be possible to reduce fluctuations in the density of the foam by half thanks to the use of fuzzy logic control in comparison with the control of the PID system that was used earlier. In addition, line control during start-up and shutdown has been greatly improved. Currently, the process can be set to steady state for an hour sooner than was previously possible.

 Thus, it is obvious that in accordance with the present invention, the process of forming a fibrous canvas from a foam fiber mixture in the production of nonwoven materials is controlled in a very successful manner so that it allows the production of a very wide quality raw material of nonwoven materials using various types of fiber or fiber mixtures, including fillers or binders, if necessary or desirable, and with optimal foam stability and final uniformity and strength of the fibrous canvas. Although the invention has been presented and described in the form that is currently considered the most practical and preferred embodiment of the invention, it is obvious to those skilled in the art that many modifications can be made within the scope of the invention, the scope of the invention giving a broad interpretation of the attached claims, so as to cover all equivalent methods and systems.

Claims (29)

 1. A method of manufacturing a non-woven material from cellulosic, synthetic or glass fibers, comprising the steps of: (a) mixing cellulosic, synthetic or glass fibers, water, air, recycled foam and a surfactant in a mixing tank (11) for preparing foam fiber mass; (b) filling the foam into the scrim; (c) molding control; (d) molding the fibrous web in the web former at a rate equal to the feed rate by removing liquid and foam from the mass in the web former and collecting at least a portion of the removed fluid and foam in a tray (23) under the mesh conveyor; (e) additionally affecting the fibrous web formed in the web former to form a finished nonwoven material, characterized in that it comprises step (f), wherein at least one of steps (a) to (e) is performed using a controller (71, 72, 73, 74, 75) with a fuzzy algorithm.  2. The method according to p. 1, characterized in that stage (a) is carried out partially by controlling the mass level in the mixing tank (11), and stage (f) is carried out partially by automatically controlling the level in the mixing tank (11) using the controller ( 71) with a fuzzy algorithm, the input parameters of which are the density and speed of the foam flow, which is repeatedly supplied to the mixing tank (11) from the pan (23) under the mesh conveyor, the pH of the foam in the tank (11), the level of foam in the pan (23) under mesh conveyor and if the amount of fiber added to the tank (11).  3. The method according to p. 2, characterized in that stage (a) is additionally performed by automatically controlling the amount of surfactant added and by recirculating a certain amount of water removed from the fibrous web during its formation and separated from the air, and stage (f) is performed partially for automatic control of the amount of surfactant added, using a controller (72) with a fuzzy algorithm, the input parameters of which are the surfactant flow rate, pressure in the collector (27) of the web forming agent, and the level of foam in Oddone (23), the wire pit, the flow rate of added fiber, and the flow rate of the recirculating water.  4. The method according to p. 3, characterized in that a canister is used comprising a movable mesh conveyor (90) and a headbox (30) and stage (c) is performed at least partially to automatically control the air / foam ratio in the canister, including the speed of the mesh conveyor in the canvas former and the pressure in the headbox (30), and stage (f) is carried out partially by using a controller (73) with a fuzzy algorithm, the input parameters of which are the surface density of the formed fiber empty canvas, the pressure in the headbox, the level of foam in the pan (23) under the mesh conveyor, the density of the recirculated foam and the amount or speed of the foam removed from the headbox.  5. The method according to p. 4, characterized in that stage (e) is performed for washing the fibrous canvas and removing liquid from the fibrous canvas during or in connection with washing, and stage (f) is carried out partially to automatically control stage (e) with using a controller (74) with a fuzzy algorithm, the input parameters of which are the speed of forming the fibrous web, the surface density of the fibrous web, the temperature of the washing liquid, the speed of the suction foam and the pressure on the washer (34).  6. The method according to p. 5, characterized in that at least two parameters from each of the mentioned input parameter groups contain all of the mentioned input parameters.  7. The method according to p. 1, characterized in that stage (a) is additionally performed by automatically controlling the amount of surfactant added and by recirculating part of the water removed from the fibrous web during molding and separated from the air, and stage (f) is performed partially for automatic control of the amount of surfactant added, using a controller (72) with a fuzzy algorithm, the input parameters of which are the surfactant flow rate, pressure in the collector (27) of the web-forming agent, the level of foam in the pan (23) under the mesh by the conveyor, the flow rate of the added fiber and the flow rate of recycled water.  8. The method according to p. 1, characterized in that a canister is used comprising a movable mesh conveyor (90) and a headbox (30), wherein step (c) is performed at least partially to automatically control the air / foam ratio in the canister, including the speed of the mesh conveyor in the canvas former and the pressure in the headbox (30), and stage (f) is carried out partially by using a controller (73) with a fuzzy algorithm, the input parameters of which are the surface density of the formed fiber blank canvas, pressure in the headbox, the level of foam in the pan (23) under the mesh conveyor, the density of the recirculated foam and the amount or speed of the foam removed from the headbox (30).  9. The method according to p. 1, characterized in that stage (e) is performed for washing the fibrous canvas and removing liquid from the fibrous canvas during or in connection with washing, and stage (f) is performed partially to automatically control stage (e) by use of a controller (74) with a fuzzy algorithm, the input parameters of which are the speed of forming the fibrous web, the surface density of the fibrous web, the temperature of the washing liquid, the speed of the suction foam and the pressure on the washer (34).  10. The method according to p. 1, characterized in that stage (e) is performed to dry the fibrous web, and most of the fiber added during stage (a) is glass fiber, to which a binder is added, and stage (f) is performed partially to control the drying process of the fibrous canvas and add a binder, using controllers with a fuzzy algorithm.  11. The method according to p. 1, characterized in that it comprises the step of using a neural control network (145) to implement quality control over essentially the entire process of manufacturing a nonwoven material.  12. The method according to p. 11, characterized in that stage (e) is performed to dry the fibrous canvas, and most of the fibers added during stage (a) are glass fibers to which a binder is added, and stage (f) is performed partially to control the drying process of the fibrous canvas and add a binder using controllers with a fuzzy algorithm.  13. The method according to p. 11, characterized in that a canister is used comprising a movable mesh conveyor (90) and a headbox (30), and stage (c) is performed at least partially to automatically control the air / foam ratio in the canister, including the speed of the mesh conveyor in the canvas former and the pressure in the headbox (30), and stage (f) is performed in part by using a controller (73) with a fuzzy algorithm, the input parameters of which are the surface density of the molded fiber of the canvas, the pressure in the headbox, the level of foam in the pan (23) under the mesh conveyor, the density of the recirculated foam and the amount or speed of the foam removed from the headbox.  14. The method according to p. 11, characterized in that stage (a) is carried out partially by controlling the mass level in the mixing tank (11), and stage (f) is performed partially to automatically control the level in the mixer (11) using the controller (71 ) with a fuzzy algorithm, the input parameters of which are: the density and speed of the foam flow, which is repeatedly supplied to the mixing tank (11) from the pallet (23) under the mesh conveyor, the density of the foam in the mixing tank (11), the pH of the foam in the mixing tank (11 ), the foam level in p oddone (23) under the mesh conveyor and the amount of fiber added to the mixing tank (11).  15. The method according to p. 14, characterized in that at least two input parameters contain all of the mentioned input parameters.  16. The method according to p. 11, characterized in that stage (a) is carried out partially to accurately control the pH in the mixing tank (11), using a variety of pH meters to determine the pH, and stage (f) is partially performed using a fuzzy controller an algorithm for controlling and coordinating pH meters.  17. A method of manufacturing a nonwoven material from cellulosic, synthetic or glass fibers, comprising the steps of: (a) mixing cellulosic, synthetic or glass fibers, water, air, recycled foam and surfactant in a mixing tank (11) for preparing foam fiber mass; (b) filling the foam into the scrim; (c) controlling the molding process; (d) molding the fibrous web in the web former at a speed equal to the feed rate by removing liquid and foam from the foam in the web and collecting at least a portion of the removed fluid and foam in a tray (23) under the mesh conveyor; (f) additional exposure to the fibrous canvas molded in the binder to obtain a finished nonwoven material, characterized by stage (f), which consists of using controllers with a fuzzy algorithm, which control at least the level in the pallet under the mesh conveyor, the level in the mixing tank, pressure in the collector of the canvas, the density of the foam and the coefficient of expiration.  18. The method according to p. 17, characterized in that stage (f) is additionally performed to control the supply of surfactants and the total surface density of the produced non-woven material.  19. The method according to p. 18, characterized in that the binder is also added in the manufacture of non-woven material containing at least 10% glass or aramid fibers, and the binder is added to the reservoir for the binder, and stage (f) is also performed to control the level in the binder tank.  20. A system for the production of non-woven material from cellulosic, synthetic or glass fibers, comprising a mixing tank (11) for mixing cellulosic, synthetic or glass fibers, water, air, recyclable foam and surfactants for the preparation of foam-fiber mass; a web for forming a fibrous web at a speed equal to its feed rate by removing liquid and foam from the foam and collecting at least a portion of the removed liquid and foam in a pan (23) under the mesh conveyor; a pump (17, 25) for pouring foam material from the mixing tank (11) into the canvas, a means for additionally acting on the fibrous canvas molded in the canvas, to obtain a finished non-woven material, characterized in that there are many controllers with a fuzzy algorithm, including at least one controller with a fuzzy algorithm for automatically controlling the level of foam in the mixing tank (11).  21. The system according to p. 20, characterized in that it contains a controller with a fuzzy algorithm for automatically controlling the level in the mixing tank (11), the input parameters of which are the density and speed of the foam flow, which is repeatedly supplied to the mixing tank (11) from the pallet ( 23) under the mesh conveyor, the pH of the foam in the mixing tank (11), the level of foam in the pan (23) under the mesh conveyor and the amount of fiber added to the mixing tank (11).  22. The system according to p. 20, characterized in that the controllers with a fuzzy algorithm are configured to control at least the level in the pallet under the mesh conveyor, the pressure in the collector of the canvas, the density of the foam and the expiration coefficient.  23. The system according to p. 22, characterized in that the controllers with a fuzzy algorithm are configured to control the supply of surfactants and the total surface density of the produced non-woven material.  24. The system according to p. 20, characterized in that the binder is added in the manufacture of non-woven material containing at least 10% glass or aramid fibers, and the binder is added to the reservoir for the binder, the system contains a controller with a fuzzy algorithm for controlling the level in reservoir for the binder.  25. The system according to p. 20, characterized in that the canister contains a movable mesh conveyor (90) and a headbox (30), moreover, one of the controllers with a fuzzy algorithm contains a controller with a fuzzy algorithm for automatically controlling the air / foam ratio when feeding into the canister , including the speed of the mesh conveyor in the scrim and the pressure in the headbox (30), while the input parameters of the controller with a fuzzy algorithm are at least two parameters from the following: tnost molded fibrous web, the pressure in the head box, the foam level in the sump (23) under the wire pit, the density of the recirculating foam, and the amount or rate of foam being removed from the headbox.  26. The system according to p. 20, characterized in that the means for further processing the molded fibrous canvas contains a means for washing the fibrous canvas and removing liquid from the fibrous canvas during or in connection with washing, and one of the controllers with a fuzzy algorithm is configured to automatically control means for washing and removing liquid, and the input parameters of this controller with a fuzzy algorithm are the speed of formation of the fibrous canvas, surface density s batt temperature of the washing liquid, the speed of the foam sucked and the pressure in the means for washing.  27. The system according to p. 20, characterized in that the means for additional processing of the molded fibrous canvas contain a dryer for drying the fibrous canvas, and one of the controllers with a fuzzy algorithm is configured to automatically control the dryer, and the input parameters of this controller with a fuzzy algorithm are at least two of the following parameters: installation temperature of drying, speed of transportation of fibrous canvas, the amount of energy supplied to the dryer machine, the moisture level in the dryer and the pressure difference above and below the fleece web at various points along the dryer.  28. The system according to p. 20, characterized in that there is a neural control network for at least partial interaction with controllers with a fuzzy algorithm for controlling the formation of a fibrous canvas.  29. The system according to p. 20, characterized in that there is a neural control network for implementing quality control, essentially throughout the system for the manufacture of non-woven material.

Images