Online monitoring of sinter quality

Abstract

The present invention relates to a method (100) and a system (20) for monitoring a sintering process, and to a method (200) for calibrating such a system (20). The following steps are carried out: i) detecting (S1), by sensor, an impact of a sinter cake (18) on an impact surface (8a) and generating corresponding sensor data (D); ii) ascertaining (S2) an impact characteristic (C) on the basis of the generated sensor data (D); iii) ascertaining (S3) a quality of the sinter from the sinter cake (18) on the basis of the ascertained impact characteristic (C); and iv) outputting (S4) the ascertained sinter quality by means of an interface (28).

Description

[0001] 202400276 1

[0002] Description

[0003] Online monitoring of sinter quality

[0004] field of technology

[0005] The present invention relates to a method and a system for monitoring an iron ore sintering process.

[0006] State of the art

[0007] For the operation of blast furnaces, the iron ore used must be in granular form. The granular material should be largely homogeneous, and the grain size should be as uniform as possible within a predetermined range to allow for gas flow through the bed of granules formed in the blast furnace.

[0008] Such iron ore granules are typically produced using sintering plants, in which iron ore is sintered together with additives to form stable bodies with specific properties. The resulting sinter cake is then broken up to produce the granules, for example, by a crushing device known as a spike crusher.

[0009] However, the quality of the sinter produced in this way can vary, as the sintering process is a complex one with many input parameters. For example, unintentional or unnoticed changes in the raw material mix (i.e., the mixing ratios), raw materials with varying properties, planned and / or unplanned downtime of the sintering plant, and / or poorly chosen process setpoints such as desired material mix, water addition, conveying speed, cooling capacity, and / or the like can affect the sinter quality.

[0010] To characterize the produced sinter, fragments of the sinter are typically taken and sampled in the laboratory. This allows for the determination of parameters such as the tumbler index or the harmonic diameter of the sinter. Samples from several different sinter cakes are usually mixed and analyzed as a representative composite sample. Since sampling is only possible after the sinter has cooled, which can take between 1 and 5 hours depending on the equipment, and the composite sample is usually compiled over an averaging period of 4 to 8 hours, the results of the analysis are typically only available 2 to 13 hours after the actual production. Therefore, any quality changes in the finished sinter detected during this analysis can only be addressed with a significant delay.

[0011] From Michael et al: "Development of an automated single particle impact tester for iron ore sinter", Minerals Engineering, Vol. 175 (2022), a highly automated test rig for rapid particle impact tests with integrated fragment analysis is known.

[0012] EP 3 536 814 B1 discloses a method for continuously determining the particle size distribution of granules consisting of solid particles with different particle sizes. The solid particles are transported in a conveying stream and collide with at least one impact body configured as a waveguide, generating acoustic signals that propagate through the impact body as structure-borne sound waves.

[0013] EP 4 492 051 A1 concerns the determination of the hardness of cement clinker in a cement plant by evaluating an acoustic signal when the cement clinker impacts a surface after leaving a rotary kiln.

[0014] Summary of the invention

[0015] Against this background, it is an object of the present invention to improve the monitoring of an iron ore sintering process, in particular to be able to provide information relating to the sintering quality in a timely manner.

[0016] This problem is solved by a method and a system for monitoring an iron ore sintering process according to the independent claims.

[0017] Preferred embodiments of the invention are the subject of the dependent claims and the following description.

[0018] According to a first aspect of the invention, the following steps are carried out in the method for monitoring an iron ore sintering process for the production of iron ore granules for blast furnace operation, which is preferably at least partially computer-implemented: i) sensorial detection of an impact of an iron ore sinter cake on an impact surface and generation of corresponding sensor data; ii) determination of an impact characteristic based on the generated sensor data; ill) determination of the quality of the sinter from the sinter cake based on the determined impact characteristic; and iv) output of the determined sinter quality, e.g. via an interface.

[0019] A sinter cake within the meaning of the present invention preferably refers to input materials, for example metal ore, fuel (such as coke dust), and calcium silicate brick, that have been at least partially baked together by heat treatment. Such a sinter cake can be produced, for example, by baking a mixture of the input materials, also referred to as a material mix, on a sinter belt consisting of several so-called grate cars, by igniting the fuel in the mixture. A vacuum system under the sinter belt ensures that the mixture burns completely from the top down.

[0020] Sinter quality, or simply sinter quality, as defined in the present invention, is a measure of the quality of the sinter under consideration, either absolute or relative. Sinter quality can therefore provide a quantitative or a qualitative statement about the quality of the sinter. For example, sinter quality can be specified absolutely by a characteristic parameter such as the tumbler index or the harmonic diameter. Alternatively, sinter quality can also be specified relative to an average or other sinter quality, for example, as an increase or decrease.Determining the quality of the sinter from the sinter cake based on a determined impact characteristic does not necessarily result in an absolute, stand-alone measure of the sinter's quality; rather, it is also possible that a sinter quality determined in this way only allows a statement about the (relative) quality of the sinter under consideration when compared to a sinter quality determined in the same way, for example at a later time.

[0021] Sinter quality can be specified, for example, in the form of the so-called harmonic diameter. The harmonic diameter summarizes the grain size distribution of sinter in a single numerical value that is closely related to the gas permeability, which is important for the blast furnace: with

[0022] DH Harmonic Diameter

[0023] Xi proportion of grain size i

[0024] Di Mean diameter of grain size i

[0025] Alternatively, the sintering quality can be specified in the form of the so-called tumbler index. The tumbler index is a relative measure of a material's resistance to breakage or disintegration due to impact loading and is defined in ISO 3271.

[0026] One aspect of the invention is based on the approach of determining the sintering quality by evaluating the impact of a body made of sintered material on an impact surface. It has been shown that the material properties of the sinter can influence how the body behaves upon impact. Depending on the material properties, for example, the size and / or number of pieces into which the body shatters upon impact can vary, as can the amount of kinetic energy of the falling body transferred to the environment, for example, to the impact surface and / or the surrounding air. The impact, namely the vibrations generated in the impact surface and / or the sound waves emitted into the surrounding air, is detected by means of a sensor device. The sensor data generated in this way allow for a data-driven characterization of the impact.Based on these impact characteristics, conclusions can be drawn about the sintering quality. In other words, the sintering quality can be derived from the impact characteristics. A statement about the sintering quality can thus be made immediately after the sintering process, especially before the sintered body has cooled. At the same time, there is no need to wait for the results of a laboratory analysis. Consequently, the sintering process can be adjusted promptly if the sintering quality is deemed unsatisfactory. This prevents the unnecessary production of a large quantity of inferior sintered material. Quality losses can be detected early and countermeasures can be taken.

[0027] In this respect, it is also advantageous to output the sinter quality determined based on the assessed impact characteristics via an interface, for example, to the operators of the sintering plant. The operators can then initiate countermeasures if the sinter quality deteriorates. Alternatively, it is also conceivable that the sinter quality is output directly to a control system of the sintering plant, which is then preferably configured to control the sintering process based on the determined sinter quality.

[0028] Such an analysis of the sinter based on sensor data generated by the sensory detection of an impact of sinter cake on an impact surface is also possible without great effort because the sinter cake is discharged from the grate wagons at the end of the sintering belt. The sinter cake typically lands on a sinter chute, which feeds the sinter cake, or at least the fragments produced by the impact on the sinter chute, to a crushing device, for example, a spike crusher. This impact on the sinter chute can already be used to determine the sinter quality. The impact surface can therefore be formed by the sinter chute itself. The method according to the invention can thus be seamlessly integrated into the already established production process for iron ore granules, as used in a blast furnace.

[0029] Preferred embodiments of the invention and their further developments are described below. These embodiments can be combined with each other and with the aspects of the invention described below, unless expressly excluded.

[0030] In a preferred embodiment, the impact is detected by means of an acoustic sensor, particularly in a frequency range of 30 kHz to 50 kHz. In this case, the acoustic sensor is advantageously a microphone. The detection of sound waves, especially in the high-frequency range, is an easy-to-establish and industrially robust measurement method. With acoustic detection of the impact, interference with the measurement caused by adverse environmental conditions prevailing during the sintering process, such as dust and heat, can be largely eliminated or at least minimized. In particular, it is also possible to place an acoustic sensor in a protected location near the impact surface without reducing the sensitivity of the acoustic measurement.

[0031] In a further preferred embodiment, vibrations of the impact surface and / or a component connected to the impact surface, generated by the impact, are detected by means of a vibration sensor, particularly in a frequency range of 500 Hz to 10 kHz. In this case, the vibration sensor is expediently designed as a structure-borne sound transducer. Thus, structure-borne sound generated by the impact, i.e., vibrations propagating in the impact surface, can be detected. Particularly advantageous, for example, is the detection of vibrations from a sintering chute or other conveying device that transports the sinter cake or its fragments to the comminution device. It has been shown that the intensity of the vibrations or structure-borne sound is a good indicator of the strength of the sintered material. Compared to acoustic measurement, the detection of vibrations or structure-borne sound has several advantages.Structure-borne sound offers the advantage of better decoupling from the environment. Consequently, when recording vibrations or structure-borne sound, the impact characteristics can be determined even more precisely, and the sintering quality can be assessed even more reliably.

[0032] According to the invention, an impact intensity is determined based on the generated sensor data. This allows for the determination of the impact force, for example, the strength of the vibrations of the impact surface (or a component connected to it) or sound waves. The determined impact intensity then forms the basis for quality assessment. For example, the amplitude, particularly the maximum amplitude, of the sound waves generated during the impact (i.e., the volume), or the amplitude, particularly the maximum amplitude, of the vibrations generated during the impact of the impact surface or the component connected to it (i.e., a maximum structure-borne sound amplitude) can be determined. It has been shown that the impact intensity correlates with the sintering strength. Consequently, the sintering strength is determined based on the calculated impact intensity. A high impact intensity indicates a hard sinter, while a low impact intensity indicates a "soft" or less resilient sinter.indicates less durable sinter.

[0033] In a further preferred embodiment, a parameter characterizing the impact, in particular the impact intensity, for example a maximum sound or vibration amplitude, is determined for several, especially successively dropped, sinter cakes based on the sensor data generated in each case. Advantageously, this parameter, e.g., the impact intensity, is averaged over a predetermined period. The predetermined period expediently extends over the impact of the several sinter cakes onto the impact surface. Subsequently, the sintering quality can be determined based on the averaged parameter. Thus, an average value for the impact intensity or the parameter characterizing the impact, such as the maximum sound or vibration amplitude, can be determined.It is advisable to average the impact events occurring within the specified time period, for example, between 10 and 30 minutes. The resulting average value conveniently corresponds to the impact characteristics. Averaging over several drops allows for a reliable, medium-term assessment of the sintering quality.

[0034] In another preferred embodiment, granules are produced from the sinter cake by further breaking up and sieving the sinter cake or its fragments after impact, for example, using a comminution device. The particle sizes in the produced granules can then be measured. In this way, the particle size distribution in the produced granules can be determined. The sinter quality is expediently also determined based on the determined particle size distribution. By taking into account the particle size distribution in the produced granules, which is also influenced by the properties of the sintered material, a more reliable assessment of the sinter quality can therefore be made.For example, the harmonic diameter of the sinter can be determined based on the grain size distribution and then compared with the harmonic diameter determined based on the impact characteristics. A weighting can also be applied, for example, based on an error determined during the respective harmonic diameter calculation.

[0035] In a further preferred embodiment, the sintering quality is determined by a machine-learned model based on the determined impact characteristics. This machine-learned model is advantageously trained on a training dataset in which a multitude of impact characteristics are each linked to at least one parameter that characterizes the sintered material. This characteristic parameter can be, for example, the so-called tumbler index and / or a harmonic diameter of the sintered material. This characteristic parameter can be determined by laboratory analysis. This allows different impact characteristics to be assigned to different sintering qualities. Using such a machine-learned model, an automated and particularly reliable assessment of the sintering quality is then possible.

[0036] A machine-learned model within the meaning of the invention is preferably a statistical model generated by one or more algorithms based on training data. The machine-learned model can therefore also be referred to as a trained model. The machine-learned model is expediently the result of machine learning. Such a machine-learned model can also be commonly referred to as artificial intelligence, which has recognized patterns and regularities in the training data and can thus also assess or evaluate unknown data according to these patterns and regularities. The machine-learned model can, for example, be based on a neural network, a random forest, a support vector machine (SVM), or a decision tree.

[0037] In a further preferred embodiment, a measurement signal corresponding to the sensor data generated during impact detection is smoothed by applying a mean-value filter. The impact characteristic is then expediently determined based on the smoothed measurement signal. Smoothing ensures that, when evaluating the impact, for example, when determining the impact characteristic, the entire signal characterizing the impact is always taken into account, even if only individual parts of the measurement signal are used in the evaluation. If, for example, the impact intensity—i.e., the maximum amplitude of the detected sound waves or vibrations—is used as the basis for determining the impact characteristic, smoothing with the mean-value filter ensures that the remaining signal components are also included in the impact characteristic.The measured impact intensity can therefore be lower, for example, due to the application of the mean value filter, than would initially be assumed based solely on the maximum amplitude. Consequently, a more differentiated analysis of the sensor data or signal is possible, which in turn allows for a more nuanced assessment of the sintering quality.

[0038] According to a second aspect of the invention, a system for monitoring an iron ore sintering process for the production of iron ore granules for blast furnace operation comprises: i) a sensor device arranged in the area of ​​an impact surface for iron ore sinter cakes to detect an impact of an iron ore sinter cake on the impact surface, such that vibrations and / or sound waves generated by the impact on the impact surface can be detected by means of the sensor device; ii) a first evaluation module configured to determine an impact characteristic based on sensor data generated by the sensor device when the impact is detected, wherein the first evaluation module determines an impact intensity based on the sensor data;iii) a second evaluation module, which is set up to determine a sinter quality based on the determined impact characteristics, wherein the second evaluation module determines a sinter strength based on the determined impact intensity; and iv) an interface for outputting the determined sinter quality.

[0039] With such a system, an assessment of sinter quality can be made immediately after the sintering process. In particular, there is no longer any need to wait for the results of a laboratory analysis. The system can be easily integrated into an existing sintering plant without affecting the sintering process or the subsequent processing of the sinter. The system allows sintering plant operators to be informed early, especially immediately after sinter production, about any deterioration in sinter quality, enabling timely countermeasures. 202400276 8

[0040] The sensor device can be configured as an acoustic sensor, for example, a microphone. The sensor device is advantageously designed to detect sound waves generated when the sintered cake impacts the impact surface, for example, in the frequency range of 30 kHz to 50 kHz. Alternatively, the sensor device can be configured as a vibration sensor, i.e., a structure-borne sound transducer. The vibration sensor is advantageously designed to detect vibrations, i.e., structure-borne sound, generated when the sintered cake impacts the impact surface, or of a component coupled to it, for example, in the frequency range of 500 Hz to 10 kHz.

[0041] A module according to the present invention can be configured as hardware and / or software. In particular, the module can comprise a processing unit, preferably connected to a storage and / or bus system via data or signals. For example, the module can comprise a microprocessor unit (CPU) and / or one or more programs or program modules. The module can be configured to execute instructions implemented as a program stored in a storage system, to acquire input signals from a data bus, and / or to output signals to a data bus. A storage system can comprise one or more, in particular different, storage media, especially optical, magnetic, solid-state, and / or other non-volatile media. The program can be configured such that it at least partially embodies the methods described herein.is capable of executing, so that the module can perform at least some of the steps of such processes and thus, in particular, monitor a sintering process.

[0042] It is preferred that the first and / or the second evaluation module be implemented as algorithms, in particular machine-learned models. It is also conceivable that the first and second evaluation modules are formed by a single algorithm, in particular a single machine-learned model.

[0043] According to a further preferred embodiment of the method, the following additional steps are performed: iii) determining the quality of the sinter from the sinter cake, in particular independently of determining the impact characteristic; and iv) assigning the determined sinter quality to the determined impact characteristic. The quality of the sinter from the sinter cake can be determined, for example, by means of a conventional laboratory analysis, in which, for instance, the tumbler index or the harmonic diameter of the sinter is determined. This can, for example, be used to train the second evaluation module, which is designed as a machine-learned model.

[0044] In a preferred embodiment, the steps mentioned in connection with the method are performed several times. The sensor device can be positioned at different positions relative to the impact surface each time. For each iteration of the steps mentioned in connection with the method, a quality factor for the correlation of the determined impact characteristic with the corresponding determined sintering quality can then be determined. Based on the determined quality factor, the sensor device can thus be finally positioned relative to the impact surface. This ensures that the impact characteristic can be reliably determined at the final position of the sensor device.For example, it can be ruled out that the detected vibrations are only inadequately detected due to poor coupling of the sensor device to the impact surface, or that the detected vibrations are impaired, for example distorted, by the coupling.

[0045] Brief description of the drawings

[0046] The properties, features, and advantages of this invention described above, as well as the manner in which they are achieved, will become clearer and more readily understandable in connection with the following description of an exemplary embodiment, which is explained in more detail in conjunction with the drawings. These drawings show:

[0047] FIG 1 shows an example of a sintering plant with a system for monitoring a sintering process,

[0048] FIG 2 shows an example of sensor data from a sensor device designed as a vibration sensor and

[0049] FIG 3 shows an example of the correlation of an impact characteristic with a parameter characteristic of the sintering quality.

[0050] Where appropriate, the same reference numerals are used in the figures for the same or corresponding elements of the invention.

[0051] Description of the embodiments

[0052] FIG 1 shows a sintering plant 1 for producing a sintered product, in particular a granulate 2 made of sintered material, which is suitable, for example, for use in a blast furnace. The sintering plant 1 comprises, in addition to a sintering belt 6 consisting of several grate carriages 4 for carrying out a sintering process and a sintering chute 8 for conveying the produced sinter to a comminution device 10, a system 20 for monitoring the sintering process. For this purpose, the system 20 has a sensor device 22a, 22b, a first evaluation module 24, a second evaluation module 26, and an interface 28 202400276 10. In addition, the system 20 in this example also has an optional optical sensor 30 and a measuring device 32.

[0053] To produce the sinter, a mixture 12, also referred to as a material mix, is placed on one of the grate wagons 4, of which, for clarity, only one is designated with a reference numeral. The sinter conveyor 6 formed by the grate wagons 4 conveys the mixture 12 into a combustion zone 14a, where a fuel contained in the mixture 12, for example, coke dust, is ignited. A vacuum system 16 located below the grate wagons 4 loaded with the mixture 12 in the combustion zone 14a ensures that the mixture 12 burns completely from the top down until the respective grate wagon 4 reaches a discharge position 14b. At the discharge position 14b, the grate wagons 4 tilt from the horizontal to the vertical position, so that the sinter cake 18 formed during the combustion of the mixture 12 is discharged.The ejected sinter cake 18 impacts the sinter chute 8 and slides on to the crushing device 10, which in this example is designed as a spike crusher. The crushing device breaks the sinter cake 18 or, if the sinter cake 18 breaks into several pieces upon impact with the sinter chute 8, it reduces these fragments to the desired granules 2.

[0054] To determine the quality of the sinter produced in the sintering process, i.e., the quality of the sinter from the sinter cake 18, in a timely manner, the system 20 can execute a process 100 for monitoring the sintering process. Here, the sensor device 22a, 22b is expediently configured to detect the impact of the sinter cake 18 on an impact surface 8a, which in this example is formed or provided by the sinter chute 8, in a process step S1 (sensorially). In the example shown, the sensor device is designed as a vibration sensor 22a, for example as a structure-borne sound transducer, and is configured to detect vibrations, i.e., structure-borne sound waves, of the impact surface 8a or a component associated with it. Tests have shown that the vibrations caused by the impact of the typically multi-ton sinter cake 18 on the chute 8 are primarily in the range of 500 Hz to 10 kHz.Accordingly, the vibration sensor 22a is appropriately designed to detect vibrations in this frequency range.

[0055] Alternatively or additionally to detecting vibrations of the impact surface 8a or an associated component, sound waves generated by the impact of the sinter cake 18 on the impact surface 8a can also be detected in process step S1. This is indicated in Figure 1 by the sensor device 22b shown with dashed lines. The sensor device is advantageously designed as an acoustic sensor 22b, for example as a microphone. Tests have shown that the sound waves generated by the impact of the sinter cake 18 on the impact surface 8a are primarily in the range of 30 kHz to 50 kHz. Accordingly, the acoustic sensor 22b is advantageously configured to detect sound waves in this frequency range.

[0056] The first evaluation module 24 is expediently configured to evaluate the sensor data generated by the sensor devices 22a, 22b when the sinter cake 18 impacts the impact surface 8a. For example, in process step S2, the first evaluation module 24 can generate an impact characteristic based on the sensor data. This impact characteristic can be, for example, a measure of the impact force. Such a measure, which can also be referred to as impact intensity, is given, for example, by the maximum amplitude of the vibrations generated during the impact on the impact surface 8a (or the associated component) or by the sound waves generated during the impact. Accordingly, the first evaluation module 24 can be configured to filter out the maximum amplitude from the sensor data or a measurement signal corresponding to the sensor data.

[0057] If necessary, to facilitate the determination of the impact intensity and / or to take into account not only the maximum amplitude but also other signal components of the measurement signal, the measurement signal can be smoothed by applying a mean value filter. This is also expediently carried out by the first evaluation device 24, in particular within the framework of process step S2.

[0058] The second evaluation device 26 is expediently configured to determine the quality of the sinter from the sinter cake 18 in a further process step S3 based on the determined impact characteristics, i.e., to establish at least a trend in the sinter quality. The second evaluation module 26 can, for example, assign a sinter quality, i.e., a measure of the sinter quality, to the determined impact characteristics. For this purpose, the second evaluation module 26 can include or be configured as a machine-learned model 26a. This machine-learned model 26a is expediently trained to assign a sinter quality to a determined impact characteristic.

[0059] The determination of sinter quality based on the impact characteristics in process step S3 can be performed even more precisely, in particular the assignment of sinter quality to the determined impact characteristics can be facilitated by also taking into account an evaluation of optical sensor data generated when the sinter cake 18 is captured by the optical sensor 30. With the help of the optical sensor 30, it can be determined, for example, whether the sinter cake 18 still has so-called brown spots, i.e., areas that are not, or at least not completely, sintered. Alternatively or additionally, the particle size of the produced granules 2 can also be measured using the measuring device 32, and the resulting particle size distribution can be taken into account when determining the sinter quality. 202400276 12

[0060] The determined sinter quality is then output via interface 28 in a further process step S4, for example to the operating personnel of the sintering plant 1. Interface 28 can therefore be a hardware-based interface, such as a display device. Alternatively or additionally, interface 28 can also be software-based in order to make the determined sinter quality available to other technical equipment, for example, a control system of the sintering plant 1. After output via such a software-based interface 28, the determined sinter quality can, if necessary, be further processed, for example, converted into a format suitable for output to the operating personnel.

[0061] To calibrate the system 20, a corresponding procedure 200 can be applied. For this purpose, procedure steps S1 and S2 are performed several times, with the sensor device 22a, 22b being positioned at different positions P1, P2 in a procedure step S7. Simultaneously, independently of the determination of the impact characteristic in procedure step S2, the quality of the sinter from the respective sinter cake 18 is determined in a further procedure step S5. For this purpose, samples can be taken from the granules 2 and evaluated in a laboratory 34, for example. The sinter quality determined in this way (externally) is then assigned to the determined impact characteristic in a procedure step S6. In this way, a correlation between the determined impact characteristic and the absolute sinter quality, which is decisive for the impact characteristic, can be determined as a function of the sensor position P1, P2.The determined correlation with the different positions P1, P2 will vary depending on how "well" or unaffected the impact can be detected at the respective position P1, P2 by the sensor device 22a, 22b. A measure of the determined correlations can then be used as a basis for a final positioning of the sensor device 22a, 22b. The sensor device 22a, 22b can be positioned in particular where the measure of correlation is especially high, meaning the impact can be detected most reliably.

[0062] It should be noted that the sinter chute 8 in FIG. 1 is only an example of various possible conveying devices that can transport the discharged sinter cake 18 to the comminution device 10. It is equally conceivable to use a conveyor belt, conveyor carts, and / or the like instead of the sinter chute 8. The impact surface 8a can be provided accordingly by the respective conveying device. Alternatively, no conveying device may be provided at all. Instead, the sinter cake 18 can also be discharged directly onto the comminution device 10, e.g., directly onto a spike crusher or a ram arranged upstream of it for the coarse pre-division of the sinter cake 18. In this case, the vibrations generated in the comminution device 10 are expediently recorded. 202400276 13

[0063] FIG. 2 shows an example of sensor data D generated by a sensor device designed as a vibration sensor when detecting vibrations from the impact of a sinter cake on an impact surface. In this example, the sensor data D correspond to a measurement signal S, which indicates the amplitude A of the detected vibrations over time t, here over a period of slightly over 20 minutes. Approximately two impact events occurred per minute. The individual peaks, marked by a triangle in FIG. 2, each mark such an impact event. The height of these peaks, i.e., the maximum amplitude A of the detected vibrations, corresponds to an intensity I of the impact of the sinter cake on the impact surface, whereby, for the sake of clarity, only one of these impact intensities is labeled with a reference symbol.The impact intensity I in the present example lies - with one outlier - in a range between 0.15 and 0.3 m / s. 2 This corresponds to an expected fluctuation, for example, due to slight, uncontrollable changes in environmental conditions during the respective sintering process, the uncontrolled discharge of the sinter cake from the grate cart, and / or the like. It is therefore advisable to summarize the impact intensity I over the period of approximately 20 minutes shown, for example, by averaging. The averaged impact intensity represents an impact characteristic that is at least largely free of environmental influences. The quality of the sinter produced from the sinter cakes manufactured during the averaging period can be reliably assigned to this impact characteristic.

[0064] FIG. 3 shows an example of the correlation between an impact characteristic C and a characteristic quantity G for sinter quality. The impact characteristic C was determined from sensor data generated when a large number of sinter cakes impacted a surface. Specifically, the impact characteristic is the peak acceleration, averaged over a predetermined period (e.g., 20 minutes), of the vibrations generated by the impact of the sinter cakes on the surface within that period (see FIG. 2). In this example, the characteristic quantity G is the harmonic diameter of the sinter from the corresponding sinter cakes. Both the impact characteristic C and the characteristic quantity G are normalized in this example, making the correlation between them easier to see.

[0065] The data shown in FIG. 3 are derived from sinter production over a period of several days. As can be seen in FIG. 3, the characteristic value G, conventionally determined in the laboratory approximately 5–13 hours after the production of the respective sinter cake, correlates sufficiently well with the impact characteristic C determined for the corresponding sinter cakes immediately after the production process to allow conclusions to be drawn about an increase or decrease in sinter quality based solely on the impact characteristic C. Even though, in the present example, no quantitative statement about the sinter quality can yet be made based on the determined impact characteristic C, trends in quality can at least be identified. Consequently, sinter production can thus be monitored.In particular, an alarm can be triggered, for example, if the impact characteristic C falls below a predetermined threshold, which may indicate a significant decrease in sintering quality. Although the invention has been further illustrated and described by the preferred embodiments, the invention is not limited by the disclosed examples, and other variations can be derived by those skilled in the art without departing from the scope of protection of the invention.

[0066] 202400276 15

[0067] Reference symbol list

[0068] 1 sintering plant

[0069] 2 granules

[0070] 4 rust wagons

[0071] 6 Sintered belt

[0072] 8 sinter slide

[0073] 8a Impact area

[0074] 10. Shredding device

[0075] 12 Mixture

[0076] 14a Fire Zone

[0077] 14b Drop-off position

[0078] 16 Vacuum system

[0079] 18 sinter cakes

[0080] 20 System

[0081] 22a Vibration sensor

[0082] 22b acoustic sensor

[0083] 24 first evaluation module

[0084] 26 second evaluation module

[0085] 26a machine-learned model

[0086] 28 Interface

[0087] 30 optical sensors

[0088] 32 Surveying device

[0089] 34 Laboratory

[0090] 100, 200 procedures

[0091] S1 Impact detection

[0092] Determine S2 impact characteristics

[0093] Determine S3 sintering quality

[0094] S4 sintered quality output

[0095] Determine S5 sinter quality (external)

[0096] S6 assign impact characteristics to an (externally determined) sintered quality

[0097] Positioning the S7 sensor device

[0098] D Sensor data

[0099] S Measurement signal

[0100] C Impact characteristics

[0101] G characteristic size

[0102] A Amplitude 202400276 16

[0103] I Impact intensity t Time

[0104] P1, P2 Sensor position

Claims

202400276 17 Claims 1. Method (100) for monitoring an iron ore sintering process for the production of iron ore granules (2) for blast furnace operation, comprising the steps: - sensory detection (S1) of an impact of an iron ore sinter cake (18) on an impact surface (8a) and generation of corresponding sensor data (D), wherein vibrations and / or sound waves generated on the impact surface (8a) are detected by means of a sensor device (22a, 22b); - Determining (S2) an impact characteristic (C) based on the generated sensor data (D), wherein an impact intensity (I) is determined based on the sensor data (D); - Determining (S3) a quality of the sinter from the sinter cake (18) based on the determined impact characteristic (C), wherein a sinter strength is determined based on the determined impact intensity (I); and - Output (S4) of the determined sinter quality.

2. Method (100) according to claim 1, wherein the impact is detected by means of an acoustic sensor (22b) in a frequency range of 30 kHz to 50 kHz.

3. Method (100) according to claim 1 or 2, wherein vibrations of the impact surface (8a) and / or a component connected to the impact surface (8a) generated by the impact are detected by means of a vibration sensor (22a) in a frequency range of 500 Hz to 10 kHz.

4. Method (100) according to one of the preceding claims, wherein a quantity (G) characterizing the impact is determined for several sinter cakes (18) on the basis of the respective generated sensor data (D) and this quantity (G) is averaged over a predetermined period in which the several sinter cakes (18) impact the impact surface (8a) before the sinter quality is determined on the basis of the averaged quantity.

5. Method (100) according to any one of the preceding claims, wherein - a granulate (2) is produced from the sinter cake (18) by further breaking up the sinter cake (18) or its fragments after impact, - a distribution of particle sizes in the produced granules (2) is determined and - the sinter quality is additionally determined based on the calculated grain size distribution.

6. Method (100) according to one of the preceding claims, wherein the sintering quality is determined on the basis of the determined impact characteristic (C) by a machine-learned model (26a) which is trained on the basis of a training data set in which a plurality of impact characteristics (C) are each linked with at least one quantity (G) that characterizes the sintered material. 202400276 18 7. Method (100) according to one of the preceding claims, wherein a measurement signal (S) corresponding to the sensor data (D) generated during the detection of the impact is smoothed by applying a mean value filter and the impact characteristic (C) is determined on the basis of the smoothed measurement signal (S).

8. Method (100) according to any one of the preceding claims, comprising the steps: - Determining (S5) the quality of the sinter from the sinter cake (18), - Assigning (S6) the determined sinter quality to the determined impact characteristic (C).

9. Method (100) according to claim 8, wherein - the steps (S1 , S2, S5, S6) are performed several times and the sensor device (22a, 22b) is positioned at different positions (P1 , P2) relative to the impact surface (8a) (S7), - whereby for each execution of the steps (S1 , S2, S5, S6) a measure for the correlation of the determined impact characteristic (C) with the corresponding determined sintering quality is determined and - the sensor device (22a, 22b) is finally positioned relative to the impact surface (8a) based on the determined dimensions.

10. System (100) for monitoring an iron ore sintering process for the production of iron ore granules (2) for blast furnace operation, with - a sensor device (22a, 22b) which is arranged in the area of ​​an impact surface (8a) for iron ore sinter cakes (18) for detecting an impact of an iron ore sinter cake (18) on the impact surface (8a), so that vibrations and / or sound waves generated by the impact on the impact surface (8a) can be detected by means of the sensor device (22a, 22b), - a first evaluation module (24) which is set up to determine an impact characteristic (C) based on sensor data (D) generated by the sensor device (22a, 22b) when the impact is detected, wherein the first evaluation module (24) determines an impact intensity (I) based on the sensor data (D), - a second evaluation module (26) which is set up to determine a sinter quality based on the determined impact characteristic (C), wherein the second evaluation module (26) determines a sinter strength based on the determined impact intensity (I), and - an interface (28) for outputting the determined sinter quality.

Images