SYSTEM AND METHOD FOR MONITORING WEAR ON COMPONENTS OF A MILLING SYSTEM

Description

Technical Area

[0001] The invention relates to a system and a method for monitoring wear, in particular a system and a method for technical maintenance based on dynamic monitoring, of components of a grinding or rolling mill or other comminution device. In particular, the invention relates to a system and a method for monitoring surface wear on grinding rollers, in particular on roller profiles in pulverizing mills. Furthermore, the invention relates to a grinding plant with such a system.

[0002] Grinding systems, such as mills or grinders, often use cylindrical rollers or pairs of rollers for feeding and comminuting a material to be ground or comminuted. Due to the required comminution effect of grinding rollers, these are usually provided with a hard surface or are cast and surface-hardened to a depth of, for example, 10 mm. Beneath the hard surface, the material of the rollers is relatively softer and tends to wear out more quickly, which can lead to possible damage to the entire roller and thus to the failure of the grinding function. It is therefore advantageous to periodically or continuously monitor the wear on components of a grinding system, in particular the roller surfaces, as this enables timely maintenance work to be carried out and the expected remaining service life of the components to be predicted, thus ensuring reliable maintenance of the grinding systems. Background of the invention

[0003] Conventional milling systems use a roller mill as a comminution machine, in which the material to be ground is broken up and ground between rotating rollers. This type of milling system is the most common in industrial grain mills today. Milling in a modern mill comprises two processing steps that are typically repeated several times: comminution and sifting. The roller mill performs the milling. In this process, the flour kernel is separated from the hull particles in several steps. The resulting mixture is fed to a plansifter for separation. The coarse particles that have not yet been ground into flour are fed back to the roller mill by pneumatic or mechanical means. A pass through the roller mill and plansifter is called a passage. During milling, the grains are broken up by fluted rollers. Smooth rollers are used for dissolving, separating the flour and hull particles.During milling, the last flour particles are separated from the husk. The grain goes through eight to twelve passes before the flour is separated from the husk. The material to be ground, e.g. the grain, falls into a closed housing into the upper end of the roller mill. There, a feed roller feeds it to the grinding gap, ensuring a constant volume flow. This gap is formed by two rollers arranged horizontally or diagonally parallel, which rotate in opposite directions (like gears). Various typically adjustable parameters are crucial: (i) distance between the rollers (grinding gap), (ii) differential speed (lead for corrugated rollers 1:2.5 and for smooth rollers 1:1.25), (iii) number of corrugations, (iv) corrugation angle (twist), and (iv) position of the corrugations relative to one another (back:back, back:edge, edge:edge, edge:back).

[0004] In grinding plants, the feeding devices and the rollers should be arranged in such a way that the pressure on the working surfaces of the rollers is as uniform as possible. Figure 1 shows a roller mill, partly in elevation, partly in section. The material is fed to the comminution rollers a and b by means of feed rollers c and d. The latter runs at the same peripheral speed as comminution roller b, so that no accumulation of material can occur on the rollers and thus no increased load on the grinding surfaces. By means of an adjusting screw e, a slider f can be regulated so that feed roller c only feeds as much material to feed roller d as the latter can pass on to comminution rollers a and b. To prevent the material in the roller mill from heating up, the mills are now being equipped with an aspiration device. Self-pressing brushes, for example, can be used to clean the rollers.

[0005] Traditionally, the condition of the components of a grinding plant, particularly the grinding rollers, is monitored by a specialist inspecting the rollers and the ground material. However, this may require interrupting the grinding process, which leads to downtime and thus production losses. Furthermore, the detection of wear, damage, or inadequate product processing depends on the experience and attentiveness of the specialist. Results determined in this way are inaccurate and subjective. To objectively determine the wear of grinding rollers, it has been proposed to compare an actual condition with a target or initial condition. For this purpose, profile impressions of the roller surface are taken to evaluate corrugated rollers, for example, as described in DE 10063377 A1. Furthermore, analysis methods have been proposed that use light sources and photodetectors for optical detection of the roller surface, as explained in WO 2017 / 140928 A1.However, such methods for determining wear on grinding equipment are limited to analyzing the current condition of the surface, meaning they do not prevent damage or yield losses resulting from existing wear. Furthermore, such methods ignore other parameters of the grinding equipment and its surroundings that can contribute to wear and impair productivity. Furthermore, these methods are complex and time-consuming.

[0006] Electronic monitoring methods based on sensor-based data acquisition of measurement parameters have been developed to detect the condition or wear of the rollers of a grinding plant, as described, for example, in WO 2021 / 037525 A1. A sensor is arranged within one of the grinding rollers. The sensor is designed to detect measured values ​​that characterize the condition of the rollers. In particular, this can be the condition of a circumferential surface of the rollers. The condition can be, for example, a temperature, a pressure, a force (force component(s) in one or more directions), wear, vibration, deformation (extension and / or deflection path), rotational speed, rotational acceleration, ambient humidity, position, or orientation of the rollers, and can be detected using appropriately designed sensors.The condition of the grinding roller is continuously recorded while the grinding plant is in operation. A data transmitter is also provided for the contactless transmission of measured values ​​from the sensor to a data receiver and then to a control unit or a higher-level control system for evaluation, allowing the grinding plant or a part of it to be controlled and regulated. In particular, the control unit can issue a warning message if a predefined warning criterion is met. The warning criterion could, for example, be that the measured value exceeds a predefined limit. If the warning criterion is met, a warning signal can be issued or the grinding plant can be brought to a standstill, for example by the control unit. However, this monitoring method ignores influences outside the grinding roller.This can lead to unwanted downtime, loss of revenue and excessive maintenance costs.

[0007] Document EP 3500370 A1 discloses a grinding plant with grinding rollers. The plant has several sensors in the rollers for recording measured values ​​that describe the condition of the rollers. The sensors are designed to optimize the operation of the grinding plant using a control unit and to indicate the remaining service life or wear status of the rollers. The sensors are designed, for example, as temperature sensors for determining a temperature profile of the rollers, as vibration sensors for detecting roller vibration, and as accelerometers for monitoring the rotational speed, acceleration, or deceleration of the rollers, and transmit their measured values ​​to the control unit. The control unit includes a machine-learning unit with which the roller operation, such as the width of a gap between two rollers or the parallelism of the rollers, is automatically optimized based on the received measurement data.Furthermore, the machine-learning unit determines the remaining service life and wear status of the rollers. The machine-learning unit can comprise a monitoring unit for monitoring the received measured values ​​and a learning unit. The monitoring unit records, for example, the temporal development of temperature, gap width, and speed as a function of an actual and a target value. The learning unit carries out the learning process by linking the actual and target values. This allows the machine-learning unit to optimize the operation of the grinding plant with regard to the shape, twist, and surface condition of the rollers. While the remaining service life and wear status of the grinding plant components are taken into account, it does not provide any information regarding the optimization of maintenance work for the grinding plant.This can result in unnecessary costs and downtime and impair the efficiency of the grinding plant. Summary of the invention

[0008] It is an object of the present invention to provide a system and method for monitoring wear on components of a grinding or rolling plant, in particular for technical maintenance over the service life of such plants, which enable cost- and material-saving maintenance of the plant, an optimal time for replacing components of the grinding plant, and a cost- and material-saving integration of the wear behavior of the plant into its life cycle. In particular, such a system and method for grinding or rolling plants are to be provided that reduce downtime for maintenance work and simultaneously avoid excessive yield loss, quality loss, and material consumption over the life cycle of the plant.Furthermore, for such systems, a timely notification for the performance of maintenance work and a prediction of the expected remaining service life of grinding rollers in these systems are to be provided. In summary, the overarching technical problem underlying the invention is to provide a method and a system for a system for grinding and / or crushing, in particular grains, with which the grinding and / or crushing can be carried out in an optimized and automated manner and which increases the operational reliability and service life of such a grinding system.

[0009] According to the present invention, these objects are achieved by a method and a system for monitoring wear on components of a grinding plant, and by a grinding plant with such a system, having the features of the independent claims. Furthermore, further advantageous embodiments and variants can be derived from the dependent claims and the associated descriptions.

[0010] A method and a system for monitoring wear on components of a grinding plant, in particular for monitoring wear on grinding rollers or a pair of grinding rollers of the grinding plant, according to the invention determines an optimized future time for servicing the grinding plant, in particular servicing a grinding roller or a pair of grinding rollers. The method is used in particular for maintaining the grinding plant over the plant's life cycle. In the following, the method can also be referred to briefly as a monitoring method or as a maintenance method. The method is carried out by means of a system which has a control unit for controlling the grinding plant, in particular for controlling the grinding rollers orGrinding roller pairs, a sensor system for recording measurement data on technical parameters relating to a wear condition of a component, in particular a grinding roller or a pair of grinding rollers, and for recording measurement data on technical parameters relating to an operating condition of the grinding plant, an analysis unit for analyzing a wear behavior of the grinding plant and at least one data transmitter and at least one data receiver for exchanging data.

[0011] The sensor system records the measurement data on the wear status of at least one component, in particular a grinding roller or a pair of grinding rollers, and the measurement data on the operating status of the grinding system periodically or continuously over time and makes the measurement data available to the analysis unit, e.g., via a data transmitter. The sensor system can comprise several measuring sensors for this purpose.

[0012] The analysis unit assigns a degree of wear of at least one component to the measurement data for at least one of the parameters relating to the wear condition of at least one component. Furthermore, the analysis unit assigns a maintenance loss to the degree of wear, which corresponds to a loss of grinding yield due to technical maintenance of at least one component. The maintenance loss can be determined, for example, by a loss of grinding yield due to a downtime of the grinding plant and / or the costs of rectifying the wear. The maintenance loss can thus depend, for example, on the extent and duration of maintenance work and materials required to rectify wear on the component. From several pieces of measurement data relating to the wear condition recorded at different times, the analysis unit derives a future wear condition at a time T in the future.The analysis unit assigns a future degree of wear and a future maintenance loss to the future wear condition. For example, the future wear condition can be determined using standard extrapolation methods from periodically or dynamically collected measurement data. The future degree of wear and the future maintenance loss are correlated with the future wear condition and can thus be derived from it.

[0013] The analysis unit can be integrated into a digital cloud platform. This allows the analysis unit to be used from multiple geographically separated locations for different grinding plants. The analysis unit can therefore be easily updated and reused.

[0014] The analysis unit further assigns a grinding yield value to the measured data for at least one parameter relating to the operating status of the grinding plant. This value indicates the yield from the operation of the grinding plant. The analysis unit compares the grinding yield value with a specified target yield value for the grinding plant and determines a grinding yield loss relative to the target yield value. The target yield value can be specified, for example, by a technical plant specification from the plant manufacturer or plant operator. The target yield value can also be a theoretical ideal yield value for the grinding plant being monitored, calculated from the technical characteristics of this plant, or the target yield value can be determined through a practical test run of the plant, provided that no wear on plant components has yet occurred.The grinding yield value determined from the measured data on the operating state at a current point in time is usually below the target yield value specified for a grinding plant due to wear and tear on the grinding components, in particular the grinding rollers or pair of grinding rollers, which results in a loss of grinding yield. For example, the yield from the grinding operation of the grinding plant is reduced because wear and tear results in increased energy requirements for the plant or because additional passes are necessary to maintain a certain grinding quality. From several measured data on the operating state recorded at different times, the analysis unit derives a future operating state and, from this, a future loss of grinding yield at a time T operation in the future. The future operating state can be determined, for example, using extrapolation methods.The future mill yield loss correlates with the future and can therefore be determined from the measured data of the operating condition.

[0015] The analysis unit optimizes a future maintenance point in time by defining the point in time at which the future maintenance loss essentially corresponds to the future milling yield loss as the future maintenance point in time. Dynamic optimization is advantageously performed, whereby the analysis unit periodically or dynamically incorporates current measurement data regarding the wear status of a component and current measurement data regarding the operating status of the milling system into the wear behavior analysis, determining an updated degree of wear, an updated maintenance loss, and an updated milling yield loss, and from this, determines a currently optimized future maintenance point in time.

[0016] The monitoring method according to the present invention uses the technical parameters to determine the condition of the grinding plant and, based on these technical parameters, determines precautions for the technical maintenance of the grinding plant. The monitoring method can thus optimize the technical functioning of a grinding plant over its entire life cycle. In particular, undesirable or excessively long downtimes of the grinding plant can be shortened or avoided, a deterioration in the grinding yield can be minimized, and a reduction in grinding capacity over time can be prevented. This can extend or improve the service life of the grinding plant and its profitability. The monitoring method also supports the careful use of materials and other resources.

[0017] In an advantageous variant of the monitoring method according to the invention, the analysis unit determines at least one specific maintenance task required for a specific type of wear from the measured data on technical parameters relating to the wear condition of a component and / or from the measured data on the operating condition of the grinding plant. Advantageously, the analysis unit defines a future maintenance time for this required specific maintenance task as the time at which the specific future maintenance loss resulting from this specific wear corresponds to the specific future loss of grinding yield resulting from this specific wear. Examples of specific maintenance tasks include replacing a roll shaft, a roll stub, or a roll bearing, or the entire roll, balancing the rolls, and grinding and polishing smooth rolls.The monitoring procedure can determine a future maintenance time for each such specific wear.

[0018] In a further advantageous variant of the monitoring method according to the invention, the analysis unit aggregates measurement data of various parameters relating to the wear status of at least one component and / or various parameters relating to the operating status of the grinding system using an aggregation module. Based on the aggregated measurement data, the analysis unit can assign an aggregated degree of wear and a corresponding aggregated wear loss to the at least one component. Furthermore, the analysis unit can assign an aggregated loss of grinding yield value to the operating status of the grinding system. For example, measurement data relating to the temperature, vibration and rotational speed of a grinding roller, the gap spacing of a grinding roller pair, the ambient humidity, etc. can be incorporated into the aggregated degree of wear.Measurement data on the energy consumption of the grinding plant, the quantity and quality of the grinding yield, etc. can be included in the aggregated grinding yield loss. A multitude of measurement sensors in the sensor system can record measurement data on various parameters and transmit them to the analysis unit. The analysis unit aggregates the multitude of measurement parameters into an aggregated wear state or an aggregated operating state. This can also advantageously weight different parameters. For example, parameters that are more relevant for determining component wear can be given a higher weighting than less relevant parameters.

[0019] Advantageously, measurement data for different types of wear, such as the specific wear types mentioned above, and / or different components can also be taken into account and aggregated to determine a common wear level for the component or the entire grinding system. Different weightings can then be assigned to the different types of wear and components.

[0020] By using aggregated measurement results, it is possible to incorporate the influence of different types of wear and their impact on the operation of the grinding plant into the determination of maintenance loss and grinding yield loss. A holistic approach can be taken, taking into account the individual factors influencing the overall technical performance of the grinding plant.

[0021] For the monitoring method according to the invention, the sensor system of the monitoring system advantageously comprises several measuring sensors that are distributed in or near the grinding plant. Preferably, measuring sensors are used that record measurement data on a temperature of components, a roller profile and / or roughness of grinding rollers of the grinding plant, a rotational speed of the grinding rollers, a vibration of the grinding rollers, a distance between rollers of a grinding roller pair, an energy consumption of the grinding plant, a production quantity of the grinding plant, a degree of grinding of the ground material, a time since the last maintenance, and / or downtime of the grinding plant. The measuring sensors for determining the roller profile of corrugated rollers record, in particular, a contour of the corrugations, e.g., their angle, pitch, rounding, etc. For smooth rollers, the surface structure is recorded, in particular.The measuring sensors can also be designed as sensor units with multiple components. For example, an optical sensor unit can be provided, which includes, among other things, a light emitter and a light receiver, such as a camera.

[0022] In one embodiment of the monitoring system, for example, at least one temperature sensor and one vibration sensor are arranged on a grinding roller of the grinding plant. Furthermore, the grinding roller can advantageously comprise at least one data transmitter and a microprocessor to provide measurement data about the grinding roller to the analysis unit. This provides a type of roller wear measuring device on the grinding roller.

[0023] In a further advantageous variant of the monitoring method according to the invention, the analysis unit uses a regression module and / or an interpolation module to create a forecast of future maintenance loss and future milling yield loss. The regression module uses, for example, the temperature, surface condition, and vibration of a grinding roller as independent parameters to determine the degree of wear and predict it for the future. Likewise, energy consumption, the degree of grinding of the material being ground, and its quantity can be used to determine the milling yield loss and thereby determine a future value. By using a regression or interpolation module, proven statistical methods can be used to review the technical condition of the grinding system and determine wear behavior.

[0024] In yet another advantageous variant of the monitoring method according to the invention, the analysis unit comprises a machine learning module and can automatically determine the future maintenance loss and the future milling yield loss; in particular, these values ​​can be automatically optimized. The machine learning module, or machine learning module for short, makes it possible to self-adaptively optimize the future maintenance time. The machine learning module can be configured for various types of machine learning, for example, supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, transductive learning, multi-task learning, and the like. With the help of the machine learning module, the analysis unit can continuously adapt the analysis process, whereby the measurement data and target values ​​continuously recorded by the sensors can be used for the operation of the milling plant.

[0025] In an advantageous development of this variant of the monitoring method according to the invention, the machine learning module can send a control signal to the control unit for regulating the operation of the grinding plant, whereby the operation of the grinding plant is self-adaptively optimized according to a favorable or desired maintenance time such that the future maintenance time corresponds to the advantageous maintenance time. During the life cycle of the grinding plant, various times for maintenance may be appropriate in practical use. For example, maintenance of the plant can be advantageously scheduled if the grinding plant is being adapted to a changed material to be ground or if the plant is to be stopped for other reasons, e.g., non-technical reasons. Such an advantageous maintenance time can be before or after a future maintenance time determined by the monitoring system.Conversely, the machine learning module can determine optimized operation of the grinding plant.

[0026] Advantageously, the operation of the grinding plant can be optimized by optimizing the parallelism of grinding rollers in a roller pair, the spacing of the grinding rollers, and / or the rotational speed of the grinding rollers, based on the future maintenance loss and the future grinding yield loss determined by the machine learning module. For example, if a favorable maintenance time is only slightly later than a future maintenance time determined by the analysis unit, operation can be optimized by the control unit setting a somewhat more gentle or energy-saving grinding operation, or something similar. This can be the case, for example, in the event of unforeseeable events that make a standard maintenance procedure impossible.In contrast, if the advantageous maintenance time is earlier than a future maintenance time determined by the analysis unit, then operation can be intensified, which can lead to faster wear, which can, however, already be remedied at the earlier advantageous maintenance time.

[0027] In yet another advantageous variant of the monitoring method according to the invention, in addition to the measurement data from the sensor system, specification data of the grinding plant and / or a material to be ground are also processed from a data pool to optimize the future maintenance schedule, to determine future maintenance losses, and to determine future milling yield losses. The data pool can be made available, for example, via a cloud platform. The data pool advantageously provides data relating to a specific grinding plant, in particular data on material and labor costs for maintenance, energy costs for operating the grinding plant at the plant location, and / or information on the material to be ground, such as advantageous temperature ranges for grinding, the typical number of required passes, the purchase price of the material to be ground, or the selling price of the milled flour.Advantageously, these specification data can be continuously adjusted to current values, so that current specification data can also be taken into account for future maintenance. Advantageously, the data pool can provide measurement data on technical parameters regarding the wear status of a component and / or the operating status of other, comparable grinding systems in order to optimize the determination of future maintenance loss and future loss of grinding yield by comparing and / or incorporating this measurement data. The machine learning module can learn from these other, comparable grinding systems and self-adaptively further improve the optimal maintenance time.

[0028] The monitoring system according to the invention is advantageously constructed from units and / or modules that are at least partially provided in the cloud platform. Furthermore, the cloud platform can be used as a data storage for the measurement data. The cloud platform serves as a large central database and provides work units for executing the process, which can be used for different grinding plants at distributed locations. The design of the monitoring system at the individual grinding plants can therefore be simplified and made cost-effective.

[0029] In a further advantageous variant of the monitoring method according to the invention, an output unit can visually display a degree of wear of at least one component and / or a maintenance loss with regard to at least one component and / or a grinding yield loss of the grinding plant as a temporal progression. Future values ​​and the determined future maintenance time are advantageously also displayed. An operator of the grinding plant can easily read the development of the plant wear from the display. The individual degrees of wear and loss values ​​for various components, the projected values ​​and also the aggregated values ​​can be displayed. Advantageously, the output unit comprises a visualized, customized dashboard format which is designed to enable the automated display of wear reports, recommended future maintenance times with forecasts, orwith recommendations for roller replacement and information on grinding operations and their optimization. The operator can have the reports transferred to their personal company account, thereby improving process transparency. For example, a grinding plant manufacturer can provide service software in which individual operator data and data on their grinding plant can be managed. The operator uses this service software by setting up a company account on the software platform to make the grinding plant data and information easily and transparently accessible to the operator.

[0030] In yet another advantageous variant of the monitoring method according to the invention, the measurement data on technical parameters relating to the wear status of a component and the operating status of the grinding system can be uploaded to a cloud platform using the data transmitter, and / or the data pool containing data relating to a specific grinding system and / or data relating to other, comparable grinding systems can be made available on a cloud platform. The decentralized storage, storage, and processing enhances process reliability and expands the processing volume. This allows the monitoring method to create a reliable maintenance plan and continuously update it without delays, which is only partially possible with local storage capacities.

[0031] In a further advantageous variant of the monitoring method according to the invention, a service module of the monitoring system automatically recommends a schedule for maintenance of the grinding plant based on the determined future maintenance loss and the future loss of grinding yield. Additionally or alternatively, the service module can automatically create a quote for maintenance of the components and / or an associated maintenance order. For example, a service module derives a required maintenance procedure from the determined wear level. The procedure can include a schedule of individual maintenance activities and a list of materials to be replaced. From this, a quote with a schedule and cost breakdown can be automatically created. An operator of the grinding plant can accept this quote, whereby the maintenance procedure can be automatically activated.For example, materials can be ordered and maintenance specialists can be booked.

[0032] In a next advantageous variant of the monitoring method and monitoring system according to the invention, an identification means is provided on at least one of the monitored components, in particular on a grinding roller. The identification means comprises electronically stored data relating to the component. The component of the grinding system can be uniquely identified by means of the identification means. For example, the identification means can provide individual data for the component and for at least one property of the roller, such as at least one of its dimensions, its crown, and / or a product number. For this purpose, the component can, for example, comprise at least one data storage device, which is designed, for example, as an RFID chip. The data stored in the data storage device can advantageously be transmitted contactlessly to the analysis unit using a data transmitter.It is possible that the data of the identification means are transmitted using the same data transmitter by means of which the measurement data regarding the wear condition and the operating condition are transmitted.

[0033] The object of the present invention is further achieved by a grinding system with grinding rollers, in particular with fluted or smooth rollers, comprising a system as described above. The grinding system according to the invention is preferably designed for grinding wheat and / or rye and / or durum and / or oats and / or barley and / or peas and / or chickpeas and / or pulses and / or rapeseed and / or soy and / or cocoa and / or coffee as the ground material.

[0034] In a simple example of the application of the monitoring system and method according to the invention, measurement parameters of the roll wear measuring device are uploaded to a digital cloud platform, and the measured measurement parameters of the roll wear measuring device are automatically analyzed by the analysis unit of the digital platform. An operator can receive an automated display of the wear results as reports with predictions and recommendations for roll replacement and information on process transparency and optimization, and in a visualized, customized dashboard format transferred to their personal company account. The analysis unit links the measurement results to the service module for creating a corrugating and smooth roll refurbishment and replacement service and for automatically creating a quote and corresponding order for the corrugating and smooth roll refurbishment / replacement services.A similar procedure can be applied to other components, such as feed rollers, adjusting elements, feeding devices and the like.

[0035] The system and method for monitoring wear on components of a dedicated grinding plant improves and optimizes plant maintenance and the production of ground material throughout its entire life cycle. The quality and quantity of the grinding result are sustainably improved, preventing declining yield from the grinding plant. Short description of the drawings

[0036] Advantageous embodiments of the invention are presented below with reference to the figures, which are for illustrative purposes only and are not to be interpreted in a restrictive manner. Features of the invention revealed by the figures are to be considered, individually and in any combination, as part of the disclosure of the invention. The figures show: Fig. 1: a roller mill of a grinding plant according to the prior art, partly in view and partly in section, Fig. 2: schematic representation of a system for monitoring wear on components of a grinding plant, in particular on grinding rollers of the grinding plant, according to the present invention, Fig. 3: schematic representation of a sequence of the method for monitoring wear on components of a grinding plant according to the invention, Fig. 4a & 4b: graphic representation of an analysis by an analysis unit of a monitoring system according to the present invention, Fig. 5a - 5d: examples of wear on corrugated rollers, Fig. 6a & 6b: wear analysis on surfaces of smooth rollers, Fig. 7: example measurement report for wear analysis from Fig. 6a & 6b , Fig. 8a & 8b:Wear analysis of corrugating roller surfaces, Fig. 9a:Example measurement report for wear analysis from Fig. 8a & 8b , Fig. 9b:Example analysis report for wear analysis from Fig. 8a & 8b , Fig. 9c:Example measurement report for wear analysis from Fig. 8a & 8b , Fig. 10a & 10b: schematic representation of a groove technology for corrugating rolls, Fig. 11: tabular overview of corrugation types for corrugating rolls Fig. 12: schematic representation of parameter values ​​for corrugating rolls with wear, Fig. 13: interaction field for entering specification data, and Fig. 14: graphical example of an analysis model for determining a future maintenance time. Detailed description of preferred design variants

[0037] To introduce the field of the invention, Figure 1Components of a roller mill of a grinding plant, partly in elevation, partly in section, as known from the prior art. The material to be ground is fed to the comminution rollers a and b by means of feed rollers c and d. The latter runs at the same peripheral speed as comminution roller b, so that the material never accumulates on the rollers and thus places greater strain on the grinding surfaces. By means of an adjusting screw e, a slider f can be regulated so that feed roller c only feeds as much material to feed roller d as the latter can pass on to comminution rollers a and b. To prevent the material in the roller mill from heating up, the mills are now being equipped with an aspiration device. Self-pressing brushes, for example, can be used to clean the rollers.

[0038] A method and a system for monitoring wear on components of a grinding plant and for maintaining the grinding plant according to the invention and variants thereof are explained in the figures by way of example and in extracts. As an overview, Figure 2 Such a monitoring system is shown schematically. A grinding plant 1 for processing or grinding a grinding material 100 into a ground grinding product 110 comprises at least one pair of grinding rollers 2, which correspond to the comminution rollers a and b of Figure 1 In the following, the grinding rollers 2 represent various components of the grinding plant, the wear of which is monitored by the system according to the invention in order to simplify the explanation of the invention.

[0039] The monitoring system further comprises a control unit 3 for controlling the grinding system, in particular the roller pair 2, which is advantageously arranged in the grinding system. Furthermore, a sensor system 4 is provided for acquiring measurement data on technical parameters relating to the wear status of a component, in particular the grinding rollers, and the operating status of the grinding system. The sensor system 4 comprises a plurality of measuring sensors for acquiring measurement data on technical parameters relating to the wear status of a component and / or the operating status of the grinding system. For example, the sensor system comprises one or more temperature sensors 4.1 for measuring the temperature of the grinding rollers 2, their surroundings, and / or the ground product 110, one or more vibration sensors 4.2 for measuring the vibration of the grinding rollers 2, one or more accelerometers 4.3, preferably at least one for each grinding roller 2, and one or more moisture sensors 4.4 for detecting the humidity of the environment of the grinding rollers 2 and / or the ground product 110, an energy consumption sensor 4.5 for detecting energy consumption of the grinding system 1, and sensors 4.6 for detecting measurement data on the quality and quantity of the ground product 110. Furthermore, measuring sensors 4.7 for determining a surface quality or condition, a distance between the grinding rollers 2, and the parallelism of the grinding rollers 2 to one another, and measuring sensors 4.8 for determining an identification means 5 of the grinding rollers 2, may be present. The measuring sensors 4.1 ... 4.8 of the sensor system 4 detect wear measurement data 40 for technical parameters relating to a wear condition of a component, in particular the grinding rollers 2 of the grinding system 1, such as the condition of the roller surface, the roller spacing, the roller parallelism, and the roller temperature. The measuring sensors 4.1 ... 4 also detect.8 of the sensor system 4, operating measurement data 50 relating to technical parameters of an operating state of the grinding plant 1, such as energy consumption, ambient temperature and humidity, the rotation or change in rotation of the grinding rollers 2, and measurement data relating to quality (e.g., grain size of the ground product) and quantity (such as a weight or a change in the weight of the ground product). The sensor system records the measurement data 40 relating to the wear state of at least one component and the measurement data 50 relating to the operating state of the grinding plant periodically or continuously over time, so that a series of measurements extending over a defined period of time is available. The defined period of time can extend, for example, from the first commissioning of the grinding plant or from the last maintenance or from other predetermined events of the grinding plant.

[0040] An analysis unit 6 of the monitoring system analyzes the behavior of the grinding plant 1 with regard to wear of the components, or a wear-related change in the operation of the grinding plant over time. The analysis unit 6 can be provided in a computing device, such as a computer, of the grinding plant 1. However, according to the invention, the analysis unit 6 is advantageously provided in a cloud platform 7. Thus, the analysis unit 6 can be part of a monitoring system for different, independent grinding plants. By using a cloud platform as the central location for the analysis unit 6, a high working capacity can be provided for the system, and the computing device of the grinding plant can be designed with a lower working power.The measurement data 40 and 50 recorded by the sensor system 4 relating to technical parameters of the grinding plant 1 are provided, preferably in a contactless manner, to the analysis unit 6 by one or more data transmitters 8 provided in the grinding plant 1. The data transmitters 8 can advantageously be provided on the components to be monitored or in their immediate vicinity. For example, a data transmitter can be provided on one grinding roller, preferably on both grinding rollers 2. Additional data transmitters can be provided on a power supply of the grinding plant, in a storage room for the grinding material 100 and / or a storage room for the ground product 110. The measuring sensors can advantageously be equipped with the data transmitters. The data can be transmitted to the analysis unit 6 or to the cloud platform 7 via a network connection. The measurement data is advantageously stored in a storage space in the cloud platform 7 and made available to the analysis unit 6.To receive the measurement data, the analysis unit 6 or the cloud platform 7 comprises at least one data receiver 9.

[0041] Furthermore, the system for monitoring wear on components of a grinding plant according to the invention can comprise a data pool 10. The data pool 10 is advantageously also provided in the cloud platform 7, but can also be accommodated, for example, in the computer unit of the grinding plant 1. The data pool can contain additional data that is helpful for analyzing wear and future maintenance times. For example, specification data 60 of the grinding plant 1 can be stored in the data pool. The specification data 60 can comprise plant-specific data and threshold values ​​for grinding production, such as a yield target value, a temperature target value for the grinding rollers 2 and / or the grinding material 100, or the grinding product 110, a moisture target value for the grinding product, target values ​​for the surface quality, parallelism and the spacing of the grinding rollers 2, dimensions of the grinding rollers 2, information on the grinding material 100, information regardingCosts, availability, and expected energy consumption of the components; information about the milled product 110, such as target temperature, target humidity, and sales value. The specification data 60 can be continuously updated in the data pool 10, so that the analysis unit 6 can be supplied with current specification data 60 at any time.

[0042] The totality of the data recorded over time, consisting of wear measurement data 40, operational measurement data 50, and, if applicable, specification data 60, forms a multidimensional data space that forms the basis for the analysis unit 6 to determine an optimal maintenance time in the future for servicing wear and for maintaining the grinding system. For this purpose, the analysis unit is configured with an analysis algorithm that determines a future wear pattern, a future maintenance loss, and a future loss of grinding yield from the data available in the data space. From these projections, the analysis unit then determines an optimal future maintenance time. Essentially, statistical methods and / or experimental modeling methods can be used to analyze the data and for a trend forecast, whereby confidence levels and probabilities associated with the parameters can be predefined.The analysis unit 6 can also include a machine learning module 11 that automatically optimizes the determination of the future maintenance time. For this purpose, the machine learning module 11 accesses the data in the data space and applies machine learning methods, in particular methods based on pattern recognition, to continuously improve the analysis of a wear history. The machine learning module 11 identifies correlations in patterns in an analyzed data set from the data space and applies these correlations to new data sets, such as data sets from other grinding plants or from other data acquisition time windows. An algorithm based on supervised learning is advantageously used for this purpose. The algorithm is trained using existing measurement data series and associated known wear characteristics and associated yield losses.The target variable is determined as the maintenance time at which total milling yield loss is optimized. Alternatively, an algorithm based on unsupervised learning can be used, especially when only a limited number of comparative measurement data are available to determine a relationship between wear and milling yield loss.

[0043] The system for monitoring wear on components of a grinding plant according to the invention can further comprise an output unit 12, which visually displays the results of the data analysis, in particular the wear trend and a determined future maintenance time, as a time course. Advantageously, a course beyond the maintenance time is also displayed in order to illustrate an expected wear trend and loss of grinding yield if wear is not remedied, i.e., no maintenance is performed. The output unit 12 can be provided on the grinding plant 1, such as on the computer unit. Furthermore, the output unit 12 can be embodied on a mobile device, such as a smartphone or a tablet, which communicates with the analysis unit via a network.In addition to the visual representation of the technical parameters and analysis results, quantitative information regarding operating conditions and wear patterns can be displayed. Examples of graphical representations of the results from the monitoring process are provided in conjunction with the . Figures 7 and 9a to 9c explained.

[0044] Furthermore, the monitoring system according to the invention can comprise a service module 13, which, preferably based on the analysis results, recommends a schedule for maintenance of the grinding plant and / or automatically creates a quotation for maintenance of the components and / or a maintenance order. For this purpose, the service module 13 can, for example, access specification data 60 from the data pool 10, such as the availability of replacement components, costs for the components, labor costs, etc. The quotation can advantageously be displayed on the output unit 12. The service module 13 is advantageously provided in the cloud platform 10 and can be a component of several monitoring systems provided for different grinding plants.

[0045] In Figure 3By way of example, a sequence of the method for monitoring wear on components of a grinding plant, in particular on grinding rollers of the grinding plant, according to the invention is set out. The individual steps of the method sequence can essentially run in real time and continuously, or the method can be carried out at predetermined time intervals, whereby the analysis results can be updated periodically. Alternatively, the method can also be triggered by manual activation, for example if an update is desired. First, in the monitoring method according to the present invention, in a data acquisition step 200, the sensor system 4 records the measurement data 40 relating to the wear state of at least one component and the measurement data 50 relating to the operating state of the grinding plant periodically or continuously over time. The measurement data 40 and 50 are provided to the analysis unit 6 as described above.In a further data collection step 210, the specification data 60 can be provided to the analysis unit 5. As shown in this process example, the specification data 60 can contain information on production capacity, production quantity, energy costs, purchase and sale prices for the ground material and the ground product, and a target yield. In particular, the specification data can include information regarding specific maintenance work, such as costs for refurbishing and replacing a grinding roller, transport costs, costs due to production downtime, and labor costs. The specification data 60 can, for example, be retrieved from the data pool 10, stored in the computer unit of the grinding plant, or made available in another way.

[0046] In an analysis step 220, the analysis unit 6 assigns a degree of wear to the measurement data 40 for at least one parameter relating to the state of wear of at least one component and a value for a maintenance loss 300 to the degree of wear (see Figure 4), which corresponds to a loss of yield due to technical maintenance of at least one component. Furthermore, the analysis unit derives a future wear state at a point in time in the future from a plurality of measurement data 40 relating to the state of wear, recorded at different times, and assigns a future degree of wear and a future maintenance loss 310 to the future wear state. At the same time, the analysis unit assigns a grinding yield value, which indicates the yield from the operation of the grinding plant, to the measurement data 50 relating to at least one parameter relating to the operating state of the grinding plant, and determines a grinding yield loss 320 relative to a target yield value of the grinding plant 1. Furthermore, the analysis unit 6 derives a future operating state from a plurality of measurement data 50 relating to the operating state, recorded at different times, and from this, a future grinding yield loss at a point in time in the future.From these analysis steps, the analysis unit 6 determines an optimized maintenance time TW in the future by defining the time at which the future maintenance loss 310 corresponds to the future milling yield loss as the future maintenance time TW. The individual analysis values ​​300, 310, 320 as well as the future maintenance time TW can be determined by the analysis unit 6 using statistical methods and / or empirical modeling methods, as explained above. For example, the analysis unit 6 can create a forecast of the future maintenance loss and the future milling yield loss using a statistics module 14 of the monitoring system, such as a regression and / or interpolation module. The statistics module 14 can be provided in the analysis unit 6 and / or the cloud platform 7. In particular, the statistics module 14 can be a component of the machine learning module 11.The machine learning module 11 can further improve the analysis of the data and the determination of the future maintenance time TW based on machine learning methods, as previously mentioned. After determining the analysis results, they are made available to the operator of the grinding plant 1 in an analysis report via the output unit 12, as shown, for example, in the following figures. Figures 7 and 9a to 9c will be explained in more detail. Furthermore, a maintenance plan for servicing the grinding plant 1 can be suggested to the operator using the service module 13 and the output unit 12. Finally, the analysis unit can advantageously transmit at least the current degree of wear and the current grinding yield value to the control unit for automated control of the grinding plant. Preferably, the predicted values ​​are also sent to the control unit to optimize the control of the grinding plant.

[0047] In addition, Figure 3The further procedure for the maintenance and servicing of grinding plant 1 is described. In a maintenance step 230, the maintenance plan is implemented according to the offer of service module 13. In an update step 240, the operating data and specification data are updated, if necessary. For example, the time of the last maintenance of the affected components and their identification feature are stored, and an updated yield target value can be recorded. Optionally, in a final step 250, the maintenance work can be invoiced, for example using analysis unit 6. In this case, for example, a comparison value can also be determined and displayed, indicating an amount of loss that was avoided by servicing the grinding plant at the determined optimized maintenance time instead of at an earlier or later time.The loss amount therefore corresponds to the amount saved by carrying out maintenance at the optimized maintenance time.

[0048] In Figure 4aA graphical representation of the results from an analysis of the analysis unit 6 according to the monitoring method of the present invention is shown. The graph indicates units for a time course on the x-axis, or a unit of measurement for a lost grinding yield and a loss due to increased energy consumption caused by wear of the grinding components. On the y-axis, the graph indicates cost units for these units of measurement, or costs over time. The graph is normalized such that no costs, no lost grinding yield, and no loss due to increased energy consumption occur as long as the grinding plant produces a yield that corresponds to the yield target value for the plant. An energy loss curve 300 and a yield loss curve 310 therefore begin at the zero point of the graph. Over time, the degree of wear on the components of the grinding plant increases, and the grinding yield decreases during operation of the grinding plant.This means that the value of the lost grinding yield and the loss due to increased energy consumption increase. In the example shown, the analysis unit determined a linear increase for the lost grinding yield and the loss due to increased energy consumption over time. In addition, the graph shows the course of yield loss due to maintenance of the grinding plant over time, or rather the age of the grinding plant, as a maintenance loss curve 320. At the beginning of the operation of a grinding plant, in particular when a grinding plant is first put into operation, at time zero, there is no or only negligible wear on the components of the grinding plant. If maintenance is nevertheless carried out, for example merely to inspect the components, the grinding plant must be stopped, which results in a maintenance loss caused by the maintenance. The maintenance loss consists, for example, of:from the lost revenue when grinding operations are shut down and the effort required to carry out maintenance or inspection. When the grinding plant is first put into operation, the maintenance loss is therefore very high. With increasing age and thus increasing wear of the components, the maintenance loss decreases. In the example shown, the analysis unit has determined an exponential decline in curve 320. It should be emphasized that curves 300, 310 and 300 for the loss values ​​of the grinding plant are only determined for a period in the past using measured data. For a period in the future, the curves are determined by the analysis unit on the basis of measured data, if applicable specification data and, if applicable, additional available information.For a current time TA, a maintenance time TW in the future can thus be optimally determined by determining the time at which the future maintenance loss, represented by curve 320, corresponds to the future milling yield loss, represented by curve 310. It should be noted that instead of separate curves for the energy loss curve 300 and the yield loss curve 310, a milling yield loss curve can be determined based on the aggregated data for these curves. The aggregated milling yield loss curve thus reflects the combined value of all factors that reduce milling yield due to wear.

[0049] In Figure 4b A graphical representation of the results from an analysis of analysis unit 6 to another grinding plant is shown. The x-axis and y-axis correspond to the axes of Figure 4aHere, an aggregated mill yield loss curve 410 was determined, which indicates the loss in mill yield caused by lost profit and increased energy costs due to aging and wear of the plant components. The curve does not begin at the origin of the graph, since a mill yield loss already exists at this point. A maintenance loss is indicated by the maintenance loss curve 420. As in the previous example, the maintenance loss decreases exponentially over time. An optimized maintenance point in the future is determined by the intersection point of graphs 410 and 420, at which the future maintenance loss corresponds to the future mill yield loss. The course of a total loss is indicated by graph 460, for which the mill yield loss and the maintenance loss are added together. Figure 4bIn addition, time ranges are defined which correspond to a period for preventive maintenance 430, a period for predicted optimized maintenance 440 and a period for maintenance in the event of component failure 450. A current time TA is generally in the range 430 of preventive maintenance, where, however, there is a higher yield loss when the grinding plant is shut down than can be compensated for by the maintenance. A future optimal maintenance time TW is in the period for predicted optimized maintenance 440, in which the curves 410 and 420 approach each other and in which the peak of the total loss curve 460 lies. Around this peak, the total loss curve 460 is at least approximately horizontal, so that the total loss costs at different times in this range only differ slightly. The period for predicted optimized maintenance 440 could, for example,by a region of the total loss curve 460 in which the curve has a gradient of less than 15 degrees, preferably less than 10 degrees, and particularly preferably less than 5 degrees. This results in an optimized maintenance period within the meaning of the present invention. Later maintenance carries the risk of failure of the grinding system due to a failure of grinding components, which leads to a greatly increased loss of grinding yield.

[0050] In summary, the present invention improves the operation of a milling plant by optimizing its maintenance and repair. This is achieved through technical monitoring of the plant components and the products processed by the plant. Operations are further improved by digitizing the monitoring and displaying monitoring results in a clear and understandable manner. A method and system according to the present invention are suitable for measuring the wear condition of grain processing rollers, including corrugated and smooth rollers for corresponding milled material, using a roller wear measuring device. Measurement parameters of the roller wear measuring device are advantageously uploaded to a digital cloud platform, and the measured parameters of the roller wear measuring device are automatically analyzed by the analysis unit of the digital platform.A user receives an automated display of wear results and reports with predictions and recommendations for roll replacement and additional process transparency / optimization in a visualized, customized dashboard format transferred to their personal company account, which can be provided by a plant manufacturer as a service. For a grinding plant operator, it is of utmost importance to avoid yield losses and downtime. From a technical perspective, yield losses arise primarily from the wear of the grinding plant components, which leads to a decrease in throughput. When grinding rolls wear out, a higher contact pressure of the rollers in a grinding roll pair is required to grind a material, which increases the energy consumption of the grinding plant.

[0051] In the Figures 5a to 5dVarious examples of surface finishes on corrugated rollers exhibiting varying degrees of wear are shown. Graph 900 shows an actual corrugation contour, while graph 910 shows an ideal line for this corrugated roller for comparison. Such differences in wear are not detectable with a purely visual inspection. The increased energy consumption leads to heating of the grinding components and their surroundings, which reduces the moisture content in the surroundings and in the ground product. Wear thus leads to reduced product quality and quantity. It is therefore crucial to find the right time to maintain or replace components, especially the grinding rollers, in order to maintain high product quality and optimize operating costs.The method and system for monitoring wear on components of a grinding plant, as well as the grinding plant equipped with such a system according to the present invention, help prevent the negative effects of wear. The method is explained below using the example of wear on smooth and corrugated rollers of grinding plants. However, the method can also be applied to other components of the grinding plant.

[0052] In simplified terms, a surface temperature of a circumferential surface of at least one roller of the grinding plant and / or the temperature of a ground product can be measured using at least two temperature sensors that measure the temperature at different points on the at least one roller or a product curtain. With a monitoring method according to the invention, it is possible to detect and monitor the temperature where it is generated, namely on the surface of the rollers. Furthermore, to monitor the roller temperature, the temperature of a ground product is monitored directly, since the roller transfers heat to the ground product, and by measuring the temperature of the ground product, a conclusion can be drawn about the temperature of the grinding rollers.

[0053] In one embodiment of the method and system for monitoring wear on components of a grinding plant, a monitoring device for automated optimization of a grinding plant control system is used, as described in the applicant's patent specification EP 3500370 B1. In this monitoring device, the control of the grinding plant, in particular its grinding rollers, is automatically optimized by evaluating measurement data on components of the grinding plant over their service life using a machine-learning unit. However, the monitoring device does not determine an optimized maintenance time based on wear measurement data, operational measurement data, specification data, and possibly additional information, as proposed by the present invention.At least for the design of the grinding plant, the arrangement of measuring sensors on grinding rollers, the description of wear on smooth rollers and corrugated rollers, and other features related to the operation of grinding plants, the present invention is based entirely on the description in patent specification EP 3500370 B1. This description is hereby incorporated into the description of the present invention. This applies in particular to the description of the following figures from EP 3500370 B1. In . Figure 2 The detection of a corrugation type of a corrugating roll is described by means of measured values ​​on the roll, whereby the surface structure of the roll is indicated both in the x-axis and in the y-axis. Figures 3 and 4 A grinding system is described that has a grinding roller with several temperature sensors and data transmitters. Figures 5a and 5bshow two possible designs for integrating sensors into one or both rollers of the grinding plant's roller pair. Figure 5a As in 5b, the acceleration measurement is carried out using an accelerometer. In Figure 6 An exemplary embodiment of the arrangement of measuring sensors on a grinding plant is shown. The measuring sensors used may include, for example, a wear sensor, a pressure sensor, a temperature sensor, a vibration sensor, an acceleration sensor / accelerometer, a force sensor, or a deformation sensor, etc. Figure 7 describes the arrangement of measuring sensors. Figures 8 and 9 explain the integration of measuring sensors.

[0054] In a variant of the monitoring method according to the present invention, the surface quality of smooth rollers 2.1, as shown by way of example in Figure 6ashown, continuously or periodically. The measurement data from the measuring sensors can be used to monitor the temporal progression of the surface roughness. Figure 6bThe surface roughness is shown along the length of the roller. The condition of the surface can be recorded using optical measuring sensors, for example. Preferably, however, the roughness of the surface is determined using a scanning measuring sensor comprising a sensitive diamond tip that scans the surface. From the scanning measurement data, the analysis unit determines the surface profile and uses this to determine the wear status or degree of wear of the smooth roller. This is done, for example, by comparing it with a theoretical or ideal reference surface profile, or by comparing different measurements over time. The analysis unit also assigns a maintenance loss to the degree of wear, assuming that maintenance would be carried out at this degree of wear. The maintenance loss can, for example, be determined from the current measurement data on the quality and quantity of the ground product at the time of the observed degree of wear.The maintenance loss can also be determined based on a reference value for the grinding yield. Based on the wear rate or maintenance loss trend, the analysis unit determines a future wear rate and assigns a future maintenance loss. Furthermore, the analysis unit determines the actual grinding yield value from the grinding plant's operating data during the time the surface roughness was measured. From this, an actual grinding yield loss is determined relative to a target yield value for the grinding plant, from which a future grinding yield loss is derived.Using these determined forecasts for the maintenance loss and the grinding yield loss, the analysis unit determines an optimized maintenance time TW in the future by setting the future maintenance time TW as the time at which the future maintenance loss due to the wear of the surface of the smooth roll corresponds to the future grinding yield loss, as defined in the . Figures 4a and 4b is shown.

[0055] In Figure 7 An example of a measurement report is shown as it is generated by the measurement data output unit Figure 6b can be displayed. In addition to information on the measurement method, such as the measurement location, measurement type, and roll identification, the report also provides information on an average roughness value according to DIN EN ISO 4287 and a roughness classification, such as "OK," "monitor," or "critical."

[0056] In a further variant of the monitoring method according to the present invention, the surface quality of corrugating rollers 2.2, as shown for example in Figure 8a are shown, continuously or periodically. The measurement data from the measuring sensors can be used to monitor the wear of the flutes over time. Figure 8bA surface contour of the corrugation determined from the measured data is shown. In the monitoring process, the wear of this contour is monitored over time, and a wear profile for the future is generated from this. As with the previous variant of the process for smooth rolls, the analysis unit determines a wear status of the corrugating roll from the measured data on the surface contour and assigns a degree of wear and a maintenance loss to this. Furthermore, a future wear status is derived from several measured data recorded at different times, and an associated future degree of wear and future maintenance loss is assigned. At the same time, the analysis unit determines a grinding yield value from measured data on the operating status of the grinding plant. This value indicates the yield from the operation of the grinding plant and uses this to determine a grinding yield loss relative to a target yield value for this grinding plant.From several measurement data on the operating status recorded at different times, a future operating status and, from this, a future loss of milling yield are derived. According to the invention, the analysis unit optimally determines a future maintenance time by determining the time at which the future maintenance loss due to wear on the corrugating roller corresponds to the future loss of milling yield as the future maintenance time.

[0057] In Figure 9a An example of a measurement report is shown as it is sent from the output unit for the corrugating roller Figure 8bcan be displayed. The report shows the Graph 900 of the measured corrugation contour within a defined measuring range. Furthermore, the operator receives information on current corrugation contour values, target values, and typical tolerance values. Furthermore, a categorization of the wear condition of corrugation edges, the entire roll, and the roll butt can be specified, such as "sharp," "blunt," and "critical." Furthermore, angles characterizing the corrugations can be specified, as described below. Figure 9b An analysis report is shown as it is for the corrugating roller from Figure 8b In this, the graph of the corrugation contour 900 is shown according to the measurement data, as shown in Figure 9a, shown. Furthermore, a theoretical ideal line 910 for the corrugation contour is shown for comparison. Furthermore, additional characteristics of the current corrugation contour are given, such as a wear ratio, a wear height, a reduction in height, a gap width to a parallel corrugating roll, a value for the parallelism of the rolls, etc. Finally, a classified recommendation is given, such as «corrugation still intact», «corrugation at the end of its service life» or «corrugation should be replaced». Further analysis results are presented in connection with Figure 13 explained. In Figure 9c A forecast report for this corrugating roller is shown as an example. The forecast report shows the analysis graph as shown in Figure 4aThe forecast report also provides values ​​for yield loss, loss due to increased energy consumption, current maintenance costs, a wear rate, an expected service life, and the time until the next recommended service for the corrugating roller. In this example, the remaining service life is 15.25 months. Service of the roller is recommended in 11 months.

[0058] In the Figures 10a and 10b Examples of corrugation contours are shown to explain the characteristics of corrugating rollers. The corrugations of a corrugating roller are described using Figure 10adescribed. Corrugations of corrugating rollers have the following surfaces: a squeezing surface 500, which is also described as a land, a free surface or rear surface 510, which leads behind a corrugation to a base surface 520, and a cutting surface 530, which forms a cutting edge 540 with the squeezing surface 500. The preferred direction of rotation when using the corrugating roller is indicated by arrow 550. Flat corrugation contours (left in Figure 10b ) are used for small grinding material and deep riffle contours (right in Figure 10b ) is used for coarser ground material. The height or depth of the ridges extends from the base area 520 to the crushing area 500. In Figure 11The table shows various designs of corrugation contours used today in corrugating rolls. The various designs differ primarily in the angles that the cutting surface and the flank face form relative to a normal of the roll surface. Different designs of corrugating rolls are used for different grinding materials and for different quality requirements. A grinding plant is equipped with suitable corrugating rolls according to its application.

[0059] Figure 12 shows a schematic representation of the characteristics of a corrugating roll with wear. Graph 900 shows the currently measured surface contour and graph 910 an ideal line, as in the Figures 10a and 10bexplained. Furthermore, distances between neighboring corrugations are defined: T1 corresponds to a distance from the base to an edge of the crushing surface 500 opposite the cutting edge 540, T2 corresponds to a distance from the base to the cutting edge 540, L corresponds to the length of the crushing surface and T corresponds to a distance from one cutting edge 540 to the next cutting edge of a neighboring corrugation. Furthermore, the corrugations are described with characteristic heights: Hb corresponds to a height from the base to the crushing surface and Ha to a height from the base to the highest point of a measured crushing surface. In addition, different rounding radii describe the wear of a contour shape: r2 describes a radius at the base and r3 a radius at the crushing surface. An angle a specifies the angle between a flank surface and the normal and an angle b specifies the angle between a cutting surface and the normal.The angles a and b are used to designate types of corrugations, as shown in . Figure 11 All of the characteristics mentioned can be detected by the sensors of the monitoring system according to the invention and used to determine the degree of wear of the corrugating roller, as described above.

[0060] To determine energy-saving potential, the analysis unit uses a mathematical model for different wheat varieties, developed based on real-world experiments using statistical experimental design on a laboratory roller mill. The model uses roller wear as a percentage of the cross-sectional area of ​​the flutes as variable A, mass flow in kilograms per hour as variable B, and rejection on an 1120 micrometer sieve as a percentage as variable C. The experiments resulted in a model for calculating the specific milling energy according to the following formula: Spez . Mahlenergie = 1.845 + 0.057 A + 10.00185 B − 0.05 C − 0.00042 AB − 8.000184 A 2

[0061] This allows the energy saving potential to be determined by comparing any worn corrugation with a new corrugation.

[0062] In Figure 13An analysis report for a corrugating roll is shown, which was created according to the monitoring method according to the invention. With this form of report, certain characteristic values ​​of the corrugating roll can be entered manually, such as the orientation of the corrugations, a type specification and dimensions of the squeezing area of ​​the corrugation type. The orientation is used by the analysis unit to insert the measured corrugation contour into the ideal contour. The type specification can be used, for example, to determine reference angles or target angles of the corrugation contour. The dimensions of the squeezing area can also serve as a reference value. The reference values ​​can be used to determine threshold values ​​for the analysis by the analysis unit. The monitoring method then determines, on the basis of this information, the measured data, the operating data and, if applicable,From specification data, the degree of wear, maintenance loss, milling yield loss, and optimized future maintenance time are determined, as described above. For example, specification data for different wheat varieties is used. In the analysis report, the analysis results can be expressed, for example, as wear as a percentage of the target value, as a ratio of measured value to target value, or by classifications, as described above.

[0063] In Figure 14 is an example of a forecast report for the analysis report from Figure 13The forecast here relates to the energy-saving potential, as described above, and is illustrated using a three-dimensional graph that graphically displays variables from the formula for the specific grinding energy. The forecast report also provides the energy costs with and without wear on the corrugating rollers. This calculation can then be used to determine the optimal future maintenance time according to the monitoring method of the invention.

[0064] In a grinding plant equipped with a monitoring system according to the present invention, unnecessary downtime due to maintenance work or material failure can be prevented, productivity and service life can be improved, and the costs of operating the plant can be transparently planned and optimized. Reference list 1 Grinding plant 410 Milling yield loss curve 2, 2.1, 2.2 grinding roller 420 Maintenance loss curve 3 Control unit 430 preventive maintenance period 4 Sensor technology 440 predicted maintenance period 4.1 ...4.8 measuring sensors 450 Maintenance period in case of failure 5 Identification means 6 Analysis unit 500 crush area 7 Cloud platform 510 open space 8 Data transmitter 520 Floor area 9 Data recipient 530 Cutting surface 10 Data pool 540 cutting edge 11 Machine Learning Module 550 Direction of rotation 12 Output unit 13 Service module 900 Contour corrugating roller 14 Statistics module 910 Ideal line corrugated roller 15 16 Calculating device TA current time 17 Feed roller TW future date 18 Feed roller 19 Data transfer 40 Wear data 50 Operational measurement data 60 Specification data 100 Ground material 110 Milled product 200 Data collection 210 Collection of specification data 220 Analysis step 230 Maintenance step 240 Update step 250 Final step 300 Maintenance loss 310 future maintenance loss 320 Loss of income

Claims

1. Process for optimized monitoring of wear on components of a grinding mill (1), particularly on grinding rolls (2) of the grinding mill, that detect an optimized future time for maintenance of the grinding mill by means of a control unit (3) for controlling the grinding mill, a sensory system (4) for detecting measurement data (40, 50) on technical parameters relating to a wear condition of a component and an operating state of the grinding mill, an analysis unit (6) to analyze wear behavior of the grinding mill and at least one data transmitter (8) and at least one data receiver (9) to exchange data, wherein the sensory system (4) detects measurement data on the wear condition (40) of at least one component (2) and on the operating state (50) of the grinding mill periodically or continuously over time, the analysis unit (6) assigns the measurement data (40) to at least one parameter on the wear condition of at least one component a degree of wear of the at least one component and the degree of wear to a maintenance loss, corresponding to a yield loss based on technical maintenance of at least one component, the analysis unit (6) derives from several pieces of measurement data (40) detected at different times a future wear condition at a time in the future and assigns the future wear condition to a future degree of wear and a future maintenance loss, the analysis unit (6) assigns the measurement data (50) on at least one parameter of the operating state of the grinding mill a grinding yield value that specifies the yield from operation of the grinding mill, and determines a grinding yield loss relative to a target yield level of the grinding mill. the analysis unit (6) derives a future operating state from several pieces of measurement data (50) detected at different times, and derives a future grinding yield loss at a time in the future therefrom and the analysis unit (6) optimizes a maintenance time (TW) in the future, by establishing as the future maintenance time (TW) that time at which the future maintenance loss corresponds to the future grinding yield loss.

2. Process for monitoring wear on components of a grinding mill in accordance with claim 1, wherein the analysis unit (6) aggregates the measurement data (40, 50) of various parameters on the wear condition of at least one component and / or on the operating state of the grinding mill by means of an aggregation module, and, on the basis of the aggregated measurement data, assigns an aggregated degree of wear to the at least one component and a relevant maintenance loss and / or assigns the operating state of the grinding mill an aggregated grinding yield loss.

3. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein the analysis unit compiles a forecast of the future maintenance loss and the future grinding yield loss by means of a regression module and / or an interpolation module.

4. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein the analysis unit (6) optimizes the future maintenance loss and the future grinding yield loss automatically by means of a module for machine learning (11).

5. Process for monitoring wear on components of a grinding mill according to the preceding claim, wherein the module for machine learning (11) sends a control signal to the control unit (3) for regulation of the operation of the grinding mill (1), wherein the operation of the grinding mill is optimized corresponding to a desired maintenance time, self-adaptively in such a way that the future maintenance time corresponds to the optimized maintenance time.

6. Process for monitoring wear on components of a grinding mill in accordance with any one of claims 4 or 5, wherein there is an optimization of the operation of the grinding mill by an optimization in relation to a parallelity of grinding rolls of a roll pair, a mutual gap of the grinding rolls and / or a rotational velocity of the grinding rolls, based on the future maintenance loss and the future grinding yield loss ascertained from the module for machine learning (11).

7. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein the sensory system (4) comprises several measurement sensors (4.1 - 4.8) detecting measurement data on a temperature of components, a roll profile and / or a roughness of grinding rolls of the grinding mill, a rotational velocity of the grinding rolls, a vibration of the grinding rolls, a gap of rolls of a grinding roll pair, energy consumption of the grinding mill, production quantity of the grinding mill, a grinding level of the material being ground, a time since last maintenance and / or downtimes of the grinding mill.

8. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein specification data from the grinding mill (1) and / or a grinding material to be ground (100) from a data pool (10) are processed to optimize the maintenance time in the future, for determining the future maintenance loss and the future grinding yield loss.

9. Process for monitoring wear on components of a grinding mill in accordance with claim 8, wherein the data pool (10) provides data relevant to a specific grinding mill, particularly data on costs for material and work for a maintenance task, energy costs for operation of the grinding mill and / or information on the grinding material (100).

10. Process for monitoring wear on components of a grinding mill in accordance with claim 8 or 9, wherein the data pool (10) provides measurement data (40, 50) on technical parameters relating to a wear condition of a component and / or an operating state of other, comparable grinding mills, for optimization of determining the future maintenance loss and future grinding yield loss by comparison and / or inclusion of these measurement data.

11. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein an output unit (12) displays visually a degree of wear of at least one component and / or a maintenance loss relating to at least one component and / or a grinding yield loss of the grinding mill (1) as a curve over time.

12. Process for monitoring wear on components of a grinding mill in accordance with the preceding claim, wherein the output unit (12) comprises a visualized, tailored dashboard format, designed in such a way that wear reports, recommended future maintenance times and information on grinding operation are displayed automatically to optimize this.

13. Process for monitoring wear on components of a grinding mill according to any one of the preceding claims, wherein the measurement data on technical parameters in relation to a wear condition of a component and an operating state of the grinding mill are loaded by means of the data transmitter (8) into a cloud platform (7) and / or the data pool with data relating to a specific grinding mill and / or data relating to other, comparable grinding mills is provided in a cloud platform (7).

14. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein a servicing module (13) based on the future maintenance loss ascertained and the future grinding yield loss automatically triggers a cycle of maintenance for the grinding mill and / or actuated by means of automated signal generation and transmission and / or an offer for maintenance of the components and / or a maintenance order is automatically compiled by means of which maintenance of the grinding mill can be triggered.

15. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein a component of the grinding mill, particularly a grinding roll, is uniquely identified by a means of identification (5) exhibiting electronically stored data relating to the component.

16. Process for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein the analysis unit (6) transfers at least the actual degree of wear and the actual grinding yield level to the control unit for automated control of the grinding mill.

17. Process of monitoring wear on components of a grinding mill in accordance with any one of the preceding claims, wherein the analysis unit (6) ascertains from measurement data (40, 50) on technical parameters relating to a wear condition of a component and / or from measurement data on the operating state of the grinding mill at least one specific maintenance task required for specific wear and a future maintenance time for this required specific maintenance task is established as that time at which the specific future maintenance loss resulting from this specific wear corresponds to the specific future grinding yield loss resulting from this specific wear.

18. System for monitoring of wear on components of a grinding mill (1), particularly on grinding rolls (2) of the grinding mill, exhibiting a control unit (3) for controlling the grinding mill (1), a sensory system (4) for detecting measurement data (40, 50) on technical parameters relating to a wear condition of a component and an operating state of the grinding mill, an analysis unit (6) to analyze wear behavior of the grinding mill and at least one data transmitter (8) and data receiver (9) to exchange data with the analysis unit, the sensory system (4) is designed in such a way that measurement data (40, 50) on the wear condition of at least one component and the operating state of the grinding mill can be detected periodically or continuously over time, the analysis unit (6) is configured to assign the measurement data (40) to at least one parameter on the wear condition of at least one component the degree of wear of the at least one component and the degree of wear to a maintenance loss, corresponding to a grinding yield loss based on technical maintenance of at least one component, the analysis unit (6) is configured to derive from several pieces of measurement data (40) detected, a wear condition at a time in the future and to assign to it a future degree of wear and a future maintenance loss, the analysis unit (6) is configured to assign the measurement data (50) on at least one parameter of the operating state of the grinding mill a grinding yield value that specifies the yield from operation of the grinding mill, and determines a grinding yield loss relative to a target yield level of the grinding mill, the analysis unit (6) is duly configured to derive a future operating state from several pieces of measurement data (50) detected at different times, and derives a future grinding yield loss at a time in the future therefrom and the analysis unit (6) is duly configured to optimize a maintenance time in the future, by establishing as the future maintenance time that time at which the future maintenance loss corresponds to the future grinding yield loss.

19. System for monitoring wear on components of a grinding mill in accordance with the preceding claim that is designed to implement the process in accordance with claims 1 to 17.

20. System for monitoring wear on components of a grinding mill in accordance with any one of preceding claims 18 and 19, wherein on a grinding roll (2, 2.1, 2.2) of the grinding mill (1) at least one temperature sensor (4.1) and a vibration sensor (4.2), a data transmitter (8) and a microprocessor are arranged to provide the measurement data on the grinding roll for the analysis unit (6).

21. System for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims 18 to 20, wherein the analysis unit (6) and / or a data pool (10) are provided with data regarding a specific grinding mill in a digital cloud platform (7).

22. System for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims 18 to 21, wherein a component of the grinding mill, particularly a grinding roll (2) exhibits a means of identification (5), exhibiting electronically stored data relating to the component for unique identification.

23. System for monitoring wear on components of a grinding mill in accordance with any one of the preceding claims 18 to 22, wherein the analysis unit (6) comprises a module for machine learning (11) and a servicing module (13), particularly also an aggregation module and / or a regression module, that is present in a cloud platform (7).

24. Grinding mill for grinding of grinding material with at least one grinding roll pair comprising a system in accordance with any one of claims 18 to 23.

25. Grinding mill in accordance with preceding claim, that is designed for grinding wheat and / or rye and / or durum wheat and / or oats and / or barley and / or peas and / or chick peas and / or pulses and / or rape and / or soya and / or cacao and / or coffee as a grinding material.