Plant for crushing metal waste and method of using the plant

Abstract

A plant for shredding metal waste, comprising at least one working chamber, means for shredding the metal waste, and motor means operatively connected to the shredding means, the plant further comprising sensor means operatively associated with the motor means and / or the shredding means for monitoring at predetermined time intervals at least one operating parameter related to wear thereof, and at least one data processing unit operatively connected to the sensor means, the data processing unit including data storage means for storing values ​​of the at least one operating parameter detected by the sensor means over a predetermined time period, and an artificial intelligence software program for enabling a plant manager to plan maintenance of the motor means and / or the shredding means.

Description

[Technical Field]

[0001] The present invention relates generally to the technical field of plant engineering and is particularly directed to a plant for crushing metal waste and a method for using the plant.

[0002] The present invention also includes a system for planning the maintenance of such a plant. [Background technology]

[0003] Conventional waste shredding plants typically perform reactive or "post-failure" maintenance, which involves technical intervention as a result of a reported failure and therefore following a failure.

[0004] This type of maintenance involves very high costs, both in terms of lost production and repairs to the machine itself.

[0005] On the other hand, in these types of plants, so-called preventive maintenance is carried out, which consists of a series of interventions scheduled at certain time intervals, in order to reduce the likelihood of breakdowns.

[0006] Again, this results in expensive maintenance due to repeated interventions whether or not the machine is aware of its condition, using resources that could otherwise be saved. Summary of the Invention

[0007] SUMMARY OF THE INVENTION It is an object of the present invention to at least partially overcome the above-mentioned drawbacks by providing a highly efficient and relatively cost-effective metal waste crushing plant.

[0008] It is an object of the present invention to provide a metal waste crushing plant which requires minimal maintenance costs and time.

[0009] Another object of the present invention is to provide a metal waste crushing plant that has a long life and is highly durable.

[0010] These objectives, as well as others that will become more apparent hereinafter, are accomplished by what is described, illustrated, and / or claimed herein.

[0011] Preferred and advantageous embodiments of the invention are defined in the dependent claims.

[0012] Further features and advantages of the present invention will become more apparent from a reading of the detailed description of a preferred but not exclusive embodiment of the plant 1, given as a non-limiting example, with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0013] [Figure 1] The scheme of Plant 1 is shown. DETAILED DESCRIPTION OF THE INVENTION

[0014] With reference to the above-mentioned figure, a plant 1 for crushing metal waste will be described below.

[0015] Such a plant 1, which is known per se, one inlet 10 for the metal waste S to be crushed; one outlet 20 for crushed metal waste S'; one working chamber 30 located between the inlet 10 and the outlet 20; a means 40 for crushing metal waste, of a type known per se, arranged in the working chamber 30; a motor means 50, such as a diesel engine or an electric motor, operably connected to the crushing means 40; Equipped with.

[0016] To predict the probability of damage to the crushing means 40 and / or the engine / motor 50 and to enable the plant operator to plan their maintenance during machine downtimes, sensor means 60 operatively associated with the motor means 50 and / or the crushing means 40 for monitoring at least one operating parameter at predetermined time intervals; at least one data processing unit 70 operatively connected to the sensor means 60; It may be possible to provide a system including:

[0017] The parameter monitored by the sensor means 60 may be any parameter as long as it is directly or indirectly related to the wear of the crushing means 40 and / or the motor means 50 .

[0018] For example, the sensor means may comprise one or more sensors operatively associated with the motor means for monitoring the energy absorption of the motor means 50 at predetermined time intervals.

[0019] Additionally or alternatively, the sensor means may advantageously comprise one or more sensors operatively associated with the motor means for monitoring the temperature of the motor means at predetermined time intervals.

[0020] Additionally or alternatively, the sensor means may suitably include one or more sensors operatively associated with the rotation shaft of the crushing means for monitoring vibrations of the rotation shaft of the crushing means at predetermined time intervals.

[0021] Additionally or alternatively, the sensor means may suitably include one or more sensors operatively associated with the rotating shaft of the crushing means for monitoring the rotational speed of the rotating shaft of the crushing means at predetermined time intervals.

[0022] If wear occurs, the above parameters may change.

[0023] The data processing unit, which may for example be a programmable logic controller (PLC) of the plant or a workstation located remotely from the plant, may comprise data storage means 71 for storing values ​​of one or more of the above operating parameters for a predetermined period of time, and an artificial intelligence software program, the artificial intelligence software program comprising instructions for performing the following steps: calculating, based on the stored values ​​in the data storage means 71, the standard deviation and the moving average of these stored values; processing, through a first neural network, a trend forecast of the operating parameter within a predetermined forecast time period based on the standard deviation and the calculated moving average; calculating, via a second neural network, a percentage probability value of damage to the motor means 50 and / or the crushing means 40 within said predicted time period based on the stored values ​​and the trend value;

[0024] This allows the plant manager to plan maintenance of the motor means and / or crushing means during downtime of the plant itself if the damage percentage probability value calculated by the artificial intelligence software exceeds a predetermined threshold.

[0025] More precisely, the artificial intelligence software consists of a multi-layer neural network with a recursive structure. This software uses as input data the historically consecutively received values ​​from the sensor means 60 (connected to the input layer with a neuron-sensor ratio of 1:1). The collected data, even in large quantities, are processed by a normalization layer, which makes it possible to obtain the necessary statistical parameters, such as the above-mentioned moving average and standard deviation. Such statistical information, with the above-mentioned ratio greater than or equal to 2 for each sensor, is transmitted to the first neural network using an auxiliary input layer, which is directly connected to the second hidden layer of the software.

[0026] Thus, the first neural network makes a prediction, and the shorter the prediction time period, the more accurate the prediction itself.

[0027] At this point, the generated trend values ​​are transferred to a second neural network with a classifier function (with a multi-layer structure based on deep learning levels), which has the task of refining a binary classification of each of the motor means 50 and / or crushing means 40 parts with respect to their failure within the period considered. The software output is a percentage value for the probability of failure.

[0028] Because machine failures are potentially infinite, the second model is trained not only on maintenance / failure history, but also on the standard deviation for the expected moving average as an error minimization function, allowing for the identification of correlations between sensor outliers (or linear or non-linear combinations thereof) and the probability of failure.

[0029] From the above, it is clear that the present invention achieves its objects.

[0030] The present invention is susceptible to numerous modifications and variations: all the details may be replaced by other technically equivalent elements and the materials may be varied according to requirements, without departing from the scope of the invention as defined in the appended claims. DISCLOSURE OF THE INVENTION (Section 1) A plant for crushing metal waste, comprising: at least one inlet for metal waste to be shredded; at least one outlet for shredded metal waste; at least one working chamber disposed between the at least one inlet and the at least one outlet; means for crushing metal waste disposed within said working chamber; motor means operably connected to said crushing means; sensor means operatively associated with said motor means and / or said crushing means for monitoring at predetermined time intervals at least one operating parameter related to wear of said motor means and / or said crushing means; at least one data processing unit operatively connected to said sensor means; Equipped with The at least one data processing unit data storage means for storing values ​​of said at least one operating parameter sensed by said sensor means for a predetermined period of time; 1. An artificial intelligence software program comprising: calculating a standard deviation and a moving average of the stored values ​​based on the stored values; processing a trend forecast of the at least one operating parameter within a predetermined forecast time period based on the standard deviation and the calculated moving average; calculating a percentage probability value of damage to said motor means and / or said crushing means within said predetermined prediction time period based on said stored values ​​and said trend value; an artificial intelligence software program including instructions for executing the Including, enabling a plant manager to schedule maintenance of the motor means and / or the crushing means during the same plant downtime if the damage percentage probability value exceeds a predetermined threshold; plant. (Section 2) Item 1. The plant of item 1, wherein the sensor means includes at least one first sensor operatively associated with the motor means for monitoring energy absorption of the motor means over a first predetermined time interval. (Section 3) 3. The plant of claim 1 or 2, wherein the sensor means includes at least one second sensor operatively associated with the motor means for monitoring the temperature of the motor means over a second predetermined time interval. (Section 4) Item 4. The plant of claim 1, 2 or 3, wherein the crushing means includes at least one rotating shaft, and the sensor means includes at least one third sensor operatively associated with the at least one rotating shaft for monitoring vibrations of the at least one rotating shaft over a third predetermined time interval. (Section 5) 5. The plant of any one of claims 1 to 4, wherein the crushing means includes at least one rotating shaft, and the sensor means includes at least one fourth sensor operatively associated with the at least one rotating shaft for monitoring the rotational speed of the at least one rotating shaft over a fourth predetermined time interval. (Section 6) 6. The plant of any one of claims 1 to 5, wherein the step of processing the forecasted values ​​is performed by a first neural network. (Section 7) 7. The plant of any one of claims 1 to 6, wherein the step of calculating the damage percentage probability value is performed by a second neural network. (Section 8) A method for using a metal waste crushing plant according to any one of items 1 to 7, shredding the metal waste for a first predetermined period of time; turning off the system for a second predetermined period of time; performing maintenance on said motor means and / or said crushing means during said second predetermined time period; Including, the maintenance step is performed only if the damage percentage probability value exceeds a predetermined threshold. method. (Section 9) 1. A system for planning the maintenance of a plant for crushing metal waste, said plant comprising: at least one inlet for metal waste to be shredded; at least one outlet for shredded metal waste; at least one working chamber disposed between the at least one inlet and the at least one outlet; means for crushing metal waste disposed within said working chamber; motor means operably connected to said crushing means; Equipped with The system comprises: sensor means operatively associated with said motor means and / or said crushing means for monitoring at least one operating parameter related to wear of said motor means and / or said crushing means over a predetermined time interval; at least one data processing unit operatively connected to said sensor means; Including, The at least one data processing unit data storage means for storing values ​​of said at least one operating parameter sensed by said sensor means for a predetermined period of time; 1. An artificial intelligence software program comprising: calculating a standard deviation and a moving average of the stored values ​​based on the stored values; processing a trend forecast of the at least one operating parameter within a predetermined forecast time period based on the calculated standard deviation and the moving average; calculating a percentage probability value of damage to said motor means and / or said crushing means within said predetermined prediction time period based on said stored values ​​and said trend value; an artificial intelligence software program including instructions for executing the Including, enabling a plant manager to schedule maintenance of the motor means and / or the crushing means during downtime of the plant itself if the damage percentage probability value exceeds a predetermined threshold; system.

Claims

1. A plant for crushing metal waste, comprising: at least one inlet (10) for the metal waste (S) to be crushed; At least one outlet (20) for shredded metal waste (S'); at least one working chamber (30) disposed between said at least one inlet (10) and said at least one outlet (20); a crushing means (40) for crushing the metal waste disposed in the working chamber (30); a motor means (50) operatively connected to said crushing means (40); sensor means (60) operatively associated with said motor means (50) and / or said crushing means (40) for monitoring at predetermined time intervals at least one operating parameter related to wear of said motor means (50) and / or said crushing means (40); at least one data processing unit (70) operatively connected to said sensor means (60); Equipped with the crushing means (40) comprises at least one rotating shaft, and the at least one operating parameter is one or more of the temperature of the motor means (50), the vibration of the at least one rotating shaft, and the rotation speed of the at least one rotating shaft; The at least one data processing unit (70) data storage means (71) for storing values ​​of said at least one operating parameter sensed by said sensor means (60) for a predetermined period of time; 1. An artificial intelligence software program comprising: calculating a standard deviation and a moving average of the stored values ​​based on the stored values; processing a trend forecast of the at least one operating parameter within a predetermined forecast time period based on the standard deviation and the calculated moving average; calculating a percentage probability value of damage to said motor means and / or said crushing means within said predetermined prediction time period based on said stored values ​​and said trend prediction value; an artificial intelligence software program including instructions for executing the Including, enabling a plant manager to schedule maintenance of the motor means and / or the crushing means during the same plant downtime if the damage percentage probability value exceeds a predetermined threshold; plant.

2. 2. The plant of claim 1, wherein said sensor means (60) includes at least one first sensor operatively associated with said motor means (50) for monitoring said motor means (50) for a first predetermined time interval.

3. 3. The plant of claim 1, wherein the sensor means (60) includes at least one second sensor operatively associated with the motor means (50) for monitoring the temperature of the motor means (50) for a second predetermined time interval.

4. 4. The plant of claim 1, 2 or 3, wherein the sensor means (60) includes at least one third sensor operatively associated with the at least one rotating shaft for monitoring vibrations of the at least one rotating shaft over a third predetermined time interval.

5. 5. The plant of claim 1, wherein the sensor means (60) includes at least one fourth sensor operatively associated with the at least one rotating shaft for monitoring the rotational speed of the at least one rotating shaft over a fourth predetermined time interval.

6. 6. The plant of claim 1, wherein the step of processing the trend forecast value is performed by a first neural network of the artificial intelligence software program.

7. 7. The plant of claim 1, wherein the step of calculating the damage percentage probability value is performed by a second neural network of the artificial intelligence software program.

8. 8. A method for using a plant for crushing metal waste according to any one of claims 1 to 7, comprising: shredding the metal waste for a first predetermined period of time; turning off the plant for a second predetermined period of time; performing maintenance on said motor means (50) and / or said crushing means (40) during said second predetermined time period; Including, the maintenance step is performed only if the damage percentage probability value exceeds a predetermined threshold. method.

Images