CN117707054A - Intelligent digital controller upper box and implementation method thereof - Google Patents

Abstract

The invention provides an intelligent digital control machine upper box, which can capture machine data of at least one machine in an operation state, filter and compare the machine data, continuously provide monitoring information of the working progress and the machine state of the machine based on the collected machine data, and intelligently analyze the machine data according to historical processing data to provide predictive information about the working progress and the machine state, analytical information about abnormal phenomena of the machine and decision information about productivity and exchange period. A method for implementing the upper box of the intelligent digital controller is also provided.

Description

Intelligent digital controller upper box and implementation method thereof

Technical Field

The invention relates to a technology of distributed operation, cloud operation and machine processing behavior analysis, in particular to an intelligent digital control machine upper box which is beneficial to a factory manager to carry out production line decision information and an implementation method thereof.

Background

The service life and performance state of the parts of the machine may be faulty due to the use of the machine for years, and when the machine has abnormal conditions, maintenance personnel must be arranged to inspect or maintain the machine, so that productivity and exchange period are affected, and the problem that the rest of processed products need to be scrapped may be caused. In addition, in the product manufacturing process, the factory end usually aims at productivity, exchange period and cost, if the operation schedule, raw materials, personnel, production lines and preparation/modification production lines of the factory can be properly arranged and controlled, the factory operation can be effectively controlled, and how to decide the exchange period recovery and the customer satisfaction degree of the factory manager or the first line business can be also facilitated. Along with the trend of industry 4.0 and the development of intelligent machinery industry, how to use production and manufacturing management and manufacturing services as guidance, and from the perspective of factory managers, a technical means is provided for assisting factory managers in stably arranging production/operation scheduling, and simultaneously predicting cost, bottleneck, productivity and exchange period in advance before order receiving or before working, and also can effectively know the fundamental problem affecting production efficiency/quality, which is a problem to be solved.

Disclosure of Invention

The invention aims to provide an intelligent digital control machine upper box and an implementation method thereof, wherein each machine upper box can collect machine data of a filter machine, and can carry out big data intelligent analysis based on historical processing and overhaul experience after integration, comparison and statistics, thereby being beneficial to a factory manager to carry out production line decision information.

In order to achieve the above purpose, the invention discloses an intelligent digital control machine upper box. The upper box of the intelligent digital controller comprises a processor, and further comprises a data acquisition module, a memory, a communication module and an input module which are respectively and electrically connected with the processor. The input module can receive an input instruction of a user. The data acquisition module can acquire one machine data of at least one machine in an operation state based on the input instruction. The communication module can selectively establish a wired or wireless communication connection with the monitored machine using a first communication protocol. The processor can filter and compare the collected machine data, so that the working progress and the monitoring information of the machine state of the monitored machine are continuously updated based on the collected machine data, and the communication module can selectively receive the machine data and the monitoring information of other machines from other machine boxes by using the first communication protocol.

On the other hand, the machine-box processor may perform an intelligent analysis based on a machine learning algorithm based on a historical processing data (or a combination thereof and monitoring information) stored in the memory to provide a predictive information associated with the machine work schedule and the machine status, to provide an analytical information (i.e., one or more important factors) associated with the machine abnormality or affecting the machine work schedule, and to provide a decision information that may include any combination of recommended schedule, capacity, and schedule based on the historical processing experience reflected by the predictive information and the analytical information.

After the invention is implemented, a user can monitor the state of the machine, the performance and the service life of the material parts and the production/operation progress in real time, and can realize the beneficial effects of knowing the characteristics of the machine with various specifications and different years, realizing stable arrangement of capacity and exchange period, realizing the pre-known warning of the machine needing to be maintained or maintained, and simultaneously, leading a factory manager or business to have higher mastery degree and credibility on production/operation scheduling based on an experience model for continuously collecting historical processing experience and historical maintenance experience and using the processing experience for recycling.

The invention can assist the manager/decision maker at the factory end to decide whether to order and arrange the key of operation scheduling by referencing the history processing and history maintenance experience, and can help the user to efficiently find out the real reasons that the production/processing efficiency is not as expected or the abnormal phenomenon of the machine occurs in the operation/processing process by rapidly providing the guidance of maintenance or problem elimination, thereby reducing the manpower and time of the factory end for investigating the real reasons that the production/processing efficiency and quality are influenced.

In one embodiment, the processor of the on-board box can provide a review instruction for the user to input the module or from an external device (such as a mobile phone), and provide the user with review of the historical processing data stored in the memory of the on-board box to find out the historical processing experience with the best processing quality for the same or similar requirements.

In one embodiment, the communication module of the on-board box can be in communication connection with the server by using a second communication protocol so as to compare and analyze the machine data, the collected monitoring information, the predictive information, the analytical information and the decision information based on the machine data, the historical processing data and the historical maintenance data from an external data source by the server, thereby disposing the optimized machine learning algorithm on one or more on-board boxes at regular or irregular time to improve the accuracy of the analysis result calculated after each on-board box executes the intelligent analysis.

In one embodiment, the memory of the set-top box stores historical maintenance data, and when the monitoring information of the monitored machine is determined by the processor to contain abnormal phenomena, the processor can calculate the analysis information related to the machine fault or abnormal phenomena based on statistical data such as maintenance exclusion knowledge, failure mode knowledge, failure phenomenon experience knowledge and the like in the historical maintenance data.

In one embodiment, the continuously updated monitoring information can be continuously input into the data set of the historical processing data along with the operation process of the machine, so as to improve the accuracy of calculating predictive, analytical and decision-making information of each machine box.

The invention also provides an implementation method based on the intelligent digital control computer upper box, a computer readable recording medium and a computer program product. The implementation method based on the intelligent digital control machine upper box is used for monitoring and collecting machine data of at least one machine, and comprises the following steps:

a processor executing an intelligent analysis based on a plurality of historical processing data stored in a memory, calculating predictive information associated with one or a combination of a work progress and a machine status of the machine;

the processor performing the intelligent analysis based on the historical process data to calculate an analytical information associated with an anomaly of the tool or affecting a scheduling of operations of the tool, the analytical information including one or more importance factors; and

the processor performing the intelligent analysis based on the predictive information and the analytical information to calculate decision information that is one or a combination of a recommended schedule, a capacity, and a date of delivery; wherein the method comprises the steps of

The processor can drive a data acquisition module to acquire a machine data from the machine in an operation state through a first communication protocol, and the machine data can be input into a data set of the historical processing data after being compared and collected into monitoring information.

In one embodiment, the processor continuously updates the predictive information, the analytical information, and the decision information while the machine is in operation, based on a combination of the historical process data and the monitoring information while the processor is performing the intelligent analysis.

In one embodiment, when the processor determines that the monitored monitoring information of the machine includes the anomaly, the processor calculates the analytical information including at least one of an anomaly type or a true cause classification tag based on a historical overhaul data stored in the memory and associated with a plurality of the machines, wherein the historical overhaul data includes one or a combination of a maintenance exclusion knowledge, a failure mode knowledge, and a failure experience knowledge.

In one embodiment, when the processor performs the intelligent analysis, a second communication protocol is used to establish a communication connection with a server at regular or irregular time to compare and analyze the machine data, the monitoring information, the predictive information, the analytical information and the decision information by the server based on the machine data, the historical processing data and the historical maintenance data from an external data source, so that an optimized machine learning algorithm can be deployed on one or more intelligent digital control machines through the second communication protocol.

In one embodiment, the processor responds to a review request such that the processor performs the intelligent analysis based on the historical process data to calculate the predictive information, the analytical information, and the decision information before the data acquisition module acquires the machine data of the machine in the operating state.

For the purpose of making clear the objectives, technical features and effects of the present invention, reference will be made to the following description in conjunction with the accompanying drawings.

Drawings

FIG. 1 is a block diagram of an embodiment of the present invention;

FIG. 2 is a flow chart of a method according to an embodiment of the invention;

FIG. 3 is a schematic diagram of a scenario according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a scenario (II) according to an embodiment of the present invention;

fig. 5 is a schematic diagram of a scenario (iii) according to an embodiment of the invention.

Description of the reference numerals

10. Intelligent digital controller upper box

101. Processor and method for controlling the same

102. Data acquisition module

103. Memory H_W historical process data

M_U monitoring information

H_R historical overhaul data

P predictive information

A analytical information

D decision information

104. First communication protocol of communication module P1

P2 second communication protocol

105. Input module

106. Output module

M-M ' ' ' machine C controller

S-sensor

M_D machine data

CS server

Implementation method of S intelligent digital control machine upper box

S1, carrying out relevance analysis or important factor analysis according to historical processing data

S2, providing a prediction result related to the working progress and the machine state

S3 providing analysis results related to abnormal phenomena or influence scheduling

S4, providing decision information which is helpful for scheduling capacity, schedule and operation based on the predictive and analytic information.

Detailed Description

Referring to fig. 1, an intelligent digital controller upper box 10 is capable of monitoring and collecting machine data m_d of at least one machine M, the intelligent digital controller upper box 10 includes a processor 101, a data acquisition module 102, a memory 103, a communication module 104, an input module 105 and an output module 106 respectively connected to the processor 101 in communication, and the communication module 104 is also connected to the data acquisition module 102 in information.

Referring to fig. 1, and referring to the context diagrams of fig. 3 to 5, in an embodiment, the data capturing module 102 may capture the machine data m_d (such as the operation information of the rotating speed, the time, the load, the processing program, etc.) of the machine M in the operating state from one or a combination of the controller C and the sensor S of the machine M, the processor 101 may filter and compare the machine data m_d collected by the data capturing module 102, so as to continuously update a monitoring information m_u of the working schedule of the machine M in the operating state and the machine state based on the machine data m_d after the compiling, and the processor 101 may also provide an abnormality prediction information m_u of the working schedule based on a machine learning algorithm (such as a decision tree, a cluster analysis, a reinforcement learning, a bellevil network, etc.), but not limited thereto, to provide an abnormality prediction information of the working schedule of the machine M in the operating state, a prediction model, or a combination of the abnormality information providing the abnormality information of the working schedule of the machine M in the operating state, a prediction model, or a combination of the prediction model providing the abnormality information.

Referring to fig. 2, a method S for implementing an upper box of an intelligent digital controller according to the present invention is applied to an upper box 10 of an intelligent digital controller, and referring to fig. 1, the method S includes the following steps.

In step S1 (performing correlation analysis or importance factor analysis according to the historical processing data), the information capturing module 101 of the upper box 10 captures the processing data m_d of the processing operation state from the machine table M according to the first communication protocol P1, the processor 101 of the upper box 10 further filters and compares the machine table data m_d collected by the information capturing module 102, so as to continuously update the monitoring information m_u of the working schedule and the machine table state of the machine table M based on the collected machine table data m_d, and the processor 101 may further perform an intelligent analysis including at least one correlation analysis and one importance factor analysis based on the combination of the historical processing data h_w or the historical processing data h_w and the monitoring information m_u of the machine table M stored in the memory 103 according to the machine learning algorithm.

On the other hand, when the step S1 is executed, the continuously updated monitoring information m_u is continuously input to the data set of the historical processing data H along with the processing/operation process of the machine station M, so as to improve the reliability of the intelligent analysis performed by the upper box 10 of the intelligent digital controller.

In step S2 (providing a prediction result regarding the working schedule and the machine state), the processor 101 calculates the predictive information P related to one or a combination of the working schedule and the machine state of the machine M according to the historical processing data h_w (or the combination of the historical processing data h_w and the monitoring information m_u) through the analysis logic of the historical processing experience reuse.

In step S3 (providing analysis results on anomalies or influencing schedules), the processor 101 calculates analysis information a of a work schedule associated with the machine M according to the historical processing data h_w (or a combination of the historical processing data h_w and the monitoring information m_u) by using the analysis logic of the historical processing experience reuse.

In step S4 (decision information for facilitating scheduling of capacity, lead time, and work based on predictive and analytical information), the processor 101 calculates decision information D, which may be one of recommended schedule, capacity, and lead time, or a combination thereof, based on the combination of predictive information P and analytical information a, in addition to the historical process data h_w (or the combination of the historical process data h_w and the monitor information m_u).

Referring to fig. 1 again, in one embodiment, the continuously updated monitor information m_u may be continuously input to the (training) dataset of the historical processing data h_w along with the operation/processing process of the machine M, so as to improve the accuracy of calculating the predictive information P and the analytical information a by the smart digital controller upper box 10.

For example, the machine tool M may be a multi-axis machine tool, a lathe, a milling machine, a welding machine, a robot module, or the like, but is not limited thereto.

For example, the controller C may be disposed on the machine M, and the controller C may be a PLC controller (Programmable Logic Controller) or a CNC controller (Computer Numerical Control), but is not limited thereto.

The sensor S may be disposed on a main shaft or a key element of the machine M, or outside the machine M, or the environment of the machine M is one of a proximity sensor, a photo-sensor, a laser displacement meter, a zone sensor, a pressure sensor, a vision sensor, a gas flow meter, a temperature sensor, or a combination thereof, but is not limited thereto.

As shown in fig. 5, the monitoring information m_u related to the working progress of the machine M may be, for example, one of a processing mode, a current total working time of the machine, an oil usage time, an actual processing time, a residual processing time, a station changing time, or a combination thereof, but is not limited thereto.

As shown in fig. 5, the monitoring information m_u related to the machine state of the machine M may be, for example, one of a spindle load monitor, a servo load monitor, a spindle rotation speed monitor, a component life monitor, a machine I/O state, a utilization rate, a power consumption monitor, or a combination thereof, but is not limited thereto.

As an example, as shown in fig. 5, the aforementioned device lifetime monitoring data may be, for example, a proximity switch, an oil pressure on/off switch, a locking button, a pressure switch, a speed switch, an indicator/warning lamp, a buzzer, a relay (the type and number of devices can be automatically adjusted and defined by a user according to the specification of the machine, so that the device is not limited thereto), or a usage count, a signal generation/generation count, or a remaining lifetime count of an input signal (for example, an oil pressure/spindle/electric clamp start signal, a frequency converter fault signal, an inching signal, a front-back continuous signal, etc., but not limited thereto).

As an example, as shown in fig. 3, the historical processing data h_w may be a historical processing time history, which may be one of a historical client, a historical product name, an order number, a work order number, a lot number, a change number, a historical stock number, a historical processing time, a historical station exchange time, and a historical processing parameter, or a combination thereof, but is not limited thereto.

Referring to fig. 1 and 4, in one embodiment, the communication module 104 may establish a communication connection with a communication module (not shown) of other intelligent digital control boxes (not shown) using a first communication protocol P1, so that each box can transmit and receive the machine data m_d extracted from the machine (M, M ', M' …) and the collected monitoring information m_u to each other, and the communication module 104 may selectively establish a wired or wireless communication connection with the monitored machine M using the first communication protocol P1, so that the data acquisition module 102 may acquire the machine data m_d of the machine M.

For example, the output module 106 may be a display screen (providing touch function), and may present the machine data m_d, the monitor information m_u, the predictive information P, the analytical information a, and the decision information D of the machine M in a manner of, for example, a graphical user interface and a statistical icon.

As an example, as shown in fig. 5, the predictive information P associated with the work progress of the machine M may be, for example, one of a predicted cycle time and a predicted residual processing time, or a combination thereof, but not limited thereto.

For example, the predictive information P related to the machine states of the machines M with the same or different specifications may be one of or a combination of a component lifetime warning, a machine lifetime warning, a bottleneck device warning, a machine maintenance timing and a part maintenance timing of the machines M. Therefore, the method can help a factory manager to diagnose and schedule the maintenance time and the part replacement time of the machine in advance, reduce the loss of maintenance and production cost caused by unexpected faults and shutdown, reduce unexpected shutdown and maintenance frequency, and simultaneously help to promote and stabilize the benefit of production line scheduling.

For example, the analysis information a of the abnormal phenomenon associated with the machine M may be that the processor 101 performs a correlation analysis on the monitoring information m_u generated during each operation/processing process of the machine M to calculate one or more important factors that help to investigate the production efficiency and quality.

For example, taking the abnormal situation as "poor processing quality" and taking the "actual processing time" and the "change time" included in the monitor information m_u as an example, if a production process involves three stations, the actual processing time of the machine M of each station is 10 minutes (i.e. the actual processing time is the same), and the change time of the station 2 to the station 3 is 3 minutes, but because the change time between the station 1 and the station 2 is 5 minutes (more than the change time between the station 2 to the station 3 by 2 minutes) due to the unknown reasons, after the processor 101 performs the correlation analysis through the statistics result of the historical processing data h_w, the analytical information a of "the change time affects the processing quality and the productivity" can be calculated, which is helpful for the manager to make a decision of adjusting the station position early or confirm whether the two stations are far from each other as soon as possible.

In one embodiment, when the processor 101 determines that the monitored monitoring information m_u of the machine M includes the abnormal phenomenon related to the machine M failure or the operation abnormality, the processor 101 may also perform an important factor analysis based on the multiple machine data m_d and the multiple monitoring information m_u generated by the machine M during each operation/processing process to calculate the actual cause (i.e. one or more important factors) of the abnormality or the failure occurring during the production process.

On the other hand, as illustrated in fig. 5, if the anomaly is taken as an example of "the processing error ratio is not proportional to the processing time", for example, the processing time is 20 minutes, the error is 5% (the average or median) of the machine M of the a specification, but if the processing time is 40 minutes, the error is not 10% (5% ×2) but 15% of the same machine M, the processor 101 analyzes the important factor through the statistical result of the historical processing data h_w, and after the analysis, the manager can replace the important factor by the important factor, and can schedule the machine M of the a specification or the machine M of the a specification with the estimated value of the anomaly according to the fact that the service life of the machine M is still normal and the parameter setting is normal, because the service life of the machine M is 10 years, the oil pressure valve is exhausted according to the processing experience, and because the frequency of the Z axis (i.e. before and after the processing) is particularly high or the number of times is particularly high, each time of the Z axis processing is caused, the delay is increased with the increase of the processing time, and the total processing time is affected, and the manager can replace the important factor of the anomaly by the important factor when the machine M is not replaced by the historical processing data h_w, and the machine M is estimated to be replaced by the machine M, and the service life of the quality is stable.

Similarly, the ratio of the processing time to the error of the "a-specification and 15-year-old machine M" may be due to an abnormal change (for example, the error is not 15% (5% ×3) but 25%) which is caused by the operation for 60 minutes, the machine characteristics which are difficult to change or are not to be improved for a short time, the productivity and the date of exchange may be estimated (if the machine M is not scheduled for maintenance or elimination), and the aforesaid machine characteristics are not limited to the error, i.e. "the X-part of the a-specification machine M may possibly be unstable during a certain operation (production/processing) stage", and other known unstable machine characteristics may be estimated (if the machine M is not scheduled for maintenance or elimination).

On the other hand, taking the abnormal phenomenon as an example of "the vibration amplitude of the machine tool structure is large, and the ripple is poor", after the processor 101 performs the analysis of the important factor according to the statistical result of the historical processing data h_w, the user knows that although the source of the vibration of the machine tool M is generally the mismatch of the spindle, the rotation speed or the cutting depth, the real reason is irrelevant to the parts and the performance of the machine tool M (for example, the result shows that the service life of the element is still long, the rotation speed monitoring is normal, the parameters are the same as the normal settings of the machine tool M with other specifications similar to those of the year), and on the contrary, the processor 101 can know that the floor thickness at the place where the machine tool M is located is not enough (too thin) may be one important factor causing the abnormal phenomenon according to the historical processing experience, so the processor 101 can provide a suggestion that the user moves the machine tool M to the place where the floor is thicker or more resistant to the vibration, thereby being helpful for improving the processing efficiency and quality.

On the other hand, taking the abnormal phenomenon as "the temperature of the machine is overheated" as an example, after the processor 101 analyzes the important factor according to the statistical result of the historical processing data h_w, the user can understand that, although the reason for the abnormal temperature of the machine is generally the problem that the cutting force or the rotation speed is not fixed, the reasons are irrelevant to the processing parameters of the machine M (for example, the result display parameters are the same as the normal setting of the machine M with other specifications similar to those of the years), and on the contrary, the processor 101 can know according to the historical processing experience that the ratio of the amount of oiling (generating heat) to the amount of water (for cooling) is not equal to the ratio of the amount of oiling (generating heat) to the amount of water (for cooling) during the groove cutting, which may be one important factor causing the abnormal phenomenon, so that the manager can educe or remind the operator to pay attention to the use ratio of the oil and the water without the need to stop the maintenance of the machine M.

Taking the abnormal phenomenon as an example of "abnormal or fault of the motor", although unstable voltage may be a common cause, "according to processing experience, it may be unstable power supply of the factory" is also an important factor causing the abnormal phenomenon, and a manager can confirm whether the power supply source of the factory has any problem or not without stopping to arrange maintenance of the machine M or purchasing a voltage stabilizer first according to the important factor on the premise of not replacing the motor.

Referring to fig. 1 to 5, in an embodiment, if the processor 101 confirms that the abnormal situation is related to a machine part or maintenance, the processor 101 may calculate, based on a combination of the monitoring information m_u and the historical maintenance data h_w, further based on a historical maintenance data h_r stored in the memory 103 and associated with a plurality of machines M, an analysis information a including one or a combination of an abnormal cause type and a true cause classification tag, wherein the historical maintenance data h_r may include one or a combination of a maintenance exclusion knowledge, a failure mode knowledge and a failure phenomenon experience knowledge, and the historical maintenance data h_r may be packaged as a pre-trained maintenance knowledge model.

For example, if the anomaly is "A, B, C gauge grinding machine, left-right anomaly" occurs, and the processor 101 considers that the anomaly is related to machine part and maintenance according to the historical maintenance data h_r, possible important factors should be found according to the historical maintenance data h_r to find out the real cause. For this, the processor 101 can list five abnormal cause types of "left and right hit single side 40.28%", "left and right leak oil 14.78%", "left and right slow speed 9.72%", "left and right hit machine 8.70%", "left and right abnormal sound 6.88%", and "left and right abnormal sound 6.28%", from the maintenance list statistics of the historical maintenance data h_r, whereby the user can learn from these ratios that the higher probability should be attributable to "left and right hit single side".

Then, the processor 101 may further compare and collect the distribution of the "true cause classification labels" corresponding to the "TOP 3 anomaly cause", for example, if the "side to side collision single side" occupying the highest proportion is taken as an example, the proportion of the "electrical component failure, damage" and the "hydraulic valve damage, oil leakage" true cause classification labels is higher, and the proportion of the "wire breakage, damage", "inaccurate grinding", "oil cylinder oil leakage", "manual transmission mechanism" and the remaining true cause classification labels corresponding to the "side to side collision single side" is obviously not higher, so that the user can learn from the proportions that the higher probability is attributable to the "electrical component failure, damage" or "hydraulic valve damage, oil leakage".

Next, the processor 101 can further compare and integrate the distribution of, for example, "TOP 3" and "repair/replace/maintain parts" in the real cause of classification label "and" repair/replace/maintain parts ", for example, it corresponds to the part classification of" near switch (20.24%) "and" relay (15.79%) "," solenoid valve (11.34%) "" and "PLC (1.21%)", whereby the user can learn from these ratios that the higher probability should be attributable to "near switch", "relay" and "solenoid valve", then the processor 101 can further compare and integrate the distribution of, for example, "A, B, C specification" and "part replacement status", for example, a specification is high and low in proportion, which can correspond to the part classification of "near switch (10.73%)", "relay (10.32%)" "and" solenoid valve (7.89%) ", and so on-demand", B is high and low in proportion, which can correspond to the "near switch (3.64%)" "and" relay (2.63%) "" and "relay (3.62%)" "and" relay (7.9%) "" and so on), which can be used as a guide the direct to the cause of the service-life of the three causes of failure and so on) respectively. It does not take other time and labor to carry out detailed maintenance on the machine M to find out where the real cause is, but the above analysis dimension of the statistical data is only an example and not a limitation.

Referring to fig. 1, in an embodiment, the communication module 104 may also establish a communication connection with a server CS by using a second communication protocol P2 to collect the machine data m_d, the monitoring information m_u, the predictive information P and the analytical information a from the processor 101, and perform comparison and analysis by the server CS based on the machine data m_d, the historical processing data h_w and the historical maintenance data h_r from an external data source through, for example, an internet, so that the optimized machine learning algorithm is deployed on one or more intelligent digital control boxes 10 at regular or irregular time through the second communication protocol P2 (different from the industrial area network applied by the first communication protocol P1), and in addition, each digital control box 10 may serve as an Edge Node (Edge Node) at the factory end to collect, filter and analyze the machine data m_d so as to reduce the waiting time of the communication module CS (for example, which is used by the Edge Computing) from the Edge Node, and reduce the waiting time of the communication module CS (for example, the server with wide bandwidth and the server.

Referring to fig. 1, in an embodiment, the processor 101 may also respond to a review request from the input module 105 or an external device (e.g. smart phone, tablet, personal computer, notebook computer) to the communication module 104, so that the data capturing module 102 may perform an intelligent analysis based on at least the historical processing data h_w (or a combination thereof with the historical maintenance data h_r) before capturing the machine data m_d of the machine M in the working state (e.g. before determining whether to take a bill or not, or before the machine M has not started to process), and thus find a historical processing experience with better processing efficiency and quality, and calculate the predictive information P, the analytical information a and the decision information D related to the working progress and the machine state of the machine M in advance.

Referring to fig. 1, in an embodiment, the smart digital controller upper box 10 may also be connected in series with a management system adopted by a factory side or an enterprise side such as ERP (Enterprise resource planning), SAP, MES (Manufacturing Execution System), WMS (Warehouse Management System), etc. through the communication module 104.

In addition, the functions and contents of the above technical features are described above, and are not repeated here, in which the above-mentioned intelligent analysis, historical processing data h_w, predictive information P, analytical information a and decision information D are mentioned in the implementation method S of the upper box of the intelligent digital controller.

Referring to fig. 1 to 2, in an embodiment, the present invention further provides a non-transitory computer readable recording medium, which is associated with at least one instruction to define the implementation method S of the smart digital controller upper case, and the relevant description of each step is detailed above and will not be repeated herein.

Referring to fig. 1 to 2, in an embodiment, the present invention further provides a computer readable recording medium associated with at least one instruction to define the implementation method S of the smart digital controller upper box, and the relevant description of each step is detailed above and will not be repeated herein.

Referring to fig. 1 to 2, in an embodiment, the present invention further provides a computer program product, after the computer system loads a plurality of instructions of the computer program product, at least the implementation method S of the smart digital controller upper box described above can be completed, and the detailed description of each step is omitted herein.

For example, the server CS of the present invention may be one or more independent server computers providing connection services, or a server running in the form of a Virtual Machine (Virtual Machine), or a server running in the form of a Virtual proprietary host (Virtual Private Server), or a public cloud, or a private cloud, but not limited thereto.

For example, the processor 101 of the present invention has functions of logic operation, temporary storage of operation results, and storage of data operation instruction positions, and may include, but not limited to, a single processor and an integration of multiple microprocessors, such as a Central Processing Unit (CPU), a virtual processor (vCPU), a Microprocessor (MPU), a micro-controller (MCU), an Application Specific Integrated Circuit (ASIC), a Programmable Logic Device (PLD), a Tensor Processor (TPU), a Digital Signal Processor (DSP), and the like.

As an example, the communication module 104 of the present invention may be applied to various communication service interfaces, such as one or any combination of a bluetooth communication unit, a WLAN communication unit, a mobile communication unit, an NFC communication unit, a ZigBee communication unit, a Z-Wave communication unit, and a UWB communication unit, where the mobile communication unit may be applied to wireless communication interfaces of 2G, 2.5G, 3G, 3.5G, 4G LTE, and 5G, but not limited thereto.

For example, the memory 103 of the present invention may be eMMC (embedded MultiMedia Card) flash memory, UFS (Universal Flash StoR _age) flash memory, NVMe (NVM Express) flash memory, solid state drive (solid state drive/disk), LPDDR (Low Power r_ Andom Access Memory) memory, dynamic random access memory (dr_am) or static random access memory (sr_am), and if it is a non-transitory (tr_address) computer readable medium, the memory may further store at least one instruction of implementing method S associated with the smart digital controller upper box, which is accessible and executable by the processor 101.

The first protocol P1 of the present invention may be EIA-485/RS485, CAN/CAN Bus (Controller Area Network) protocol; wi-Fi architecture-based WIA-PA, haLow Wi-Fi (IEEE 802.11 ah), wiGig (IEEE 802.11 ad) wireless communication protocol; wireless communication protocols based on IEEE 802.15.4 standards, such as 6LoWPAN, wirelessHART, zigBee; a Bluetooth Low Energy (BLE) wireless communication protocol; loRA (remote wide area network modulation) protocol; the Sub-GHz solution based LoRa, NB-IoT, 6TiSCH protocols, etc. are one or any combination of the above listed wireless protocols, but the above are only examples and not limiting.

The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention; any equivalent changes and modifications within the spirit and scope of the present invention will occur to those skilled in the art from this disclosure.

Claims (9)

1. An intelligent digital control machine top box for monitoring and collecting machine data of at least one machine, the intelligent digital control machine top box comprising:

the processor comprises a data acquisition module, a memory and a communication module which are respectively and electrically connected with the processor, wherein the communication module establishes communication connection with the machine by using a first communication protocol;

the data acquisition module is used for acquiring the machine data of the machine in an operation state;

the processor is used for filtering and comparing the collected machine data, so that based on the collected machine data, the working progress and the machine state of the machine are calculated, and the monitoring information is continuously input into a data set of historical processing data along with the working process of the machine; and

The processor is also used for performing intelligent analysis based on a machine learning algorithm and the historical processing data or the combination of the historical processing data and the monitoring information stored in the memory to provide predictive information, analytical information and decision information; wherein the method comprises the steps of

The predictive information is a predicted result of the processor for the working progress and the machine state of the machine, the analytical information is one or more important factors causing an abnormal phenomenon of the machine or affecting an operation schedule of the machine, and the decision information is one of a recommended schedule, a capacity and a traffic period or a combination thereof.

2. The intelligent digital controller upper box according to claim 1, wherein when the processor determines that the monitored monitoring information of the machine includes the abnormal phenomenon, the processor is also configured to calculate the analytical information including at least one of an abnormality cause type or a true cause classification tag based on a historical overhaul data stored in the memory and associated with a plurality of the machines, the historical overhaul data including one or a combination of a overhaul exclusion knowledge, a failure mode knowledge, and a failure phenomenon experience knowledge.

3. The intelligent digital controller upper box of claim 1, further comprising an input module electrically connected to the processor, the processor also providing a review request from the input module or an external device, such that the processor can perform the intelligent analysis based on the historical process data before the data acquisition module acquires the machine data of the machine in the operating state.

4. The intelligent digital controller upper box of claim 1, wherein the communication module is configured to establish a communication connection with a server using a second communication protocol to integrate the machine data, the monitoring information after the processor integration, the predictive information, the analytical information, and the decision information, and the server is configured to compare and analyze the machine data, the historical process data, and a historical overhaul data from an external data source, such that the optimized machine learning algorithm is deployed to one or more intelligent digital controllers upper boxes via the second communication protocol.

5. An implementation method based on an intelligent digital control machine upper box, adopting the intelligent digital control machine upper box as claimed in any one of claims 1-4 for monitoring and collecting machine data of at least one machine, characterized by comprising the following steps:

a processor executing an intelligent analysis based on a plurality of historical processing data stored in a memory, calculating predictive information associated with one or a combination of a work progress and a machine status of the machine;

the processor performing the intelligent analysis based on the historical process data to calculate an analytical information associated with an anomaly of the tool or affecting a scheduling of operations of the tool, the analytical information including one or more importance factors; and

the processor performing the intelligent analysis based on the predictive information and the analytical information to calculate decision information that is one or a combination of a recommended schedule, a capacity, and a date of delivery; wherein the method comprises the steps of

The processor can drive a data acquisition module to acquire a machine data from the machine in an operation state through a first communication protocol, and the machine data can be input into a data set of the historical processing data after being compared and collected into monitoring information.

6. The method of claim 5, wherein the processor continuously updates the predictive information, the analytical information, and the decision information while the machine is in the operating state based on a combination of the historical process data and the monitoring information while the processor is performing the intelligent analysis.

7. The method of claim 5, wherein when the processor determines that the monitored monitoring information of the machine includes the anomaly, the processor calculates the analytical information including at least one of an anomaly type or a true cause classification tag based on historical overhaul data stored in the memory and associated with a plurality of the machines, wherein the historical overhaul data includes one or a combination of overhaul exclusion knowledge, failure mode knowledge and failure experience knowledge.

8. The method of claim 5, wherein the processor establishes communication with a server using a second communication protocol at regular or irregular intervals when performing the intelligent analysis to compare and analyze the machine data, the monitoring information, the predictive information, the analytical information, and the decision information with one or more intelligent digital control boxes based on the machine data from an external data source and the historical process data by the server, such that an optimized machine learning algorithm can be deployed to one or more intelligent digital control boxes via the second communication protocol.

9. The method of claim 5, wherein the processor responds to a review request to cause the data acquisition module to perform the intelligent analysis based on the historical process data to calculate the predictive information, the analytical information and the decision information before acquiring the machine data of the machine in the operating state.