Numerical control machine tool operation state monitoring system based on big data

Abstract

The invention discloses a numerical control machine tool operation state monitoring system based on big data, and belongs to the technical field of equipment operation monitoring. Presetting a working temperature interval [m, n], and determining whether the machine tool equipment is to-be-determined equipment or fault equipment according to the preset working interval; selecting any to-be-determined equipment to obtain a temperature sequence of the to-be-determined equipment in a past time period t ', marking the temperature sequence as a to-be-determined temperature sequence, extracting a reference temperature sequence from historical temperature sequences of the to-be-determined equipment by taking the past time period t' as an interval, and calculating fault scores of all the to-be-determined equipment; if the fault score of any to-be-determined device is smaller than a preset threshold value, judging the to-be-determined device as a normal device; if the fault score of any to-be-determined device is larger than or equal to the preset threshold value, the to-be-determined device is judged to be a fault device, alarm information is sent to all the fault devices, and accurate recognition of the operation fault of the machine tool device is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment operation monitoring, and in particular to a CNC machine tool operation status monitoring system based on big data. Background Art

[0002] CNC machine tools, short for digitally controlled machine tools, are automated machine tools that integrate a program control system. The system can logically process programs containing control codes or other symbolic instructions, convert these instructions into coded numbers, and input them into the CNC device through information carriers. The CNC device sends out control signals after calculation processing to accurately guide the movement of the machine tool, thereby automatically processing parts of specified shapes and sizes according to the requirements of the drawings.

[0003] The failure of CNC machine tools is not always caused by excessive temperature. The abnormal operation status may be caused by many factors, such as poor contact of components, electrical system problems or other mechanical failures, which may cause a sudden change in the operation status of the machine tool at a specific moment, thereby causing abnormal temperature changes. Even if these temperature changes may not exceed the normal range, the CNC machine tool has already failed. Therefore, relying solely on detecting whether the temperature is too high to monitor the failure of CNC machine tools often results in insufficient accuracy of fault detection and cannot fully reflect the actual operation status of the machine tool. Summary of the invention

[0004] The purpose of the present invention is to provide a CNC machine tool operation status monitoring system based on big data to solve the following technical problems:

[0005] Monitoring CNC machine tool faults by simply detecting whether the temperature is too high often results in insufficient accuracy in fault detection and is unable to fully reflect the actual operating conditions of the machine tool.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A CNC machine tool operation status monitoring system based on big data, comprising:

[0008] A data acquisition module, used to collect the current body temperature T of all machine tool equipment and generate a curve Y of the temperature changing with the running time;

[0009] A data analysis module is used to preset a working temperature interval [m, n], and compare the body temperature T of all machine tools with the working temperature interval [m, n] in turn. If the body temperature of any machine tool satisfies (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a pending device; if the body temperature of any machine tool does not satisfy (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a warning device;

[0010] The data optimization module is used to select any pending device and obtain the body temperature sequence of the pending device in the past time period t' and mark it as the pending temperature sequence; taking the past time period t' as an interval, extracting several historical temperature sequences of the pending device whose similarity X with the pending temperature sequence reaches a preset threshold, and screening out a reference temperature sequence from the several historical temperature sequences; calculating the fault score Q of the pending device according to the calculation formula Q=(Ta-Tb)*|Sa-Sb|*X, ​​where Ta is the maximum temperature in the pending temperature sequence, Tb is the maximum temperature in the reference temperature sequence, Sa is the variance in the pending temperature sequence, and Sb is the variance in the reference temperature sequence;

[0011] The result generation module is used to determine the pending device as a warning device if the fault score Q of any pending device is greater than or equal to a preset threshold; if the fault score Q of any pending device is less than the preset threshold, the pending device is determined as a normal device and an alarm message is issued to all warning devices.

[0012] As a further solution of the present invention: the data analysis module further includes reconfirming the pending equipment, and the specific process is as follows:

[0013] Select any pending device and obtain the tangent slope k of the point corresponding to the current temperature in the corresponding curve Y of the pending device, obtain the temperature change rate value v according to the slope k, preset the future observation time t and calculate the expected body temperature T' of the machine tool equipment according to the calculation formula T'=T+Vt, calculate the temperature difference between the body temperature T' and the working temperature n, if the difference is greater than the preset threshold, the pending device is judged as a faulty device, if the difference is less than or equal to the preset threshold, it is still a pending device.

[0014] As a further solution of the present invention: the process of obtaining the temperature change rate value V is:

[0015] Establish a rectangular coordinate system, with temperature and time as the ordinate and abscissa respectively, obtain the coordinates of the current temperature (x1, y1), and obtain a point on the curve (x2, y2), x1≠x2, y1≠y 2， Then the slope of the straight line passing through the points (x1, y1) and (x2, y2) is k1 = (y2-y1) / (x2-x1) = Δy / Δx. When Δx→0, k1 infinitely approaches the slope k of the curve at the current temperature. Obtaining the value of k1 at this time, the temperature change rate value v = k1.

[0016] As a further solution of the present invention: in the data optimization module, the specific calculation process of the similarity X is:

[0017] The number of data points of the pending temperature sequence is obtained and marked as n. A distance matrix between the pending temperature sequence and the temperature in the historical temperature sequence is constructed according to the data corresponding to the n points. The distance matrix is:

[0018]

[0019] in | B n -A n | is the difference distance value between the pending temperature sequence and the historical temperature sequence; the matrix shortest cumulative path distance Ddtw(B,A) between the pending temperature sequence A and the historical temperature sequence B is calculated in sequence, and the curve similarity function is constructed according to the distance matrix using the dynamic time warping method. The similarity x function formula is:

[0020]

[0021] X is the similarity, and the comparison similarity values ​​X1, X2, ..., Xm of all historical temperature sequences are calculated in sequence, and the historical temperature sequences corresponding to the curve comparison similarity values ​​greater than the preset threshold are screened out.

[0022] As a further solution of the present invention: In the data optimization module, the specific process of screening out the reference temperature sequence is:

[0023] The temperature mean differences between the pending temperature sequence and each historical temperature sequence are calculated in turn and the calculated temperature mean differences are normalized to obtain the difference degree C. The reference score W of each historical temperature sequence is calculated according to the calculation formula Wi=Xi*(1-Ci), and the historical temperature sequence corresponding to the maximum reference score is selected and calibrated as the reference temperature sequence, where i is any historical temperature sequence, Xi is the similarity between the historical temperature sequence and the pending temperature sequence, and Ci is the difference between the historical temperature sequence and the pending temperature sequence.

[0024] As a further solution of the present invention: in the data optimization module, the historical temperature sequence is the body temperature of the pending equipment without any fault.

[0025] As a further solution of the present invention: the result generation module also includes continuing to monitor the normal equipment, obtaining the body temperature T' of any normal equipment at the next time node, if T' is greater than T and satisfies (3m+n) / 4≤T'≤(m+3n) / 4, then the normal equipment is re-determined as a pending equipment; if T' is greater than T and does not satisfy (3m+n) / 4≤T'≤(m+3n) / 4, then the normal equipment is re-determined as a faulty equipment; if T' is less than or equal to T, it is still a normal equipment.

[0026] Beneficial effects of the present invention:

[0027] The present invention first collects the body temperature T of all current machine tools, determines the pending equipment according to the body temperature of the machine tools, and determines the pending equipment that exceeds the working temperature range during the preset observation time by obtaining the temperature change speed of the machine tools according to the change speed and marks it as a faulty equipment, so as to timely issue an alarm before the abnormal situation occurs in the machine tools, and avoid the machine tools being in an abnormal operation state for a long time, which may cause equipment damage. At the same time, due to the abnormal temperature change caused by poor contact of components, electrical system problems or other mechanical failures in the machine tools, even if these temperature changes may not exceed the normal range, the CNC machine tool has already failed. Therefore, the temperature sequence of any pending equipment in the past time period is obtained and compared with the historical temperature sequence, and the fault score is calculated. If there is any pending equipment whose fault score Q is greater than or equal to the preset threshold, the pending equipment is determined as a faulty equipment, indicating that the temperature of the current pending equipment is abnormal but does not exceed the normal range, indicating that the CNC machine tool may have some faults. Therefore, the pending equipment is determined as a warning equipment, and alarm information is issued to all faulty equipment, thereby realizing accurate identification of the operation fault of the CNC machine tool equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention will be further described below in conjunction with the accompanying drawings.

[0029] Figure 1 It is a flow chart of a CNC machine tool operation status monitoring system based on big data of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] See also Figure 1 As shown, the present invention is a CNC machine tool operation status monitoring system based on big data, comprising:

[0032] A data acquisition module, used to collect the current body temperature T of all machine tool equipment and generate a curve Y of the temperature changing with the running time;

[0033] A data analysis module is used to preset a working temperature interval [m, n], and compare the body temperature T of all machine tools with the working temperature interval [m, n] in turn. If the body temperature of any machine tool satisfies (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a pending device; if the body temperature of any machine tool does not satisfy (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a faulty device.

[0034] The data optimization module is used to select any pending device and obtain the body temperature sequence of the pending device in the past time period t' and mark it as the pending temperature sequence; taking the past time period t' as an interval, extracting several historical temperature sequences of the pending device whose similarity X with the pending temperature sequence reaches a preset threshold, and screening out a reference temperature sequence from the several historical temperature sequences; calculating the fault score Q of the pending device according to the calculation formula Q=(Ta-Tb)*|Sa-Sb|*X, ​​where Ta is the maximum temperature in the pending temperature sequence, Tb is the maximum temperature in the reference temperature sequence, Sa is the variance in the pending temperature sequence, and Sb is the variance in the reference temperature sequence;

[0035] The result generation module is used to determine that any pending device is a faulty device if its fault score Q is greater than or equal to a preset threshold; if any pending device has a fault score Q less than the preset threshold, the pending device is determined to be a normal device and an alarm message is issued to all faulty devices.

[0036] The present invention first collects the body temperature T of all current machine tools, determines the pending equipment according to the body temperature of the machine tools, and determines the pending equipment that exceeds the working temperature range during the preset observation time by obtaining the temperature change speed of the machine tools according to the change speed and marks it as a faulty equipment, so as to timely issue an alarm before the abnormal situation occurs in the machine tools, and avoid the machine tools being in an abnormal operation state for a long time, which may cause equipment damage. At the same time, due to the abnormal temperature change caused by poor contact of components, electrical system problems or other mechanical failures in the machine tools, even if these temperature changes may not exceed the normal range, the CNC machine tool has already failed. Therefore, the temperature sequence of any pending equipment in the past time period is obtained and compared with the historical temperature sequence, and the fault score is calculated. If there is any pending equipment whose fault score Q is greater than or equal to the preset threshold, the pending equipment is determined as a faulty equipment, indicating that the temperature of the current pending equipment is abnormal but does not exceed the normal range, indicating that the CNC machine tool has some faults. Therefore, the pending equipment is determined as a faulty equipment, and alarm information is issued to all faulty equipment, thereby realizing accurate identification of the operation fault of the CNC machine tool equipment.

[0037] In another preferred embodiment of the present invention, the data analysis module further includes reconfirming the pending device, and the specific process is as follows:

[0038] Select any pending device and obtain the tangent slope k of the point corresponding to the current temperature in the corresponding curve Y of the pending device, obtain the temperature change rate value v according to the slope k, preset the future observation time t and calculate the expected body temperature T' of the machine tool equipment according to the calculation formula T'=T+Vt, calculate the temperature difference between the body temperature T' and the working temperature n, if the difference is greater than the preset threshold, the pending device is judged as a faulty device, if the difference is less than or equal to the preset threshold, it is still a pending device.

[0039] In another preferred embodiment of the present invention, the process of obtaining the temperature change rate value V is as follows:

[0040] Establish a rectangular coordinate system, with temperature and time as the ordinate and abscissa respectively, obtain the coordinates of the current temperature (x1, y1), and obtain a point on the curve (x2, y2), x1≠x2, y1≠y 2， Then the slope of the straight line passing through the points (x1, y1) and (x2, y2) is k1 = (y2-y1) / (x2-x1) = Δy / Δx. When Δx→0, k1 infinitely approaches the slope k of the curve at the current temperature. Obtaining the value of k1 at this time, the temperature change rate value v = k1.

[0041] In another preferred embodiment of the present invention, in the data optimization module, the specific calculation process of the similarity X is:

[0042] The number of data points of the pending temperature sequence is obtained and marked as n. A distance matrix between the pending temperature sequence and the temperature in the historical temperature sequence is constructed according to the data corresponding to the n points. The distance matrix is:

[0043]

[0044] in | B n -A n | is the difference distance value between the pending temperature sequence and the historical temperature sequence; the matrix shortest cumulative path distance Ddtw(B,A) between the pending temperature sequence A and the historical temperature sequence B is calculated in sequence, and the curve similarity function is constructed according to the distance matrix using the dynamic time warping method. The similarity x function formula is:

[0045]

[0046] X is the similarity, and the comparison similarity values ​​X1, X2, ..., Xm of all historical temperature sequences are calculated in sequence, and the historical temperature sequences corresponding to the curve comparison similarity values ​​greater than the preset threshold are screened out.

[0047] In another preferred embodiment of the present invention, in the data optimization module, the specific process of screening out the reference temperature sequence is:

[0048] The temperature mean differences between the pending temperature sequence and each historical temperature sequence are calculated in turn and the calculated temperature mean differences are normalized to obtain the difference degree C. The reference score W of each historical temperature sequence is calculated according to the calculation formula Wi=Xi*(1-Ci), and the historical temperature sequence corresponding to the maximum reference score is selected and calibrated as the reference temperature sequence, where i is any historical temperature sequence, Xi is the similarity between the historical temperature sequence and the pending temperature sequence, and Ci is the difference between the historical temperature sequence and the pending temperature sequence.

[0049] In another preferred embodiment of the present invention, in the data optimization module, the historical temperature sequence is the body temperature of the pending equipment without any fault.

[0050] In another preferred embodiment of the present invention, the result generation module also includes continuing to monitor the normal equipment, obtaining the body temperature T' of any normal equipment at the next time node, if T' is greater than T and satisfies (3m+n) / 4≤T'≤(m+3n) / 4, then the normal equipment is re-determined as a pending equipment; if T' is greater than T and does not satisfy (3m+n) / 4≤T'≤(m+3n) / 4, then the normal equipment is re-determined as a faulty equipment; if T' is less than or equal to T, then it is still a normal equipment.

[0051] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A CNC machine tool operation status monitoring system based on big data, characterized in that: include: A data acquisition module, used to collect the current body temperature T of all machine tool equipment and generate a curve Y of the temperature changing with the running time; A data analysis module is used to preset a working temperature interval [m, n], and compare the body temperature T of all machine tools with the working temperature interval [m, n] in turn. If the body temperature of any machine tool satisfies (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a pending device; if the body temperature of any machine tool does not satisfy (3m+n) / 4≤T≤(m+3n) / 4, the machine tool is determined to be a faulty device. The data optimization module is used to select any pending device and obtain the body temperature sequence of the pending device in the past time period t' with the current time as the starting point and mark it as the pending temperature sequence; extract several historical temperature sequences of the pending device whose similarity X with the pending temperature sequence reaches a preset threshold from the historical temperature sequence of the pending device with the past time period t' as the interval, and select the reference temperature sequence from the several historical temperature sequences; The fault score Q of the pending device is calculated according to the calculation formula Q=(Ta-Tb)*|Sa-Sb|*X, ​​wherein Ta is the maximum temperature in the pending temperature sequence, Tb is the maximum temperature in the reference temperature sequence, Sa is the variance in the pending temperature sequence, and Sb is the variance in the reference temperature sequence; the historical temperature sequence is the historical body temperature sequence of the pending device when it is operating without fault; The result generation module is used to determine that any pending device is a faulty device if the fault score Q of any pending device is greater than or equal to a preset threshold; if the fault score Q of any pending device is less than the preset threshold, the pending device is determined as a normal device.

2. According to the big data-based CNC machine tool operation status monitoring system of claim 1, it is characterized in that: The data analysis module also includes reconfirming the pending equipment, and the specific process is as follows: Select any pending device and obtain the tangent slope k of the point corresponding to the current temperature in the corresponding curve Y of the pending device, obtain the temperature change rate value v according to the slope k, preset the future observation time t and calculate the expected body temperature T' of the machine tool equipment according to the calculation formula T'=T+Vt, calculate the temperature difference between the body temperature T' and the working temperature n, if the difference is greater than the preset threshold, the pending device is judged as a faulty device, if the difference is less than or equal to the preset threshold, it is still a pending device.

3. The CNC machine tool operation status monitoring system based on big data according to claim 2 is characterized in that: The process of obtaining the temperature change rate value V is as follows: Establish a rectangular coordinate system, with temperature and time as the ordinate and abscissa respectively, obtain the coordinates of the current temperature (x1, y1), and obtain a point on the curve (x2, y2), x1≠x2, y1≠y 2， Then the slope of the straight line passing through the points (x1, y1) and (x2, y2) is k1 = (y2-y1) / (x2-x1) = Δy / Δx. When Δx→0, k1 infinitely approaches the slope k of the curve at the current temperature. Obtaining the value of k1 at this time, the temperature change rate value v = k1.

4. The CNC machine tool operation status monitoring system based on big data according to claim 1 is characterized in that: In the data optimization module, the specific calculation process of the similarity X is: The number of data points of the pending temperature sequence is obtained and marked as n. A distance matrix between the pending temperature sequence and the temperature in the historical temperature sequence is constructed according to the data corresponding to the n points. The distance matrix is: in | B n -A n | is the difference distance value between the pending temperature sequence and the historical temperature sequence; the matrix shortest cumulative path distance Ddtw(B,A) between the pending temperature sequence A and the historical temperature sequence B is calculated in sequence, and the curve similarity function is constructed according to the distance matrix using the dynamic time warping method. The similarity x function formula is: X is the similarity. All historical temperature sequence comparison similarity values ​​X1, X2, ..., Xm are calculated in sequence, and the historical temperature sequences corresponding to the similarities greater than the preset threshold are screened out.

5. The CNC machine tool operation status monitoring system based on big data according to claim 1 is characterized in that: In the data optimization module, the specific process of screening out the reference temperature sequence is as follows: The temperature mean differences between the pending temperature sequence and each historical temperature sequence are calculated in turn and the calculated temperature mean differences are normalized to obtain the difference degree C. The reference score W of each historical temperature sequence is calculated according to the calculation formula Wi=Xi*(1-Ci), and the historical temperature sequence corresponding to the maximum reference score is selected and calibrated as the reference temperature sequence, where i is any historical temperature sequence, Xi is the similarity between the historical temperature sequence and the pending temperature sequence, and Ci is the difference between the historical temperature sequence and the pending temperature sequence.

6. The CNC machine tool operation status monitoring system based on big data according to claim 1 is characterized in that: The result generation module also includes sending early warning information to all faulty equipment and notifying relevant personnel to carry out maintenance.

7. The CNC machine tool operation status monitoring system based on big data according to claim 1 is characterized in that: The result generation module also includes continuing to monitor the normal equipment to obtain the body temperature T' of any normal equipment at the next time node. If T' is greater than T and satisfies (3m+n) / 4≤T'≤(m+3n) / 4, the normal equipment is re-determined as a pending equipment; if T' is greater than T and does not satisfy (3m+n) / 4≤T'≤(m+3n) / 4, the normal equipment is re-determined as a faulty equipment; if T' is less than or equal to T, it is still a normal equipment.

Images