Multi-point measurement system and method thereof

Abstract

The present invention provides a multi-point measurement system and method thereof. The multi-point measurement system includes sensors and a computing device respectively attached to the measured points of the measured equipment. The computing unit of the computing device inputs the processed sensing data obtained by data preprocessing the initial sensing data generated by the sensor into the learning module of the computing unit for data analysis, and obtains the reference discriminations corresponding to the sensors respectively. At least two adjacent sensors form a combination, and the computing unit sequentially inputs the processed sensing data corresponding to the sensors in the combination into the learning module for merging operation, and obtains the merged discriminations corresponding to the combinations respectively. The computing unit performs judgment operation on the merged discrimination of each combination and the reference discrimination corresponding to the sensor in the combination, so as to generate a suitability judgment of the measured point.

Description

Technical Field

[0001] The present invention relates to a multi-point measurement system and method thereof. Specifically, the present invention relates to a multi-point measurement system and method thereof for mechanical equipment. Background Art

[0002] In the manufacturing industry, how to monitor production machines and handle failures immediately to avoid affecting the normal operation of the production line and affecting product yield and factory capacity has always been a topic of concern in the manufacturing industry.

[0003] A sensor is a device used to detect events or changes in the environment (e.g., position, length, height difference, displacement, appearance, etc.) and transmit this information to other electronic devices. Depending on the type of sensor, the data collected by the sensor includes physical quantities such as temperature, light, color, air pressure, magnetism, speed, acceleration, etc.

[0004] If sensors are used to monitor production machines, multiple sensors need to be deployed at multiple locations on the production machines to monitor various physical quantities of the production machines. Therefore, the placement of sensors is crucial. How to find the most appropriate placement location in a more efficient way and place the sensors at the best measurement points of the measured equipment is an important issue that needs to be solved in the method of using sensors to monitor production machines. Summary of the invention

[0005] An object of the present invention is to provide a multi-point measurement system that can find the most suitable measurement point of the device to be measured in a more efficient manner through the characteristics of the sticky sensor.

[0006] Another object of the present invention is to provide a multi-point measurement method, which can find the most suitable measurement point of the measured device through data analysis.

[0007] One embodiment of the present invention relates to a multi-point measurement system. The multi-point measurement system is used to measure a device to be measured, and includes a plurality of sensors and a computing device. The sensors are respectively attached to a plurality of points to be measured of the device to be measured. The computing device includes a computing unit and a storage unit; the computing unit includes a learning module, and the computing devices are respectively connected to the sensors in a communication manner. The sensor generates initial sensing data and transmits it to the storage unit for storage. The computing unit inputs the processed sensing data obtained by data preprocessing the initial sensing data into the learning module for data analysis, and obtains the reference discriminations corresponding to the sensors respectively. At least two adjacent sensors form a combination, and the computing unit sequentially inputs the processed sensing data corresponding to the sensors in the combination into the learning module for merging operations, and obtains the merged discriminations corresponding to the combinations respectively. The computing unit performs a judgment operation using the merged discrimination corresponding to each combination and the reference discrimination corresponding to the sensors in the combination, respectively, to generate a suitability judgment of the points to be measured.

[0008] In some embodiments, the plurality of points to be tested include a first point to be tested and an adjacent second point to be tested; the combination is a first combination, the first combination includes a first sensor and a second sensor, the first sensor is located at the first point to be tested, and the second sensor is located at the second point to be tested; the plurality of reference discriminations corresponding to the first combination include a first reference discrimination and a second reference discrimination, the first reference discrimination corresponds to the first sensor, and the second reference discrimination corresponds to the second sensor; the combined discrimination corresponding to the first combination The first combined discrimination degree is a first combined discrimination degree, and the first combined discrimination degree corresponds to the first sensor and the second sensor; and the determination operation is a first determination operation, and the first determination operation comprises: the calculation unit compares the first reference discrimination degree and the second reference discrimination degree with the first combined discrimination degree respectively; and if the first reference discrimination degree and the second reference discrimination degree are both greater than the first combined discrimination degree, the calculation unit determines that the to-be-tested point corresponding to one of the first sensor and the second sensor is unsuitable, and performs a second determination operation.

[0009] In some embodiments, the plurality of test points further include a third test point adjacent to the second test point; the combination is a second combination, the second combination includes the second sensor and a third sensor, the third sensor is located at the third test point; the plurality of reference discriminations corresponding to the second combination include the second reference discrimination and a third reference discrimination, the third reference discrimination corresponding to the third sensor; the combined discrimination corresponding to the second combination is a second combined discrimination, the second combined discrimination corresponding to the second sensor and the third sensor; and the determination operation is the second determination operation; the second determination operation includes: the calculation unit compares the second reference discrimination and the third reference discrimination with the second combined discrimination respectively; and if the second reference discrimination and the third reference discrimination are both greater than the second combined discrimination, the calculation unit determines that the test point corresponding to the second sensor is unsuitable.

[0010] In some embodiments, each of the plurality of initial sensing data includes a plurality of initial time-series sensing data; and the plurality of initial time-series sensing data respectively correspond to a plurality of time windows.

[0011] In some embodiments, the device under measurement has an operation cycle; and the time unit range of the multiple time windows is 1 to 10 times of the operation cycle.

[0012] In some embodiments, the data pre-processing is a label setting; and the plurality of processed sensing data are a plurality of labeled sensing data.

[0013] In some embodiments, the label setting includes setting multiple labels; and the multiple labels include a loose screw, a power shortage, a maximum speed, a first speed, equipment abnormality, yield, other identifiable variation factors, or a combination thereof.

[0014] In some embodiments, the plurality of tags correspond to a plurality of setting scenarios respectively; and the calculation unit performs the determination operation once according to each setting scenario.

[0015] In some embodiments, the plurality of sensors are temperature sensors, humidity sensors, vibration sensors, or a combination thereof; and the plurality of initial sensing data are temperature data, humidity data, vibration data, or a combination thereof.

[0016] In some embodiments, the calculation unit performs the determination operation once according to each initial sensing data.

[0017] Another embodiment of the present invention relates to a multi-point measurement method. The multi-point measurement method includes a multi-point measurement method for measuring a device to be measured, comprising the following steps: using a plurality of sensors respectively attached to a plurality of points to be measured of the device to be measured to generate a plurality of initial sensing data and transmit them to a storage unit of a computing device for storage, the computing device being respectively connected to the sensors in communication; using a computing unit of the computing device to pre-process the initial sensing data to obtain a plurality of processed sensing data, inputting the processing sensing data into a learning module of the computing unit for data analysis, and obtaining a plurality of reference discriminations respectively corresponding to the sensors; at least two adjacent sensors form a combination, the computing unit sequentially inputting the processing sensing data corresponding to the sensors in the combination into the learning module for merging operation, and obtaining the combined discriminations respectively corresponding to the combination; and using the computing unit to perform a determination operation respectively using the combined discrimination corresponding to each combination and the reference discrimination corresponding to the sensor in the combination, so as to generate a suitability determination of the point to be measured.

[0018] In some embodiments, the determination operation is a first determination operation, the combination is a first combination, the first combination includes a first sensor and a second sensor, the plurality of reference discriminations corresponding to the first combination include a first reference discrimination and a second reference discrimination, the combined discrimination corresponding to the first combination is a first combined discrimination, and the first determination operation includes: comparing the first reference discrimination and the second reference discrimination with the first combined discrimination respectively by the computing unit; and if the first reference discrimination and the second reference discrimination are both greater than At the first combined resolution, the calculation unit determines that the test point corresponding to one of the first sensor and the second sensor is unsuitable, and performs a second determination operation; wherein: the plurality of test points include a first test point and an adjacent second test point; the first sensor is located at the first test point, and the second sensor is located at the second test point; the first reference resolution corresponds to the first sensor, the second reference resolution corresponds to the second sensor; and the first combined resolution corresponds to the first sensor and the second sensor.

[0019] In some embodiments, the determination operation is the second determination operation, the combination is a second combination, the second combination includes the second sensor and a third sensor, the multiple reference discriminations corresponding to the second combination include the second reference discrimination and a third reference discrimination, and the combined discrimination corresponding to the second combination is a second combined discrimination; the second determination operation includes: using the calculation unit to compare the second reference discrimination and the third reference discrimination with the second combined discrimination respectively; and if the second reference discrimination and the third reference discrimination are both greater than the second combined discrimination, using the calculation unit to determine that the test point corresponding to the second sensor is unsuitable; wherein: the multiple test points further include a third test point adjacent to the second test point; the third sensor is located at the third test point; the third reference discrimination corresponds to the third sensor; and the second combined discrimination corresponds to the second sensor and the third sensor.

[0020] In some embodiments, each of the plurality of initial sensing data includes a plurality of initial time-series sensing data; and the plurality of initial time-series sensing data respectively correspond to a plurality of time windows.

[0021] In some embodiments, the device under measurement has an operation cycle; and the time unit range of the multiple time windows is 1 to 10 times of the operation cycle.

[0022] In some embodiments, the data pre-processing is a label setting; and the plurality of processed sensing data are a plurality of labeled sensing data.

[0023] In some embodiments, the label setting includes setting multiple labels; and the multiple labels include a loose screw, a power shortage, a maximum speed, a first speed, equipment abnormality, yield, other identifiable variation factors, or a combination thereof.

[0024] In some embodiments, the calculation unit performs the determination operation once according to each of the plurality of setting contexts respectively corresponding to the plurality of tags.

[0025] In some embodiments, the plurality of sensors are temperature sensors, humidity sensors, vibration sensors, or a combination thereof; and the plurality of initial sensing data are temperature data, humidity data, vibration data, or a combination thereof.

[0026] In some embodiments, the calculation unit performs the determination operation once according to each initial sensing data. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of a multi-point measurement system according to an embodiment of the present invention;

[0028] Figure 2 A schematic diagram of a multi-point measurement scenario according to an embodiment of the present invention;

[0029] Figure 3 A schematic diagram of initial sensing data according to an embodiment of the present invention;

[0030] Figure 4 A schematic diagram of a time window according to an embodiment of the present invention;

[0031] Figure 5 A schematic diagram of the reference discrimination according to an embodiment of the present invention;

[0032] Figure 6 It is a combined schematic diagram of an embodiment of the present invention;

[0033] Figure 7 A schematic diagram of combined discrimination according to an embodiment of the present invention;

[0034] Figure 8 A schematic diagram of labeled sensing data according to an embodiment of the present invention;

[0035] Fig. 9 A schematic diagram of a multi-point measurement system according to another embodiment of the present invention;

[0036] Fig.10 FIG. 4 is a schematic diagram of a multi-point measurement method according to another embodiment of the present invention.

[0037] Description of main component symbols:

[0038] 10 Multi-point measurement system

[0039] 100 Sensors

[0040] 110 combinations

[0041] 111 First Group

[0042] 112 Second Group

[0043] 120 First Sensor

[0044] 130 Second sensor

[0045] 140 Third Sensor

[0046] 200 Computing Devices

[0047] 210 computing units

[0048] 215 learning modules

[0049] 220 storage units

[0050] 230 Communication module

[0051] 300 Intermediary Device

[0052] 20 Device under measurement

[0053] T Temperature data

[0054] H Humidity data

[0055] V Vibration Data

[0056] Tag

[0057] D Operation cycle

[0058] t time

[0059] to start time

[0060] t1 first time

[0061] t2 Second time

[0062] t3 Third time

[0063] t4 fourth time

[0064] t5 fifth time

[0065] tn Last time

[0066] W Time Window

[0067] W1 First Time Window

[0068] W2 Second Time Window

[0069] W3 Third Time Window

[0070] Wn-2 Last time window

[0071] B Benchmark discrimination

[0072] B1 First benchmark discrimination

[0073] B2 Second standard discrimination

[0074] B3 Third standard discrimination

[0075] C Merged Discrimination

[0076] C1 First Merge Discrimination

[0077] C2 Second Merged Discrimination

[0078] D Operation cycle

[0079] P Point to be measured

[0080] P1 The first point to be measured

[0081] P2 The second point to be measured

[0082] P3 The third point to be tested

[0083] P12 The first and second test points

[0084] P23 The second and third test points

[0085] N Network

[0086] S1~S5 Step 1~Step 5 DETAILED DESCRIPTION

[0087] Figure 1 A schematic diagram of a multi-point measurement system according to an embodiment of the present invention; Figure 2 Schematic diagram of a multi-point measurement scenario according to an embodiment of the present invention. The multi-point measurement system 10 is used to measure a device 20 to be measured, and includes a plurality of sensors 100 and a computing device 200; the plurality of sensors 100 (e.g., a first sensor 120, a second sensor 130, a third sensor 140, etc.) are respectively attached to a plurality of points P to be measured (e.g., a first point P1 to be measured, a second point P2 to be measured, a third point P3 to be measured, etc.) of the device 20 to be measured. Figure 1 and Figure 2 In the embodiment, the sensor 100 at least includes a first sensor 120, a second sensor 130, and a third sensor 140, and the point P to be measured includes at least a first point P1 to be measured, a second point P2 to be measured, and a third point P3 to be measured, and the first sensor 120 is attached to the first point P1 to be measured, the second sensor 130 is attached to the second point P2 to be measured, and the third sensor 140 is attached to the third point P3 to be measured, but it is not limited thereto; that is, in other embodiments, the multi-point measurement system 10 may include three sensors 100, which are respectively attached to different points P to be measured.

[0088] In a preferred embodiment, the sensor 100 is a soft and flexible thin wireless sensor that can be attached to the surface or inner side of the measured device 20 at will to detect changes in temperature, humidity, vibration, etc. generated by the measured device 20 during the production process, and then the collected data is instantly transmitted to the storage unit 220 for storage for subsequent data analysis.

[0089] In a preferred embodiment, the device 20 to be measured is a mechanical device in a production line. Through the multi-point measurement system and multi-point measurement method of the present invention, the best measurement position of the surface or inner side of the device 20 to be measured can be found, and multiple sensors 100 are further attached to the multiple best measurement positions on the surface or inner side of the device 20 to be measured. Then, the sensors 100 collect various data generated by the device 20 to be measured during the production process for subsequent data analysis.

[0090] like Figure 1 As shown, the computing device 200 includes a computing unit 210, a storage unit 220, and a communication module 230; the computing unit 210 includes a learning module 215, and the computing device 200 is respectively connected to the sensor 100 through the communication module 230. The sensor 100 generates initial sensing data and transmits it to the storage unit 220 for storage.

[0091] In a preferred embodiment, the computing device 200 is a mobile device, such as a smart phone. In another embodiment, the computing device 200 is a host of a local area network. In a preferred embodiment, the computing unit 210 is a processor of the computing device 200, and the storage unit 220 is an internal storage or an external storage of the computing device 200.

[0092] The sensor 100 is a temperature sensor, a humidity sensor, a vibration sensor, or a combination thereof; the initial sensing data is temperature data, humidity data, vibration data, or a combination thereof. In a preferred embodiment, the sensor 100 is a sensor with a composite function, such as a temperature and humidity composite sensor, a temperature, humidity and vibration composite sensor, or other types of composite sensors. In another embodiment, the sensor 100 is a single-function sensor, such as a temperature sensor, a humidity sensor, a vibration sensor, or other sensors that only include a single function; the initial sensing data is a single type of initial sensing data, such as temperature data, humidity data, vibration data, or other sensing data obtained by corresponding functions.

[0093] Figure 3 FIG. 4 is a schematic diagram of initial sensing data according to an embodiment of the present invention. Figure 3 The sensor 100 of the illustrated embodiment may be a composite sensor composed of a temperature sensor, a humidity sensor, and a vibration sensor. Therefore, the initial sensing data generated by the sensor 100 is composite initial sensing data composed of temperature data, humidity data, and vibration data. That is, the initial sensing data includes three types of sensing data: temperature data, humidity data, and vibration data.

[0094] by Figure 3 In the embodiment, for example, the first sensor 110 attached to the first test point P1 generates initial sensing data after measurement at the first test point P1 as "P1: T (25, 28, 27 ... 24); H (70, 65, 63 ... 68); V (80, 78, 81 ... 75)", wherein the temperature data T (unit: ° C) is "25, 28, 27 ... 24", the humidity data H (unit: %) is "70, 65, 63 ... 68", the vibration data V (unit: m / s 2) is "80, 78, 81...75"; the second sensor 120 attached to the second test point P2, the initial sensing data generated by the measurement at the second test point P2 is "P2: T (26, 25, 27...23); H (71, 64, 65...67); V (79, 76, 78...76)", wherein the temperature data T (unit: ° C) is "26, 25, 27...23", the humidity data H (unit: %) is "71, 64, 65...67", the vibration data V (unit: m / s 2 ) is "79, 76, 78...76"; the third sensor 130 attached to the third test point P3, the initial sensing data generated by the measurement at the third test point P3 is "P3:T(25, 24, 28...26); H(69, 67, 68...68); V(78, 75, 76...74)", wherein the temperature data T (unit: ° C) is "25, 24, 28...26", the humidity data H (unit: %) is "69, 67, 68...68", the vibration data V (unit: m / s 2 ) is "78, 75, 76…74".

[0095] Specifically, each initial sensing data includes a plurality of initial time series sensing data. Figure 3 Taking the embodiment as an example, the initial sensing data generated by measurement at the first test point P1 is "P1: T (25, 28, 27 ... 24); H (70, 65, 63 ... 68); V (80, 78, 81 ... 75)", wherein the initial time series sensing data of the temperature data T is "25, 28, 27 ... 24", that is, the value "25" is the first temperature data obtained at the first time point, the value "28" is the second temperature data obtained at the second time point, the value "27" is the third temperature data obtained at the third time point, ... (and so on), ... the value "24" is the last temperature data obtained at the last time point.

[0096] The measuring device 20 has an operation cycle D; specifically, the operation cycle D is the cycle of one operation of the measuring device 20. The initial time series sensing data correspond to a plurality of time windows W; in a preferred embodiment, the time unit range of the time window W is 1 to 10 times of the operation cycle D. For details about the time window W, see Figure 4 and the description of its corresponding paragraph.

[0097] Figure 4 FIG. 1 is a schematic diagram of a time window W according to an embodiment of the present invention. The time window W includes a first time window W1, a second time window W2, a third time window W3, ... (and so on), and a final time window Wn-2. The time t includes a start time t0, a first time t 1 , the second time t 2 , the third time t 3, the fourth time t 4 , the fifth time t 5 ...Last time t n .At Figure 4 In the embodiment shown, the operation cycle D is 3 seconds as an example; that is, it takes 3 seconds for the measuring device 20 to complete a complete operation. Figure 4 The time unit range of the time window W in the embodiment shown can be 3 seconds to 30 seconds. Figure 4 The embodiment shown takes 3 seconds as an example. Figure 4 In the embodiment shown, the start time t0 is the 0th second, the first time t1 is the 1st second, and the second time t 2 is the second second, the last time t n is the nth second. Figure 4 As shown, the first time window W1 includes the start time t 0 ~The third time t 3 The second time window W2 includes the first time t 1 ~ The fourth time t 4 , the third time window W3 includes the second time t 2 ~Fifth time t 5 ...etc.; that is, Figure 3 The first sensor 120 located at the first test point P1 of the embodiment generates the first to fourth temperature data at the 0th to 3rd seconds, generates the fifth to eighth temperature data at the 1st to 4th seconds, generates the ninth to twelfth temperature data at the 2nd to 5th seconds, ... (and so on).

[0098] Figure 5 FIG. 4 is a schematic diagram of reference discrimination according to an embodiment of the present invention. Figure 1 The calculation unit 210 will Figure 3 The processed sensing data obtained by data preprocessing of the initial sensing data is input into the learning module 215 for data analysis, and the Figure 5 Corresponding to different points to be measured P( Figure 5 Take the first test point P1, the second test point P2, and the third test point P3 as an example) as the reference discrimination B of the sensor 100. Figure 5 Taking the embodiment as an example, the reference discrimination B includes a first reference discrimination B1, a second reference discrimination B2, and a third reference discrimination B3, which respectively correspond to the first sensor 120 located at the first test point P1, the second sensor 130 located at the second test point P2, and the third sensor 140 located at the third test point P3.

[0099] Taking the temperature data of the first sensor 120 located at the first test point P1 as an example, Figure 1 The calculation unit 210 will Figure 3The temperature initial sensing data "25, 28, 27 ... 24" of the first sensor 120 located at the first test point P1 is processed to obtain processed sensing data, and then the processed sensing data is input into Figure 1 The learning module 215 is used to analyze the data, and the reference discrimination B (i.e., the first reference discrimination B1) of the first sensor 120 located at the first test point P1 is obtained to be 92%. The reference discrimination B (i.e., the second reference discrimination B2) of the second sensor 130 located at the second test point P2 is obtained to be 89%, and the reference discrimination B (i.e., the third reference discrimination B3) of the third sensor 140 located at the third test point P3 is obtained to be 95%. For details about the processed sensing data, see Figure 8 and the description of its corresponding paragraph.

[0100] Furthermore, at least two adjacent sensors 100 form a group 110. Figure 1 The computing unit 210 sequentially inputs the processed sensing data corresponding to the sensors 100 in the combination 110. Figure 1 The learning module 215 performs a merging operation and obtains the merging discrimination C corresponding to the combination 110. For details about the combination 110, see Figure 6 and the description of its corresponding paragraph. Figure 1 The calculation unit 210 performs a determination operation using the combined discrimination C corresponding to each combination 110 and the reference discrimination B corresponding to the sensor 100 in the combination 110 to generate a suitability determination of the point P to be measured. Figure 7 and the description of its corresponding paragraph.

[0101] Figure 6 A schematic diagram of a combination 110 according to an embodiment of the present invention; Figure 7 FIG. 1 is a schematic diagram of the combined discrimination according to an embodiment of the present invention. It is worth noting that Figure 6 In the illustrated embodiment, each combination 110 includes two sensors 100; in other embodiments, each combination may also include more sensors 100, for example, three sensors 100. Figure 6 As shown, the test point P includes a first test point P1, a second test point P2 adjacent to the first test point P1, and a third test point P3 adjacent to the second test point P2. The first sensor 120, the second sensor 130, and the third sensor 140 are respectively located at the first test point P1, the second test point P2, and the third test point P3. The determination operation includes a first determination operation and a second determination operation. When the suitability determination result of the first determination operation is unsuitable, the second determination operation will be performed.

[0102] Specifically, in the first determination operation, the combination 110 is the first combination 111; the first combination 111 includes the first sensor 120 and the second sensor 130. The reference discrimination B corresponding to the first combination 111 includes the first reference discrimination B1 and the second reference discrimination B2; the first reference discrimination B1 corresponds to the first sensor 120, and the second reference discrimination B2 corresponds to the second sensor 130. The combined discrimination C corresponding to the first combination 111 is the first combined discrimination C1; the first combined discrimination C1 corresponds to the first sensor 120 and the second sensor 130.

[0103] When the judgment operation is the first judgment operation, Figure 1 The calculation unit 210 compares the first reference recognition degree B1 and the second reference recognition degree B2 with the first combined recognition degree C1 respectively. If the first reference recognition degree B1 and the second reference recognition degree B2 are both greater than the first combined recognition degree C1, then Figure 1 The calculation unit 210 determines that the test point corresponding to one of the first sensor 120 and the second sensor 130 is unsuitable, and performs a second determination operation. Figure 5 The baseline discrimination B and Figure 7 Taking the combined recognition degree C as an example, the first reference recognition degree B1 is 92% and the second reference recognition degree B2 is 89%, both of which are greater than the first combined recognition degree C1 of 56%. Therefore, the calculation unit 210 determines that one of the first test point P1 and the second test point P2 is inappropriate.

[0104] Specifically, in the second determination operation, the combination 110 is the second combination 112; the second combination 112 includes the second sensor 130 and the third sensor 140, such as Figure 6 The reference discrimination B corresponding to the second combination 112 includes a second reference discrimination B2 and a third reference discrimination B3; the third reference discrimination B3 corresponds to the third sensor 140. The combined discrimination C corresponding to the second combination 112 is a second combined discrimination C2, and the second combined discrimination C2 corresponds to the second sensor 130 and the third sensor 140.

[0105] When the determination operation is the second determination operation, Figure 1 The calculation unit 210 compares the second reference recognition degree B2 and the third reference recognition degree B3 with the second combined recognition degree C2 respectively. If the second reference recognition degree B2 and the third reference recognition degree B3 are both greater than the second combined recognition degree C2, then Figure 1 The calculation unit 210 determines that the point to be measured corresponding to the second sensor 130 is not suitable. Figure 5 The baseline discrimination B and Figure 7Taking the combined discrimination C as an example, the second reference discrimination B2 is 89% and the third reference discrimination B3 is 95%, both of which are greater than the second combined discrimination C2 of 70%, so the calculation unit 210 determines that the second test point P2 is inappropriate. Figure 8 FIG. 1 is a schematic diagram of labeled sensing data according to an embodiment of the present invention. In a preferred embodiment, data preprocessing is label setting, and the processed sensing data is labeled sensing data. Figure 3 The initial sensing data and Figure 8 Taking the labeled sensing data as an example, the first sensor 110 attached to the first test point P1, the initial sensing data generated by the measurement at the first test point P1 is "P1: T (25, 28, 27 ... 24); H (70, 65, 63 ... 68); V (80, 78, 81 ... 75)". After the data preprocessing set by the tag, the labeled sensing data generated by the measurement at the first test point P1 is "P1: T (25, 28, 27 ... 24, IP); H (70, 65, 63 ... 68, IP); V (80, 78, 81 ... 75, IP)", where "IP" is the tag Tag.

[0106] Specifically, the tag setting includes setting multiple tags, including loose screws, insufficient power, maximum speed, first gear speed, equipment abnormality, yield, other identifiable variation factors, or a combination thereof. The tags correspond to multiple setting scenarios; for example, the tag "LS" corresponds to the setting scenario of "loose screws", the tag "IP" corresponds to the setting scenario of "insufficient power"...

[0107] In a preferred embodiment, Figure 1 The calculation unit 210 performs a determination operation according to each set scenario; that is, a determination operation is performed with the set scenario of "loose screws" and a result of suitability determination is obtained, a determination operation is performed with the set scenario of "insufficient power" and a result of suitability determination is obtained, ... (and so on). Specifically, after the test is performed in each round of the set scenario (for example, "loose screws") and the initial sensing data is generated, a tag Tag (for example, "LS") is added to the initial sensing data to become the tagged sensing data.

[0108] In a preferred embodiment, Figure 1 The calculation unit 210 performs a determination operation based on different types of initial data; that is, performs a determination operation based on temperature data and obtains a suitability determination result, performs a determination operation based on vibration data and obtains a suitability determination result, performs a determination operation based on humidity data and obtains a suitability determination result, ... (and so on).

[0109] In detail, Figure 1The calculation unit 210 will perform a judgment operation according to different types of initial data in different scenarios; that is, a judgment operation is performed with the temperature data of the setting scenario of "loose screws", a judgment operation is performed with the humidity data of the setting scenario of "loose screws", a judgment operation is performed with the vibration data of the setting scenario of "loose screws", a judgment operation is performed with the temperature data of the setting scenario of "insufficient power", a judgment operation is performed with the humidity data of the setting scenario of "insufficient power", a judgment operation is performed with the vibration data of the setting scenario of "insufficient power", ... (and so on). The result of the suitability judgment of each type of initial data of each test point P in each setting scenario is obtained, and then according to different design requirements, a further customized judgment process is performed to obtain unsuitable test points P, remove unsuitable test points P, and replace them with new test points P, and repeat the above process until all the most suitable test points P are found.

[0110] Fig. 9 FIG. 4 is a schematic diagram of a multi-point measurement system according to another embodiment of the present invention. Fig. 9 and Figure 1 The difference is that Figure 1 The computing device 200 is a computing device located within the communication range of the sensor 100, such as a mobile device or a host of a local area network; Figure 1 The computing device 200 and the sensor 100 are connected to each other in a communication manner via short-range wireless communication; in a preferred embodiment, the computing device 200 and the sensor 100 are connected to each other in a communication manner via Bluetooth communication.

[0111] Different from Figure 1 , Fig. 9 The computing device 200' is a cloud host, so Fig. 9 The computing device 200' needs to communicate with the sensor 100 through the intermediary device 300; in detail, the intermediary device 300 is located within the communication range of the sensor 100, and communicates with the sensor 100 in a short-range wireless communication manner, and the intermediary device 300 is then connected to the computing device 200' through the Internet N. In a preferred embodiment, the intermediary device 300 communicates with the sensor 100 in a Bluetooth communication manner. In a preferred embodiment, the intermediary device 300 is a mobile device or a host of a local area network.

[0112] Fig.10 FIG. 1 is a schematic diagram of a multi-point measurement method according to another embodiment of the present invention. Fig.10As shown, the multi-point measurement method includes the following steps: (S1) using multiple sensors 100 (first sensor 120, second sensor 130, third sensor 140...) respectively attached to multiple points P to be measured of the measured device 20 to generate multiple initial sensing data and transmit them to the storage unit 220 of the computing device 200 for storage, and the computing device 200 is respectively connected to the sensor 100 for communication; (S2) the initial sensing data is subjected to data preprocessing to obtain multiple processed sensing data; (S3) the computing unit 210 of the computing device 200 inputs the processed sensing data into the learning module 215 of the computing unit 210 for data analysis, and obtains A plurality of reference discriminations respectively corresponding to the sensors 100 (the first sensor 120, the second sensor 130, the third sensor 140, ...); (S4) at least two adjacent groups of the plurality of sensors form a combination 110, and the computing unit 210 sequentially inputs the processed sensing data corresponding to the sensors 100 in the combination 110 into the learning module 215 for merging operation, and obtains a plurality of merged discriminations respectively corresponding to the combination 110; and (S5) the computing unit 210 performs a determination operation respectively with the merged discrimination corresponding to each combination 110 and the reference discrimination corresponding to the sensor 100 in the combination, so as to generate a suitability determination of the point P to be measured. Fig.10 For the corresponding Figures 1 to 9 method, as mentioned above Figures 1 to 9 The relevant paragraphs will not be repeated here.

[0113] The first determination operation includes the following steps: using the calculation unit 210 to compare the first reference recognition and the second reference recognition with the first combined recognition respectively; if the first reference recognition and the second reference recognition are both greater than the first combined recognition, the calculation unit 210 determines that the test point corresponding to one of the first sensor and the second sensor is unsuitable, and performs the second determination operation.

[0114] The second determination operation includes the following steps: using the calculation unit 210 to compare the second reference discrimination and the third reference discrimination with the second combined discrimination respectively; if the second reference discrimination and the third reference discrimination are both greater than the second combined discrimination, the calculation unit 210 determines that the test point corresponding to the second sensor is unsuitable.

[0115] The present invention has been described by the above-mentioned relevant embodiments, however, the above-mentioned embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, modifications and equivalent settings included in the spirit and scope of the claims are included in the scope of the present invention.

Claims

1. A multi-point measurement system for measuring a device to be measured, characterized in that: Include: A plurality of sensors are respectively attached to a plurality of points to be measured of the device to be measured; and a computing device, comprising a computing unit and a storage unit, wherein the computing unit comprises a learning module, and the computing device is communicatively connected to the plurality of sensors respectively; in The multiple sensors generate multiple initial sensing data and transmit them to the storage unit for storage; The computing unit inputs a plurality of processed sensing data obtained by a data preprocessing of the plurality of initial sensing data into the learning module to perform a data analysis and obtain a plurality of reference discriminations respectively corresponding to the plurality of sensors; At least two adjacent groups of the plurality of sensors form a combination, and the calculation unit sequentially inputs the plurality of processed sensing data corresponding to the plurality of sensors in the combination into the learning module to perform a merging operation, and obtains a plurality of merging discriminations respectively corresponding to the plurality of combinations; as well as The calculation unit performs a determination operation using the combined discrimination corresponding to each combination and the multiple reference discriminations corresponding to the multiple sensors in the combination to generate a suitability determination of the multiple points to be measured; The plurality of points to be tested include a first point to be tested and an adjacent second point to be tested; The combination is a first combination, and the first combination includes a first sensor and a second sensor, the first sensor is located at the first point to be measured, and the second sensor is located at the second point to be measured; The plurality of reference discriminations corresponding to the first combination include a first reference discrimination and a second reference discrimination, the first reference discrimination corresponds to the first sensor, and the second reference discrimination corresponds to the second sensor; The combined recognition corresponding to the first combination is a first combined recognition, and the first combined recognition corresponds to the first sensor and the second sensor; and The determination operation is a first determination operation, and the first determination operation includes: The calculation unit compares the first reference discrimination and the second reference discrimination with the first combined discrimination respectively; as well as If the first reference discrimination and the second reference discrimination are both greater than the first combined discrimination, the calculation unit determines that the test point corresponding to one of the first sensor and the second sensor is unsuitable, and performs a second determination operation.

2. The multi-point measurement system according to claim 1, characterized in that: The plurality of test points further include a third test point adjacent to the second test point; The combination is a second combination, and the second combination includes the second sensor and a third sensor, and the third sensor is located at the third point to be measured; The plurality of reference discriminations corresponding to the second combination include the second reference discrimination and a third reference discrimination, and the third reference discrimination corresponds to the third sensor; The combined resolution corresponding to the second combination is a second combined resolution, and the second combined resolution corresponds to the second sensor and the third sensor; and The determination operation is the second determination operation; the second determination operation includes: The calculation unit compares the second reference discrimination and the third reference discrimination with the second combined discrimination respectively; as well as If the second reference recognition degree and the third reference recognition degree are both greater than the second combined recognition degree, the calculation unit determines that the point to be measured corresponding to the second sensor is unsuitable.

3. The multi-point measurement system according to claim 1, characterized in that: Each of the plurality of initial sensing data comprises a plurality of initial time series sensing data; and The plurality of initial time series sensing data correspond to a plurality of time windows respectively.

4. The multi-point measurement system according to claim 3, characterized in that: The device under measurement has an operation cycle; and The time unit range of the multiple time windows is 1 to 10 times of the operation cycle.

5. The multi-point measurement system according to claim 1, characterized in that: The data is preprocessed into a label setting; and The plurality of processed sensing data are a plurality of labeled sensing data.

6. The multi-point measurement system according to claim 5, characterized in that: The tag setting includes setting a plurality of tags; and The multiple labels include a loose screw, a power shortage, a maximum speed, a first speed, equipment abnormality, yield, other identifiable variation factors, or a combination thereof.

7. The multi-point measurement system according to claim 6, characterized in that: The multiple tags correspond to multiple setting scenarios respectively; and The calculation unit performs the determination operation once according to each set scenario.

8. The multi-point measurement system according to claim 1, characterized in that: The plurality of sensors are temperature sensors, humidity sensors, vibration sensors, or a combination thereof; and The plurality of initial sensing data are temperature data, humidity data, vibration data, or a combination thereof.

9. The multi-point measurement system according to claim 8, characterized in that: The calculation unit performs the determination operation once according to each initial sensing data.

10. A multi-point measurement method for measuring a device to be measured, characterized in that: It includes the following steps: A plurality of sensors respectively attached to a plurality of points to be measured of the device to be measured generate a plurality of initial sensing data and transmit the data to a storage unit of a computing device for storage, the computing device being communicatively connected to the plurality of sensors respectively; A computing unit of the computing device uses a computing unit to pre-process the plurality of initial sensing data to obtain a plurality of processed sensing data, and inputs the processed sensing data into a learning module of the computing unit to perform a data analysis, and obtain a plurality of reference discriminations respectively corresponding to the plurality of sensors; At least two adjacent groups of the plurality of sensors form a combination, and the calculation unit sequentially inputs the plurality of processed sensing data corresponding to the plurality of sensors in the combination into the learning module to perform a merging operation, and obtains a plurality of merging discriminations respectively corresponding to the plurality of combinations; as well as The calculation unit performs a determination operation using the combined discrimination corresponding to each combination and the multiple reference discriminations corresponding to the multiple sensors in the combination to generate a suitability determination of the multiple points to be measured; The determination operation is a first determination operation, the combination is a first combination, the first combination includes a first sensor and a second sensor, the plurality of reference discriminations corresponding to the first combination include a first reference discrimination and a second reference discrimination, the combined discrimination corresponding to the first combination is a first combined discrimination, and the first determination operation includes: Using the calculation unit to compare the first reference recognition degree and the second reference recognition degree with the first combined recognition degree respectively; as well as If the first reference discrimination and the second reference discrimination are both greater than the first combined discrimination, the calculation unit determines that the test point corresponding to one of the first sensor and the second sensor is unsuitable, and performs a second determination operation; wherein: The plurality of points to be tested include a first point to be tested and an adjacent second point to be tested; The first sensor is located at the first point to be measured, and the second sensor is located at the second point to be measured; The first reference discrimination corresponds to the first sensor, and the second reference discrimination corresponds to the second sensor; as well as The first combined resolution corresponds to the first sensor and the second sensor.

11. The multi-point measurement method according to claim 10, characterized in that: The determination operation is the second determination operation, the combination is a second combination, the second combination includes the second sensor and a third sensor, the plurality of reference discriminations corresponding to the second combination include the second reference discrimination and a third reference discrimination, and the combined discrimination corresponding to the second combination is a second combined discrimination; the second determination operation includes: Using the calculation unit to compare the second reference discrimination and the third reference discrimination with the second combined discrimination respectively; as well as If the second reference discrimination and the third reference discrimination are both greater than the second combined discrimination, the calculation unit determines that the point to be tested corresponding to the second sensor is unsuitable; wherein: The plurality of test points further include a third test point adjacent to the second test point; The third sensor is located at the third point to be tested; The third reference discrimination corresponds to the third sensor; as well as The second combined resolution corresponds to the second sensor and the third sensor.

12. The multi-point measurement method according to claim 10, characterized in that: Each of the plurality of initial sensing data comprises a plurality of initial time series sensing data; and The plurality of initial time series sensing data correspond to a plurality of time windows respectively.

13. The multi-point measurement method according to claim 12, characterized in that: The device under measurement has an operation cycle; and The time unit range of the multiple time windows is 1 to 10 times of the operation cycle.

14. The multi-point measurement method according to claim 10, characterized in that: The data is preprocessed into a label setting; and The plurality of processed sensing data are a plurality of labeled sensing data.

15. The multi-point measurement method according to claim 14, characterized in that: The tag setting includes setting a plurality of tags; and The multiple labels include a loose screw, a power shortage, a maximum speed, a first speed, equipment abnormality, yield, other identifiable variation factors, or a combination thereof.

16. The multi-point measurement method according to claim 15, further comprising: The calculation unit performs the determination operation once according to each of the plurality of setting scenarios respectively corresponding to the plurality of tags.

17. The multi-point measurement method according to claim 10, characterized in that: The plurality of sensors are temperature sensors, humidity sensors, vibration sensors, or a combination thereof; and The plurality of initial sensing data are temperature data, humidity data, vibration data, or a combination thereof.

18. The multi-point measurement method according to claim 17, characterized in that: The calculation unit performs the determination operation once according to each initial sensing data.

Images