Production supervision method, device, server and storage medium

Abstract

The present invention discloses a production supervision method, device, server and storage medium, the method comprising: after acquiring the posture signal of the production personnel, extracting features from the posture signal to obtain a posture vector corresponding to the posture signal; adding time features to the posture vector to obtain a target vector; determining the loss function of the target vector based on the standard vector, and when the function value is greater than a preset threshold, determining the production abnormality, and making a production alarm based on the abnormal information. The above technical scheme acquires the posture signal of the production personnel on the production line, and determines the posture vector corresponding to the posture signal, adds time features to the posture vector to obtain the target vector, and after determining the loss function of the target vector, determines the production abnormality according to the loss function value being greater than the preset threshold, and makes an alarm based on the abnormal information, so as to realize real-time supervision of the production process, so as to remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel in their actions.

Description

Technical Field

[0001] The embodiments of the present invention relate to production management technology, and in particular to a production supervision method, device, server and storage medium. Background Art

[0002] The standardization of the production process can determine the quality and reputation of the products produced. During the production process, experienced production personnel often make mistakes such as irregular movements and wrong installation steps. At the same time, new production personnel who enter the production line usually simply follow the work instructions to perform production operations without receiving effective guidance, resulting in high trial and error costs for production factories.

[0003] In the existing technology, supervision of the production process mainly relies on the mode of inspection. The supervisor in charge of the process and production inspects the entire production line and corrects the irregular actions and incorrect processes of the production personnel. Of course, cameras can also be installed at key production stations, and the supervisor can monitor the production process of the assembly line through monitoring.

[0004] However, the existing technology cannot standardize the actions of production personnel during the production process, nor can it timely detect irregular operations of production personnel during the production process. In addition, special instructors are required to train production personnel, which is costly. Summary of the invention

[0005] The present invention provides a production supervision method, device, server and storage medium, which realize real-time supervision of the production process, timely alarm for abnormal information in the production process, and further standardize production actions.

[0006] In a first aspect, an embodiment of the present invention provides a production supervision method, comprising:

[0007] After acquiring the posture signal of the production personnel, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal;

[0008] Adding the time feature to the posture vector to obtain a target vector;

[0009] A loss function of the target vector is determined based on the standard vector, and when the function value is greater than a preset threshold, a production anomaly is determined, and a production alarm is issued based on the anomaly information.

[0010] The embodiment of the present invention provides a production supervision method, including: after acquiring the posture signal of the production personnel, extracting features from the posture signal to obtain a posture vector corresponding to the posture signal; adding a time feature to the posture vector to obtain a target vector; determining the loss function of the target vector based on a standard vector, and when the function value is greater than a preset threshold, determining a production anomaly, and making a production alarm based on the abnormal information. The above technical scheme can first acquire the posture signal of the production personnel on the production line, and then determine the posture vector corresponding to the posture signal, and add a time feature to the posture vector to obtain a target vector. After determining the loss function of the target vector, the production anomaly is determined according to the loss function value being greater than a preset threshold, and when the production is abnormal, an alarm is made based on the abnormal information, so as to realize real-time supervision of the production process, and timely remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel in action.

[0011] Furthermore, the posture signal includes a human posture signal acquired based on a wearable device worn by the production personnel.

[0012] Further, the posture signal includes a current posture signal and a historical posture signal, and correspondingly, the posture vector includes a current posture vector and a historical posture vector, and the target vector includes a current target vector and a historical target vector.

[0013] Further, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal, including:

[0014] The current posture signal and the historical posture signal are input as input signals into a convolutional neural network (CNN), and feature extraction is performed on the current posture signal and the historical posture signal based on the CNN to obtain the current posture vector and the historical posture vector.

[0015] Furthermore, the time feature includes a current time feature and a historical time feature. Accordingly, the time feature is added to the posture vector to obtain a target vector, including:

[0016] The current posture vector and the historical posture vector are input into a long short-term memory network LSTM as input signals. Based on the LSTM, the current time feature is added to the current posture vector to obtain the current target vector, and the historical time feature is added to the historical posture vector to obtain the historical target vector.

[0017] Further, a loss function of the target vector is determined based on the standard vector, and when the function value is greater than a preset threshold, production abnormality is determined, and a production alarm is issued based on the abnormal information, including:

[0018] Inputting the current target vector and the historical target vector as input signals into an extreme learning machine ELM, determining the loss function of the current target vector based on the standard vector in the ELM to obtain a current function value, and determining the loss function of the historical target vector based on the standard vector to obtain a historical function value;

[0019] A minimum function value is determined according to the current function value and the historical function value, and when the minimum function value is greater than a preset threshold, a production abnormality is determined, and a production alarm is issued based on the abnormality information.

[0020] Furthermore, the historical gesture signal includes a first historical gesture signal and a second historical gesture signal acquired based on a preset period.

[0021] In a second aspect, an embodiment of the present invention further provides a production supervision device, comprising:

[0022] A feature extraction module is used to extract features of the posture signal after acquiring the posture signal of the production personnel, so as to obtain a posture vector corresponding to the posture signal;

[0023] A target vector determination module, used for adding a time feature to the posture vector to obtain a target vector;

[0024] The alarm module is used to determine the loss function of the target vector based on the standard vector, determine the production abnormality when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information.

[0025] In a third aspect, an embodiment of the present invention further provides a server, the server comprising:

[0026] One or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the production supervision method as described in any one of the first aspects.

[0027] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute the production supervision method as described in any one of the first aspects.

[0028] In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the computer instructions are executed on a computer, the computer executes the production supervision method provided in the first aspect.

[0029] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the production monitoring device, or may be packaged separately from the processor of the production monitoring device, which is not limited in this application.

[0030] The description of the second, third, fourth and fifth aspects in this application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third, fourth and fifth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0031] In this application, the name of the above-mentioned production supervision device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0032] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A flowchart of a production supervision method provided in Embodiment 1 of the present invention;

[0034] Figure 2 A flowchart of a production supervision method provided in Embodiment 2 of the present invention;

[0035] Figure 3 A flowchart of a production supervision method provided in Embodiment 2 of the present invention;

[0036] Figure 4 A schematic diagram of the structure of a production monitoring device provided in Embodiment 3 of the present invention;

[0037] Figure 5 A schematic diagram of the structure of a server provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0038] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0039] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0040] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0041] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0042] It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe various operations (or steps) as sequential processes, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.

[0043] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0044] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0045] Embodiment 1

[0046] Figure 1 This is a flowchart of a production supervision method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where it is necessary to supervise the production status of production personnel and to alarm abnormal production. The method can be executed by a production supervision device, such as Figure 1 As shown, the method specifically comprises the following steps:

[0047] Step 110: After acquiring the posture signal of the production personnel, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal.

[0048] Among them, production personnel can wear wearable devices during the production process, and the wearable devices can be used to obtain the posture signals of the production personnel, specifically, the wrist posture signals, back posture signals and upper arm posture signals of the production personnel.

[0049] In addition, the wrist posture signal may include "0" indicating "normal wrist", "1" indicating "wrist raised", and "2" indicating "wrist lowered", and of course, other signals indicating other wrist postures may also be included; the back posture signal may include "0" indicating "back straight" and "1" indicating "back bent", and of course, other signals indicating other back posture movements may also be included; the upper arm posture signal may include "0" indicating "normal upper arm", "1" indicating "upper arm raised", and "2" indicating "upper arm lowered", and of course, other signals indicating other upper arm postures may also be included. It can be seen that the posture signal may be a vector with a dimension of 3, for example, the posture signal may be (0, 1, 2).

[0050] Specifically, the posture signal of the production personnel can be obtained based on the wearable device, and after the posture signal is obtained, the feature extraction of the posture signal is performed to convert the posture signal into a posture vector of a preset dimension. The preset dimension is not specifically limited here and can be determined according to actual needs. For example, the preset dimension can be 200. Of course, in practical applications, the posture signal can be input into a convolutional neural network (CNN) to extract features of the posture signal based on CNN, and then determine the posture vector corresponding to the posture signal.

[0051] In an embodiment of the present invention, the acquired posture signal of the production personnel can be preliminarily processed. Specifically, the posture vector corresponding to the posture signal can be determined to process the posture signal into a posture vector of a preset dimension, so as to facilitate subsequent processing of the posture vector and improve the efficiency of posture signal processing.

[0052] Step 120: Add the time feature to the posture vector to obtain a target vector.

[0053] The acquisition cycle of the gesture signal may be 0.3 seconds, and the gesture signal may be acquired once every 0.3 seconds. The acquired gesture signal may be a vector with a dimension of 3. The time feature may be the time information corresponding to the acquired gesture signal.

[0054] Specifically, the time feature corresponding to the posture signal can be added to the posture vector corresponding to the posture signal, and the result obtained after the addition is the target vector. Of course, in practical applications, the posture vector can be input into a long short-term memory network (LSTM) to add the time feature to the posture vector based on the LSTM, thereby obtaining the target vector. Of course, the dimension of the target vector obtained here can also be a preset dimension.

[0055] In the embodiment of the present invention, the target vector obtained by adding the time information corresponding to the posture signal to the posture vector corresponding to the posture signal is more precise and can be used to more accurately determine whether the production is abnormal.

[0056] Step 130: determine the loss function of the target vector based on the standard vector, determine that the production is abnormal when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information.

[0057] In which, a standard production posture signal is obtained during the production process, and features of the standard production posture signal are extracted to obtain a standard posture vector, and the time information corresponding to the standard production posture signal is further added to the standard posture vector to obtain a standard vector.

[0058] In addition, the preset threshold is not specifically limited here and can be determined according to actual needs. For example, the preset threshold may be 0.95.

[0059] In practical applications, the target vector can be input into an extreme learning machine (ELM) to determine the loss function of the target vector based on the standard vector in the ELM. By setting the number of hidden nodes in the network in the ELM, a unique optimal solution can be generated, so it has the advantages of fast learning speed and good generalization performance. In the embodiment of the present invention, the target vector can be expressed as (Xi, ti), where Xi = [xi 1 ,xi 2 ,…,xin] T ∈Rn,ti=[ti 1 ,ti 2 ,…,tin] T ∈Rm. A single hidden layer neural network with L hidden nodes can be expressed as Among them, g(x) is the activation function, W i =[w i1 ,w i2 ,…,w in ] T is the input weight, β i is the output weight, b i The bias of the i-th hidden unit, W i·X i W i and X i Therefore, the loss function can be determined

[0060] In the embodiment of the present invention, the input weight may be a standard vector, and the loss function of the target vector may be determined in the ELM based on the standard vector to determine the function value. The function value may also be compared with a preset threshold value, and of course, the preset threshold value may be 0.95. When the function value is greater than 0.95, production abnormality is determined, and a production alarm is issued based on the abnormal information to remind production personnel to correct production actions, and management personnel may also be reminded to record production abnormalities.

[0061] A production supervision method provided in the first embodiment of the present invention includes: after acquiring the posture signal of the production personnel, extracting features from the posture signal to obtain a posture vector corresponding to the posture signal; adding time features to the posture vector to obtain a target vector; determining the loss function of the target vector based on the standard vector, and when the function value is greater than a preset threshold, determining the production abnormality, and making a production alarm based on the abnormal information. The above technical scheme can first acquire the posture signal of the production personnel on the production line, and then determine the posture vector corresponding to the posture signal, and add time features to the posture vector to obtain the target vector. After determining the loss function of the target vector, the production abnormality is determined according to the loss function value being greater than the preset threshold, and when the production is abnormal, an alarm is made based on the abnormal information, so as to realize real-time supervision of the production process, and timely remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel in action.

[0062] Embodiment 2

[0063] Figure 2 This is a flowchart of a production supervision method provided in Embodiment 2 of the present invention. This embodiment is specific based on the above embodiment. Figure 2 As shown, in this embodiment, the method may further include:

[0064] Step 210: Acquire the posture signal of the production personnel.

[0065] Preferably, the posture signal includes a human posture signal acquired based on a wearable device worn by the production personnel.

[0066] Likewise, the human body posture signal may include the production personnel's wrist posture signal, back posture signal, upper arm posture signal, and the like.

[0067] Furthermore, the gesture signal includes a current gesture signal and a historical gesture signal, and the historical gesture signal includes a first historical gesture signal and a second historical gesture signal acquired based on a preset period.

[0068] Since the acquisition cycle of the gesture signal is 0.3 seconds, the first historical gesture signal can be the gesture signal acquired in the previous 0.3 seconds, and the second historical signal can be the gesture signal acquired in the previous 0.6 seconds. Of course, the current gesture signal, the first historical gesture signal, and the second historical gesture signal can all be vectors with a dimension of 3. For example, the current gesture signal can be (1, 1, 1), the first historical gesture signal can be (2, 2, 2), and the second historical gesture signal can be (3, 3, 3).

[0069] In an embodiment of the present invention, since the movements of production personnel during the production process are relatively fast and the current posture is only related to the posture in a short period of time before, the current posture signal, the first historical posture signal of the previous 0.3 seconds, and the second historical posture signal of the previous 0.6 seconds can be obtained to analyze and supervise the current posture based on the current posture signal, the first historical posture signal, and the second historical posture signal.

[0070] Step 220: extract features from the gesture signal to obtain a gesture vector corresponding to the gesture signal.

[0071] The posture vector includes a current posture vector and a historical posture vector, and the historical posture vector may include a first historical posture vector corresponding to a first historical posture signal and a second historical posture vector corresponding to a second historical posture signal.

[0072] In one implementation, step 220 may specifically include:

[0073] The current posture signal and the historical posture signal are input as input signals into a convolutional neural network (CNN), and feature extraction is performed on the current posture signal and the historical posture signal based on the CNN to obtain the current posture vector and the historical posture vector.

[0074] Specifically, the current posture signal, the first historical posture signal, and the second historical posture signal can be input into CNN as input signals. Based on CNN, feature extraction is performed on the current posture signal, the first historical posture signal, and the second historical posture signal to obtain the current posture vector, the first historical posture vector, and the second historical posture vector.

[0075] In the embodiment of the present invention, the current posture vector corresponding to the current posture signal, the first historical posture vector corresponding to the first historical posture signal, and the second historical posture signal corresponding to the second historical posture signal can be determined, and the initial processing of the posture signal is realized based on CNN. Of course, the dimensions of the current posture vector, the first historical posture vector, and the second historical posture vector are all preset dimensions, and the specific dimension can be 200.

[0076] Step 230: Add the time feature to the posture vector to obtain a target vector.

[0077] Among them, the time features include current time features and historical time features, the historical time features include first historical time features and second historical time features, the target vector includes current target vector and historical target vector, the historical target vector includes first historical target vector and second historical target vector.

[0078] In one implementation, step 230 may specifically include:

[0079] The current posture vector and the historical posture vector are input into a long short-term memory network LSTM as input signals. Based on the LSTM, the current time feature is added to the current posture vector to obtain the current target vector, and the historical time feature is added to the historical posture vector to obtain the historical target vector.

[0080] Specifically, the current posture vector, the first historical target vector and the second historical target vector are input into the LSTM as input signals, and the current time feature is added to the current posture vector based on the LSTM to obtain the current target vector; the first historical time feature is added to the first historical posture vector to obtain the first historical target vector; the second historical time feature is added to the second historical posture vector to obtain the second historical target vector.

[0081] In an embodiment of the present invention, the current target vector, as well as the first historical target vector and the second historical target vector that are most correlated with the current target vector can be determined, and the dimensions of the current target vector, the first historical target vector and the second historical target vector are consistent with the dimensions of the current posture vector, the first historical posture vector and the second historical posture vector.

[0082] Step 240: determine the loss function of the target vector based on the standard vector, determine that the production is abnormal when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information.

[0083] In one implementation, step 240 may specifically include:

[0084] The current target vector and the historical target vector are input into an extreme learning machine ELM as input signals. In the ELM, a loss function of the current target vector is determined based on the standard vector to obtain a current function value, and a loss function of the historical target vector is determined based on the standard vector to obtain a historical function value. A minimum function value is determined based on the current function value and the historical function value, and when the minimum function value is greater than a preset threshold, a production abnormality is determined, and a production alarm is issued based on the abnormal information.

[0085] Specifically, the current target vector, the first historical target vector, and the second historical target vector are input into the ELM as input signals. In the ELM, the loss function of the current target vector is determined based on the standard vector to obtain the current function value; the loss function of the first historical target vector is determined based on the standard vector to obtain the first historical function value; the loss function of the second historical target vector is determined based on the standard vector to obtain the second historical function value. Then, the minimum value among the current function value, the first historical function value, and the second historical function value is determined as the minimum function value, and the minimum function value is compared with the preset threshold value. If the minimum function value is greater than the preset threshold value, the production abnormality is determined, and a production alarm is issued based on the abnormal information; if the minimum function value is less than or equal to the preset threshold value, the posture signal of the production personnel continues to be obtained.

[0086] In the embodiment of the present invention, the first historical target vector and the second historical target vector with the highest correlation with the current target vector can be used to determine the loss function value, so that the obtained loss function value is more accurate. Similarly, when the loss function value is greater than 0.95, the production abnormality is determined, and a production alarm is issued based on the abnormal information to remind the production personnel to correct the production action, and the management personnel can also be reminded to record the production abnormality. This makes the determination of production abnormalities more accurate, and further improves the accuracy and effectiveness of production supervision.

[0087] A production supervision method provided in the second embodiment of the present invention includes: after acquiring the posture signal of the production personnel, extracting features from the posture signal to obtain the posture vector corresponding to the posture signal; adding the time feature to the posture vector to obtain the target vector; determining the loss function of the target vector based on the standard vector, and when the function value is greater than a preset threshold, determining the production abnormality, and making a production alarm based on the abnormal information. The above technical scheme can first obtain the posture signal of the production personnel on the production line, and then determine the posture vector corresponding to the posture signal, and add the time feature to the posture vector to obtain the target vector. After determining the loss function of the target vector, the production abnormality is determined according to the loss function value being greater than the preset threshold, and when the production is abnormal, an alarm is made based on the abnormal information, so as to realize real-time supervision of the production process, and timely remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel in action.

[0088] In addition, the above-mentioned production supervision method can enable production personnel to standardize their own production actions according to the alarm, further improve the standardization of the production process, and reduce the learning cost of production personnel and the need for guidance.

[0089] Figure 3 The following is a flowchart of a production supervision method provided in the second embodiment of the present invention, which exemplifies one of the implementation methods. Figure 3 As shown, including:

[0090] Step 310: Acquire a current posture signal of the production personnel, and acquire a first historical posture signal and a second historical posture signal based on a preset period.

[0091] Step 320: Based on CNN, feature extraction is performed on the current posture signal, the first historical posture signal, and the second historical posture signal to obtain the current posture vector, the first historical posture vector, and the second historical posture vector corresponding to the current posture signal, the first historical posture signal, and the second historical posture signal.

[0092] Step 330: Based on LSTM, the current time feature is added to the current posture vector to obtain the current target vector; the first historical time feature is added to the first historical posture vector to obtain the first historical target vector; the second historical time feature is added to the second historical posture vector to obtain the second historical target vector.

[0093] Step 340: In ELM, determine the loss function of the current target vector based on the standard vector to obtain the current function value, determine the loss function of the first historical target vector based on the standard vector to obtain the first historical function value, and determine the loss function of the second historical target vector based on the standard vector to obtain the second historical function value.

[0094] Step 350: Determine a minimum function value according to the current function value, the first historical function value, and the second historical function value.

[0095] Step 360: Determine whether the minimum function value is greater than a preset threshold.

[0096] If the minimum function value is greater than the preset threshold, execute step 370 ; otherwise, return to execute step 310 .

[0097] Step 370: Determine production anomalies and issue production alarms based on the anomaly information.

[0098] A second embodiment of the present invention provides an implementation method of a production supervision method, including: obtaining a current posture signal of a production personnel, and a first historical posture signal and a second historical posture signal obtained based on a preset period; performing feature extraction on the current posture signal, the first historical posture signal, and the second historical posture signal based on CNN, respectively, to obtain a current posture vector, a first historical posture vector, and a second historical posture vector corresponding to the current posture signal, the first historical posture signal, and the second historical posture signal; adding a current time feature to the current posture vector based on LSTM to obtain a current target vector; adding the first historical time feature to the first historical posture vector to obtain a first historical target vector; adding the second historical time feature to the second historical posture vector vector, obtain the second historical target vector; determine the loss function of the current target vector based on the standard vector in ELM to obtain the current function value, determine the loss function of the first historical target vector based on the standard vector to obtain the first historical function value, and determine the loss function of the second historical target vector based on the standard vector to obtain the second historical function value; determine the minimum function value according to the current function value, the first historical function value and the second historical function value; determine whether the minimum function value is greater than the preset threshold; if the minimum function value is greater than the preset threshold, determine the production abnormality, and make a production alarm based on the abnormal information; otherwise, return to execute the current posture signal of the production personnel, and the first historical posture signal and the second historical posture signal obtained based on the preset period. The above technical scheme can first obtain the posture signal of the production personnel on the production line, and then determine the posture vector corresponding to the posture signal, and add time features to the posture vector to obtain the target vector. After determining the loss function of the target vector, determine the production abnormality according to the loss function value being greater than the preset threshold, and when the production is abnormal, alarm based on the abnormal information, so as to realize real-time supervision of the production process, and timely remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel.

[0099] Embodiment 3

[0100] Figure 4 This is a schematic diagram of the structure of a production monitoring device provided in the third embodiment of the present invention, which can be applied to situations where the production status of production personnel needs to be monitored and an alarm is issued for abnormal production. The device can be implemented by software and / or hardware and is generally integrated in a server.

[0101] like Figure 4 As shown, the device comprises:

[0102] The feature extraction module 410 is used to extract features from the posture signal of the production personnel after acquiring the posture signal, so as to obtain a posture vector corresponding to the posture signal;

[0103] A target vector determination module 420, configured to add a time feature to the posture vector to obtain a target vector;

[0104] The alarm module 430 is used to determine the loss function of the target vector based on the standard vector, determine the production abnormality when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information.

[0105] The production supervision device provided in this embodiment, after acquiring the posture signal of the production personnel, performs feature extraction on the posture signal to obtain the posture vector corresponding to the posture signal; adds the time feature to the posture vector to obtain the target vector; determines the loss function of the target vector based on the standard vector, and when the function value is greater than a preset threshold, determines the production abnormality, and makes a production alarm based on the abnormal information. The above technical scheme can first obtain the posture signal of the production personnel on the production line, and then determine the posture vector corresponding to the posture signal, and add the time feature to the posture vector to obtain the target vector. After determining the loss function of the target vector, the production abnormality is determined according to the loss function value being greater than the preset threshold, and when the production is abnormal, an alarm is made based on the abnormal information, so as to realize real-time supervision of the production process, and timely remind the production personnel to standardize the production actions, and also remind the instructor to guide the production personnel in action.

[0106] Based on the above embodiment, the posture signal includes a human posture signal obtained based on a wearable device worn by the production personnel.

[0107] On the basis of the above embodiment, the posture signal includes a current posture signal and a historical posture signal, and accordingly, the posture vector includes a current posture vector and a historical posture vector, and the target vector includes a current target vector and a historical target vector.

[0108] Based on the above embodiment, the feature extraction module 410 is specifically used for:

[0109] After obtaining the posture signal of the production personnel, the current posture signal and the historical posture signal are input into the convolutional neural network CNN as input signals, and the features of the current posture signal and the historical posture signal are extracted based on the CNN to obtain the current posture vector and the historical posture vector.

[0110] Based on the above embodiment, the time feature includes a current time feature and a historical time feature. Accordingly, the target vector determination module 420 is specifically configured to:

[0111] The current posture vector and the historical posture vector are input into a long short-term memory network LSTM as input signals. Based on the LSTM, the current time feature is added to the current posture vector to obtain the current target vector, and the historical time feature is added to the historical posture vector to obtain the historical target vector.

[0112] Based on the above embodiment, the alarm module 430 is specifically used for:

[0113] Inputting the current target vector and the historical target vector as input signals into an extreme learning machine ELM, determining the loss function of the current target vector based on the standard vector in the ELM to obtain a current function value, and determining the loss function of the historical target vector based on the standard vector to obtain a historical function value;

[0114] A minimum function value is determined according to the current function value and the historical function value, and when the minimum function value is greater than a preset threshold, a production abnormality is determined, and a production alarm is issued based on the abnormality information.

[0115] Based on the above embodiment, the historical gesture signal includes a first historical gesture signal and a second historical gesture signal acquired based on a preset period.

[0116] The production supervision device provided in the embodiment of the present invention can execute the production supervision method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0117] Embodiment 4

[0118] Figure 5 A schematic diagram of the structure of a server provided in Embodiment 4 of the present invention. Figure 5 A block diagram of an exemplary server suitable for implementing embodiments of the present invention is shown. Figure 5 The server 5 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0119] like Figure 5 As shown, the server 5 is in the form of a general purpose computing electronic device. The components of the server 5 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0120] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0121] The server 5 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the server 5, including volatile and non-volatile media, removable and non-removable media.

[0122] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The server 5 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0123] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0124] The server 5 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the server 5, and / or any device that enables the server 5 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the server 5 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the server 5 via the bus 18. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with the server 5, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0125] The processing unit 16 executes various functional applications and page displays by running the program stored in the system memory 28, for example, implementing the production supervision method provided by the embodiment of the present invention, the method comprising:

[0126] After acquiring the posture signal of the production personnel, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal;

[0127] Adding the time feature to the posture vector to obtain a target vector;

[0128] A loss function of the target vector is determined based on the standard vector, and when the function value is greater than a preset threshold, a production anomaly is determined, and a production alarm is issued based on the anomaly information.

[0129] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the production supervision method provided by any embodiment of the present invention.

[0130] Embodiment 5

[0131] Embodiment 5 of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the production supervision method provided by the embodiment of the present invention is implemented, for example, and the method includes:

[0132] After acquiring the posture signal of the production personnel, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal;

[0133] Adding the time feature to the posture vector to obtain a target vector;

[0134] A loss function of the target vector is determined based on the standard vector, and when the function value is greater than a preset threshold, a production anomaly is determined, and a production alarm is issued based on the anomaly information.

[0135] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0136] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0137] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0138] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0139] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0140] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A production supervision method, characterized in that: include: After acquiring the posture signal of the production personnel, feature extraction is performed on the posture signal to obtain a posture vector corresponding to the posture signal; Adding the time feature to the posture vector to obtain a target vector; Determine a loss function of the target vector based on the standard vector, determine a production abnormality when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information; Wherein, the posture signal includes a current posture signal and a historical posture signal, and correspondingly, the posture vector includes a current posture vector and a historical posture vector, and the target vector includes a current target vector and a historical target vector; The step of extracting features from the gesture signal to obtain a gesture vector corresponding to the gesture signal includes: Inputting the current posture signal and the historical posture signal as input signals into a convolutional neural network (CNN), and performing feature extraction on the current posture signal and the historical posture signal based on the CNN to obtain the current posture vector and the historical posture vector; Wherein, the time feature includes a current time feature and a historical time feature. Accordingly, the time feature is added to the posture vector to obtain a target vector, including: The current posture vector and the historical posture vector are input as input signals into a long short-term memory network LSTM, and based on the LSTM, the current time feature is added to the current posture vector to obtain the current target vector, and the historical time feature is added to the historical posture vector to obtain the historical target vector; The loss function of determining the target vector based on the standard vector, determining production abnormality when the function value is greater than a preset threshold, and issuing a production alarm based on the abnormal information, includes: Inputting the current target vector and the historical target vector as input signals into an extreme learning machine ELM, determining the loss function of the current target vector based on the standard vector in the ELM to obtain a current function value, and determining the loss function of the historical target vector based on the standard vector to obtain a historical function value; A minimum function value is determined according to the current function value and the historical function value, and when the minimum function value is greater than a preset threshold, a production abnormality is determined, and a production alarm is issued based on the abnormality information.

2. The production supervision method according to claim 1, characterized in that: The posture signal includes a human posture signal acquired based on a wearable device worn by the production personnel.

3. The production supervision method according to any one of claim 1, characterized in that: The historical gesture signal includes a first historical gesture signal and a second historical gesture signal acquired based on a preset period.

4. A production supervision device, characterized in that: include: A feature extraction module is used to extract features of the posture signal after acquiring the posture signal of the production personnel, so as to obtain a posture vector corresponding to the posture signal; A target vector determination module, used for adding a time feature to the posture vector to obtain a target vector; An alarm module is used to determine the loss function of the target vector based on the standard vector, determine the production abnormality when the function value is greater than a preset threshold, and issue a production alarm based on the abnormal information; Wherein, the posture signal includes a current posture signal and a historical posture signal, and correspondingly, the posture vector includes a current posture vector and a historical posture vector, and the target vector includes a current target vector and a historical target vector; The feature extraction module is specifically used to obtain the posture signal of the production personnel, input the current posture signal and the historical posture signal as input signals into a convolutional neural network CNN, and perform feature extraction on the current posture signal and the historical posture signal based on the CNN to obtain the current posture vector and the historical posture vector; Wherein, the time feature includes a current time feature and a historical time feature. Accordingly, the target vector determination module is specifically used to input the current posture vector and the historical posture vector as input signals into a long short-term memory network LSTM, add the current time feature to the current posture vector based on the LSTM to obtain the current target vector, and add the historical time feature to the historical posture vector to obtain the historical target vector; Among them, the alarm module is specifically used to input the current target vector and the historical target vector as input signals into the extreme learning machine ELM, and determine the loss function of the current target vector based on the standard vector in the ELM to obtain the current function value, and determine the loss function of the historical target vector based on the standard vector to obtain the historical function value; determine the minimum function value according to the current function value and the historical function value, and when the minimum function value is greater than a preset threshold, determine the production abnormality, and make a production alarm based on the abnormal information.

5. A server, characterized in that: The server comprises: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the production supervision method as described in any one of claims 1 to 3.

6. A storage medium comprising computer executable instructions, wherein the computer executable instructions are used to execute the production supervision method according to any one of claims 1 to 3 when executed by a computer processor.

Images