Smart factory edge data collection method, system and medium based on the Internet of Things

Abstract

The present invention relates to the field of edge data acquisition, and specifically to an edge data acquisition method, system and medium for a smart factory based on the Internet of Things, comprising: an edge acquisition module, a local modeling module, an abnormality recognition module, an edge communication module and a local adjustment module. The edge acquisition module is used to collect working parameters and environmental parameters of production equipment, the local modeling module is used to calculate end point coordinates and an environmental impact matrix, the abnormality recognition module is used to correct the acquired environmental data and identify abnormal sensors, the edge communication module is used to build a communication link, and the local adjustment module is used to adjust the variable parameters of the production equipment and automatically adjust them. The present invention can achieve the integration and consistency of different data sources, ensure the quality and availability of data, and at the same time help to improve data quality, timely discover sensor failures or inaccurate data, and achieve a more efficient and safer production mode.

Description

Technical Field

[0001] The present invention relates to the field of edge data collection, and specifically to an edge data collection method, system and medium for a smart factory based on the Internet of Things. Background Art

[0002] A smart factory refers to a factory that uses advanced information technology and automation technology to realize the intelligentization, automation and digitization of the production process. Edge data collection refers to the collection and processing of data through IoT devices at data sensors or devices close to the data source. In the industrial Internet, it is generally necessary to transmit the data collected by each edge device to the edge server, which determines the operating status of the device and issues instructions to adjust the device.

[0003] The data collected by edge devices include environmental factors such as temperature, brightness, production settings and energy supply, as well as production factors such as equipment operating status and product quality indicators. Collecting this data requires a sophisticated sensing system, but sensors may be affected by the environment, resulting in unstable data quality and affecting the accuracy of subsequent data analysis and decision-making.

[0004] In addition, the IoT devices and edge servers are connected through a gateway, and the protocol conversion process consumes a lot of time, so that the data analysis and adjustment instructions of the edge server cannot reach the IoT control system in time, causing a certain degree of lag in the equipment adjustment behavior, and it is impossible to timely analyze the impact of the adjustment behavior on the factory production quality. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent factory edge data acquisition system and method based on the Internet of Things to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an edge data acquisition system for a smart factory based on the Internet of Things, comprising: an edge acquisition module, a local modeling module, an anomaly recognition module, an edge communication module and a local adjustment module;

[0007] The edge acquisition module is used to set up edge IoT devices at factory equipment, collect working parameters of the devices through sensors on the edge IoT devices, and the IoT devices record all collected data in the database. The database is connected to the edge server via an edge gateway, and the link can transmit data bidirectionally according to a preset communication protocol;

[0008] The local modeling module is used to calculate the dependency of various types of data, and divide the collected data of the IoT device into dependent data and environmental data according to the dependency. A multidimensional space model is constructed in the edge device with the dependent data. The environmental data constitutes a pointing vector, and the end point of the pointing vector in the structure is recorded. According to the historical environmental data and the end point coordinates, a multidimensional influence function of the dependent data and the environmental data is calculated for each dependent data, and the influence function is matrixed to obtain an environmental influence matrix;

[0009] The abnormal identification module is used to transmit environmental data and the end point coordinates of the pointing vector to the edge server in real time through the transmission interface in the edge device. The edge server calculates the theoretical pointing interval through the environmental impact matrix. When the end point coordinate exceeds the interval range, the impact ratio is calculated according to the components of the difference vector between the center point of the interval and the end point in each dependent variable dimension, the second communication protocol is activated, the probability distribution function of the environmental data is obtained, the sensor with a probability less than the threshold is replaced, and the acquired environmental data is corrected according to the impact ratio;

[0010] The edge communication module is used to build a communication link between the edge device and the edge server, and provides protocol conversion and communication proxy paths. The supported conversion protocols include: Modbus protocol, OPC protocol, Profibus protocol, MQTT protocol and EtherNet / IP protocol. It also provides a data rollback function to restore the data status when an error or unexpected situation occurs in the database communication;

[0011] The local adjustment module is used to adjust the variable parameters in the environmental data in the edge server so that the theoretical pointing point coincides with the initial pointing point, and to recover the coordinates of the end point of the pointing vector before and after the adjustment. When the coordinates of the end point after adjustment are closer to the center point of the theoretical pointing interval than before the adjustment, the adjustment behavior is recorded in the edge device as a positive adjustment mode. The edge device calculates the matching degree of the collected data and each positive adjustment mode in real time. When it is higher than the threshold, the device parameters are adjusted according to the positive adjustment mode.

[0012] Furthermore, the edge acquisition module includes: an IoT device unit and a database unit;

[0013] The IoT device unit is used to collect operating parameters of factory production equipment, environmental parameters in the production environment, and setting parameters of production equipment;

[0014] The database unit is connected to the edge server via a network interface, and is used to store the collected data, and to classify, store and analyze the data.

[0015] Further, the local modeling module includes: a data classification unit, a space pointing unit and a dependent variable influence unit;

[0016] The data classification unit is used to calculate the dependent variability of each type of data, and the parameters whose dependent variability is not zero are used as dependent parameters, and the parameters whose dependent variability is zero are used as environmental parameters;

[0017] The spatial pointing unit is used to establish a multi-dimensional spatial model in the edge device, the dimension of which is consistent with the number of dependent parameters, and to generate a multi-dimensional pointing vector from the environmental data, the dimension of which is consistent with the number of environmental parameters;

[0018] The dependent variable influencing unit is used to project the multi-dimensional pointing vector on the multi-dimensional space model, obtain the pointing end point coordinates of the projected vector, and send them to the edge server.

[0019] Furthermore, the abnormality identification module includes: an interval identification unit and a parameter troubleshooting unit;

[0020] The interval identification unit is arranged in the edge server and is used to determine whether the end point coordinates transmitted back by the edge device are within the theoretical pointing interval;

[0021] The parameter checking unit is used to calculate the influence ratio of the dependent variable data on each environmental data according to the components of the difference vector between the center point and the end point of the interval in each dependent variable dimension.

[0022] Further, the edge communication module includes: a protocol switching unit and a data rollback unit;

[0023] The protocol switching unit activates a second communication protocol in the edge server, the second communication protocol enabling the exchange of environmental data records in the edge device database;

[0024] The data rollback unit is used to detect the communication link between the edge device and the edge server, and to restore the data status when an error or unexpected situation occurs in the database communication.

[0025] Further, the local adjustment module includes: a central command unit, a mode recording unit and an edge adjustment unit;

[0026] The central command unit is used to calculate the adjustment value of the variable parameter of the device when the end point coordinates do not exceed the interval range, so that the adjusted end point coordinates coincide with the theoretical pointing point, and determine the parameter adjustment weight according to the influence ratio;

[0027] The mode recording unit is used to receive feedback data after parameter adjustment, and when the distance between the end point coordinates in the feedback data and the theoretical pointing point decreases, record the adjustment conditions and adjustment behaviors in the edge device;

[0028] The edge adjustment unit is used to calculate the matching degree between the collected environmental parameters and the adjustment conditions in the edge device, and when it is higher than a threshold, it points to the adjustment behavior corresponding to the adjustment condition.

[0029] The edge data collection method of a smart factory based on the Internet of Things includes the following steps:

[0030] Step S1. The edge device sensor collects various types of data at fixed intervals and stores them in the database. The average correlation of the sampling sequence is calculated according to the first method. The data with correlation higher than the threshold is stored as dependent parameters, otherwise it is stored as environmental parameters.

[0031] Step S2. Establish a multidimensional space model based on dependent parameters, the dimension of which is consistent with the number of dependent parameters, generate a multidimensional pointing vector based on environmental data, the dimension of which is consistent with the number of environmental parameters, project the multidimensional pointing vector in the multidimensional space model, and obtain the pointing end point coordinates of the projected vector;

[0032] Step S3. Set an unknown processing matrix between the pointing vector and the pointing end point coordinates, substitute the historical dependent parameters and the pointing end point coordinates into the unknown processing matrix, solve the environmental impact matrix, and the edge server processes the currently collected environmental parameters through the environmental impact matrix to obtain theoretical coordinates;

[0033] Step S4. Use the actual coordinates to point to the theoretical coordinates, generate an error vector, calculate the projection length of the error vector in each environmental parameter dimension, and obtain the change in the environmental parameter. When the change in the environmental parameter within a unit period exceeds a threshold, the corresponding sensor is repaired;

[0034] Step S5. Take the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjust the variable parameters in the environmental data, so that the adjusted theoretical coordinates coincide with the optimal endpoint coordinates. If the distance between the actual coordinates and the optimal endpoint coordinates decreases before and after the adjustment, the adjustment process is recorded to the edge device. When the edge device collects the same environmental parameters, it actively adjusts the variable parameters.

[0035] Further, step S1 includes:

[0036] Step S11. The edge device is set in the production environment of the factory to collect the working parameters of the production equipment, which include variable parameters and immutable parameters. The variable parameters also include: production speed, production quality, production energy consumption, setting parameters, energy supply parameters and setting parameters of environmental conditioning facilities of the production equipment. The immutable parameters also include: temperature, humidity, sound and brightness.

[0037] Step S12. The edge device partitions and stores the sampled data according to the parameter type, arranges them in chronological order, and forms a sampling sequence. For each parameter, the average correlation of the sampling sequence is calculated according to the first method. The first method is as follows;

[0038] ;

[0039] Where Q represents the average correlation of the sampling sequence, m represents the number of working parameters, n represents the length of the sampling sequence, Wi represents the sampling sequence corresponding to the i-th working parameter, and Wi j represents the jth element in the sequence Wi, represents the mean of the elements in the sequence Wi, represents the jth element in the sampling sequence of the current parameter, Represents the mean of the elements in the sampling sequence of the current parameter;

[0040] Step S13. Preset a threshold Q0. If the average correlation Q>Q0 of the parameter, the parameter is stored as a dependent function. If Q≤Q0, the parameter is stored as an environmental parameter.

[0041] Further, step S2 includes:

[0042] Step S21. The database of the edge device is connected to the edge server via the edge gateway, and data is transmitted bidirectionally according to the preset communication protocol. The edge server obtains the environmental parameters and dependent parameters stored in the database. The number of environmental parameters is recorded as t1, and the number of dependent parameters is recorded as t2;

[0043] Step S22: Establish a t2-dimensional virtual space in the edge server, and generate actual coordinates W in the virtual space according to the current values ​​of the dependent parameters: (W1, W2, ..., W t2 ), where W1, W2, …, W t2 Represent the 1st, 2nd, …t2th dependent parameter values ​​respectively;

[0044] At the same time, starting from zero, points (H1, H2, ..., H t1 ) generates a t1-dimensional pointing vector for the end point, where H1, H2, …, H t1 Represent the 1st, 2nd, …, t1th environmental parameter values ​​respectively, map the pointing vector in the t2-dimensional virtual space, and obtain the end point coordinate H of the mapping vector: (HE1, HE2, …, HE t2 ), where HE1, HE2, …, HE t2 They represent the mapping values ​​of the projection vector in the 1st, 2nd, …t2 dimensions respectively.

[0045] Further, step S3 includes:

[0046] Step S31. Preset the unknown matrix A with t2 rows and t2 columns, and list the state transformation equation:

[0047] ;

[0048] Among them A (1,1) To A (t2,t2)Respectively represent the values ​​of the 1st row and 1st column to the t2th row and t2th column in the unknown matrix A;

[0049] Step S32. Find history t2 in the database 2 Substitute the environmental parameters and dependent parameters obtained in the sampling into the state transformation equation, solve the value of each unknown element in the unknown matrix A, and obtain the environmental impact matrix B;

[0050] Step S33: The edge server obtains the environmental parameters currently collected by the edge device, calculates the end point coordinates according to the method of step S22, and processes the end point coordinates with the environmental impact matrix B to obtain the theoretical coordinates E.

[0051] Further, step S4 includes:

[0052] Step S41. Generate error vector , project the error vector into the t1-dimensional space, and calculate the projection length of the error vector in each dimension, which is recorded as the set R={R1,R2,…,R t1}, where R1, R2, …, R t1 Respectively represent the environmental parameter changes in the 1st, 2nd, …t1th dimensions of the error vector;

[0053] Step S42. Compare the elements in the set R with the maximum environmental change threshold within the unit sampling period in turn. When the threshold is exceeded, mark the environmental parameter corresponding to the element as an abnormal parameter, and repair the sensor that collects the abnormal parameter in the edge device.

[0054] Further, step S5 includes:

[0055] Step S51. Take the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjust the variable parameters in the environmental data, so that the projection endpoint of the pointing vector of the adjusted environmental parameters in the t2-dimensional virtual space coincides with the optimal endpoint coordinates after being processed by the environmental impact matrix B, and record the adjustment parameters and adjustment amounts of the edge server;

[0056] Step S52. In the next sampling cycle after the variable parameters are adjusted, the actual coordinates corresponding to the sampled data are obtained. If the distance between the actual coordinates and the optimal end point coordinates is less than the distance between the actual coordinates and the optimal end point coordinates in the previous cycle, the adjustment parameters, adjustment amounts and current environmental parameter values ​​of the edge server are sent to the edge device, forming a positive adjustment mode;

[0057] Step S53: When the edge device collects the same environmental parameters as those recorded in the forward adjustment mode, it automatically inspects and adjusts the variable parameters according to the adjustment parameters and adjustment amounts recorded in the forward adjustment mode.

[0058] An edge data acquisition medium for a smart factory based on the Internet of Things, wherein the medium stores computer software, a network transmission protocol, and an agent communication protocol for executing an edge data acquisition method for a smart factory based on the Internet of Things, and is capable of establishing a communication connection between an edge device and an edge server.

[0059] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0060] 1. The present invention can calculate the dependency of various types of data, divide the collected data of IoT devices into dependent data and environmental data, use the dependent data to form a multi-dimensional structure, and the environmental data to form a pointing vector, so as to model the collected data and better analyze the correlation status of the data, realize the integration and consistency of different data sources, and improve the quality and availability of data.

[0061] 2. The present invention calculates the theoretical pointing interval through the environmental impact matrix. When the actual pointing point exceeds the interval range, the second protocol is turned on for the dependent data sensor based on the components of the difference vector in each dimension, and the change rate of the pointing vector in different dimensions is calculated. The corresponding sensor for the parameter with a change rate greater than the threshold is replaced, which helps to improve data quality, timely discover sensor failures or inaccurate data, avoid making wrong decisions, and improve work efficiency and production quality.

[0062] 3. The present invention enables the edge server to adjust the variable parameters in the environmental data so that the theoretical pointing point coincides with the initial pointing point, and stores the adjustment feedback as a positive adjustment mode in the edge IoT device. When the collected data is consistent with the stored adjustment mode, the adjustment is directly performed, thereby improving the equipment adjustment efficiency and realizing a more efficient and safer production mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0064] Figure 1 It is a structural schematic diagram of the smart factory edge data acquisition system based on the Internet of Things of the present invention.

[0065] Figure 2 It is a schematic diagram of the steps of the smart factory edge data collection method based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] See also Figure 1 ,The present invention provides a technical solution: an edge data acquisition system for a smart factory based on the Internet of Things, comprising: an edge acquisition module, a local modeling module, an anomaly recognition module, an edge communication module and a local adjustment module;

[0068] The edge acquisition module is used to set up edge IoT devices at factory equipment, collect working parameters of the devices through sensors on the edge IoT devices, and the IoT devices record all collected data in the database. The database is connected to the edge server via an edge gateway, and the link can transmit data bidirectionally according to a preset communication protocol;

[0069] The local modeling module is used to calculate the dependency of various types of data, and divide the collected data of the IoT device into dependent data and environmental data according to the dependency. A multidimensional space model is constructed in the edge device with the dependent data. The environmental data constitutes a pointing vector, and the end point of the pointing vector in the structure is recorded. According to the historical environmental data and the end point coordinates, a multidimensional influence function of the dependent data and the environmental data is calculated for each dependent data, and the influence function is matrixed to obtain an environmental influence matrix;

[0070] The abnormality identification module is used to transmit environmental data and the endpoint coordinates of the pointing vector to the edge server in real time through the transmission interface in the edge device. The edge server calculates the theoretical pointing interval through the environmental impact matrix. When the endpoint coordinate exceeds the interval range, the impact ratio is calculated according to the components of the difference vector between the center point of the interval and the endpoint in each dependent variable dimension, the second communication protocol is activated, the probability distribution function of the environmental data is obtained, the sensor with a probability less than a threshold is replaced, and the acquired environmental data is corrected according to the impact ratio.

[0071] The edge communication module is used to build a communication link between the edge device and the edge server, and provides protocol conversion and communication proxy paths. The supported conversion protocols include: Modbus protocol, OPC protocol, Profibus protocol, MQTT protocol and EtherNet / IP protocol. It also provides a data rollback function to restore the data status when an error or unexpected situation occurs in the database communication;

[0072] The local adjustment module is used to adjust the variable parameters in the environmental data in the edge server so that the theoretical pointing point coincides with the initial pointing point, and to recover the coordinates of the end point of the pointing vector before and after the adjustment. When the coordinates of the end point after adjustment are closer to the center point of the theoretical pointing interval than before the adjustment, the adjustment behavior is recorded in the edge device as a positive adjustment mode. The edge device calculates the matching degree of the collected data and each positive adjustment mode in real time. When it is higher than the threshold, the device parameters are adjusted according to the positive adjustment mode.

[0073] Furthermore, the edge acquisition module includes: an IoT device unit and a database unit;

[0074] The IoT device unit is used to collect operating parameters of factory production equipment, environmental parameters in the production environment, and setting parameters of production equipment;

[0075] The database unit is connected to the edge server via a network interface, and is used to store the collected data, and to classify, store and analyze the data.

[0076] Further, the local modeling module includes: a data classification unit, a space pointing unit and a dependent variable influence unit;

[0077] The data classification unit is used to calculate the dependent variability of each type of data, and the parameters whose dependent variability is not zero are used as dependent parameters, and the parameters whose dependent variability is zero are used as environmental parameters;

[0078] The spatial pointing unit is used to establish a multi-dimensional spatial model in the edge device, the dimension of which is consistent with the number of dependent parameters, and to generate a multi-dimensional pointing vector from the environmental data, the dimension of which is consistent with the number of environmental parameters;

[0079] The dependent variable influencing unit is used to project the multi-dimensional pointing vector on the multi-dimensional space model, obtain the pointing end point coordinates of the projected vector, and send them to the edge server.

[0080] Furthermore, the abnormality identification module includes: an interval identification unit and a parameter troubleshooting unit;

[0081] The interval identification unit is arranged in the edge server and is used to determine whether the end point coordinates transmitted back by the edge device are within the theoretical pointing interval;

[0082] The parameter checking unit is used to calculate the influence ratio of the dependent variable data on each environmental data according to the components of the difference vector between the center point and the end point of the interval in each dependent variable dimension.

[0083] Further, the edge communication module includes: a protocol switching unit and a data rollback unit;

[0084] The protocol switching unit activates a second communication protocol in the edge server, the second communication protocol enabling the exchange of environmental data records in the edge device database;

[0085] The data rollback unit is used to detect the communication link between the edge device and the edge server, and to restore the data status when an error or unexpected situation occurs in the database communication.

[0086] Further, the local adjustment module includes: a central command unit, a mode recording unit and an edge adjustment unit;

[0087] The central command unit is used to calculate the adjustment value of the variable parameter of the device when the end point coordinates do not exceed the interval range, so that the adjusted end point coordinates coincide with the theoretical pointing point, and determine the parameter adjustment weight according to the influence ratio;

[0088] The mode recording unit is used to receive feedback data after parameter adjustment, and when the distance between the end point coordinates in the feedback data and the theoretical pointing point decreases, record the adjustment conditions and adjustment behaviors in the edge device;

[0089] The edge adjustment unit is used to calculate the matching degree between the collected environmental parameters and the adjustment conditions in the edge device, and when it is higher than a threshold, it points to the adjustment behavior corresponding to the adjustment condition.

[0090] like Figure 2 As shown, the edge data collection method of a smart factory based on the Internet of Things includes the following steps:

[0091] Step S1. The edge device sensor collects various types of data at fixed intervals and stores them in the database. The average correlation of the sampling sequence is calculated according to the first method. The data with correlation higher than the threshold is stored as dependent parameters, otherwise it is stored as environmental parameters.

[0092] Step S1 includes:

[0093] Step S11. The edge device is set in the production environment of the factory to collect the working parameters of the production equipment, which include variable parameters and immutable parameters. The variable parameters also include: production speed, production quality, production energy consumption, setting parameters, energy supply parameters and setting parameters of environmental conditioning facilities of the production equipment. The immutable parameters also include: temperature, humidity, sound and brightness.

[0094] Step S12. The edge device partitions and stores the sampled data according to the parameter type, arranges them in chronological order, and forms a sampling sequence. For each parameter, the average correlation of the sampling sequence is calculated according to the first method. The first method is as follows;

[0095] ;

[0096] Where Q represents the average correlation of the sampling sequence, m represents the number of working parameters, n represents the length of the sampling sequence, Wi represents the sampling sequence corresponding to the i-th working parameter, and Wi j represents the jth element in the sequence Wi, represents the mean of the elements in the sequence Wi, represents the jth element in the sampling sequence of the current parameter, Represents the mean of the elements in the sampling sequence of the current parameter;

[0097] Step S13. Preset a threshold Q0. If the average correlation Q>Q0 of the parameter, the parameter is stored as a dependent function. If Q≤Q0, the parameter is stored as an environmental parameter.

[0098] Step S2. Establish a multidimensional space model based on dependent parameters, the dimension of which is consistent with the number of dependent parameters, generate a multidimensional pointing vector based on environmental data, the dimension of which is consistent with the number of environmental parameters, project the multidimensional pointing vector in the multidimensional space model, and obtain the pointing end point coordinates of the projected vector;

[0099] Step S2 includes:

[0100] Step S21. The database of the edge device is connected to the edge server via the edge gateway, and data is transmitted bidirectionally according to the preset communication protocol. The edge server obtains the environmental parameters and dependent parameters stored in the database. The number of environmental parameters is recorded as t1, and the number of dependent parameters is recorded as t2;

[0101] Step S22: Establish a t2-dimensional virtual space in the edge server, and generate actual coordinates W in the virtual space according to the current values ​​of the dependent parameters: (W1, W2, ..., W t2 ), where W1, W2, …, W t2 Represent the 1st, 2nd, …t2th dependent parameter values ​​respectively;

[0102] At the same time, starting from zero, points (H1, H2, ..., H t1 ) generates a t1-dimensional pointing vector for the end point, where H1, H2, …, H t1 Represent the 1st, 2nd, …, t1th environmental parameter values ​​respectively, map the pointing vector in the t2-dimensional virtual space, and obtain the end point coordinate H of the mapping vector: (HE1, HE2, …, HE t2 ), where HE1, HE2, …, HE t2 They represent the mapping values ​​of the projection vector in the 1st, 2nd, …t2 dimensions respectively.

[0103] Step S3. Set an unknown processing matrix between the pointing vector and the pointing end point coordinates, substitute the historical dependent parameters and the pointing end point coordinates into the unknown processing matrix, solve the environmental impact matrix, and the edge server processes the currently collected environmental parameters through the environmental impact matrix to obtain theoretical coordinates;

[0104] Step S3 includes:

[0105] Step S31. Preset the unknown matrix A with t2 rows and t2 columns, and list the state transformation equation:

[0106] ;

[0107] Among them A (1,1) To A (t2,t2) Respectively represent the values ​​of the 1st row and 1st column to the t2th row and t2th column in the unknown matrix A;

[0108] Step S32. Find history t2 in the database 2 Substitute the environmental parameters and dependent parameters obtained in the sampling into the state transformation equation, solve the value of each unknown element in the unknown matrix A, and obtain the environmental impact matrix B;

[0109] Step S33: The edge server obtains the environmental parameters currently collected by the edge device, calculates the end point coordinates according to the method of step S22, and processes the end point coordinates with the environmental impact matrix B to obtain the theoretical coordinates E.

[0110] Step S4. Use the actual coordinates to point to the theoretical coordinates, generate an error vector, calculate the projection length of the error vector in each environmental parameter dimension, and obtain the change in the environmental parameter. When the change in the environmental parameter within a unit period exceeds a threshold, the corresponding sensor is repaired;

[0111] Step S4 includes:

[0112] Step S41. Generate error vector , project the error vector into the t1-dimensional space, and calculate the projection length of the error vector in each dimension, which is recorded as the set R={R1,R2,…,R t1}, where R1, R2, …, R t1 Respectively represent the environmental parameter changes in the 1st, 2nd, …t1th dimensions of the error vector;

[0113] Step S42. Compare the elements in the set R with the maximum environmental change threshold within the unit sampling period in turn. When the threshold is exceeded, mark the environmental parameter corresponding to the element as an abnormal parameter, and repair the sensor that collects the abnormal parameter in the edge device.

[0114] Step S5. Take the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjust the variable parameters in the environmental data, so that the adjusted theoretical coordinates coincide with the optimal endpoint coordinates. If the distance between the actual coordinates and the optimal endpoint coordinates decreases before and after the adjustment, the adjustment process is recorded to the edge device. When the edge device collects the same environmental parameters, it actively adjusts the variable parameters.

[0115] Step S5 includes:

[0116] Step S51. Take the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjust the variable parameters in the environmental data, so that the projection endpoint of the pointing vector of the adjusted environmental parameters in the t2-dimensional virtual space coincides with the optimal endpoint coordinates after being processed by the environmental impact matrix B, and record the adjustment parameters and adjustment amounts of the edge server;

[0117] Step S52. In the next sampling cycle after the variable parameters are adjusted, the actual coordinates corresponding to the sampled data are obtained. If the distance between the actual coordinates and the optimal end point coordinates is less than the distance between the actual coordinates and the optimal end point coordinates in the previous cycle, the adjustment parameters, adjustment amounts and current environmental parameter values ​​of the edge server are sent to the edge device, forming a positive adjustment mode;

[0118] Step S53: When the edge device collects the same environmental parameters as those recorded in the forward adjustment mode, it automatically inspects and adjusts the variable parameters according to the adjustment parameters and adjustment amounts recorded in the forward adjustment mode.

[0119] An edge data acquisition medium for a smart factory based on the Internet of Things, wherein the medium stores computer software, a network transmission protocol, and an agent communication protocol for executing an edge data acquisition method for a smart factory based on the Internet of Things, and is capable of establishing a communication connection between an edge device and an edge server.

[0120] Embodiment: The edge device collects four types of parameters: temperature, humidity, equipment running speed and set speed. Temperature, humidity and set speed are environmental parameters, and set speed is a dependent parameter. At a certain moment, the values ​​of the four types of parameters collected are [20, 0.2, 30, 20], respectively. The theoretical projection length of (20, 0.2, 30) in one-dimensional space is calculated to be 18, which is inconsistent with the actual projection length of 20. The reverse projection is repaired with a length of 20. The parameter change of temperature is 2, which exceeds the preset threshold, and the temperature sensor is repaired.

[0121] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0122] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A smart factory edge data collection method based on the Internet of Things, characterized in that: The method comprises the following steps: Step S1. The edge device sensor collects various types of data at fixed intervals and stores them in the database. The average correlation of the sampling sequence is calculated according to the first method. The data with correlation higher than the threshold is stored as dependent parameters, otherwise it is stored as environmental parameters. Step S2. Establish a multidimensional space model based on dependent parameters, the dimension of which is consistent with the number of dependent parameters, generate a multidimensional pointing vector based on environmental data, the dimension of which is consistent with the number of environmental parameters, project the multidimensional pointing vector in the multidimensional space model, and obtain the pointing end point coordinates of the projected vector; Step S3. Set an unknown processing matrix between the pointing vector and the pointing end point coordinates, substitute the historical dependent parameters and the pointing end point coordinates into the unknown processing matrix, solve the environmental impact matrix, and the edge server processes the currently collected environmental parameters through the environmental impact matrix to obtain theoretical coordinates; Step S4. Use the actual coordinates to point to the theoretical coordinates, generate an error vector, calculate the projection length of the error vector in each environmental parameter dimension, and obtain the change in the environmental parameter. When the change in the environmental parameter within a unit period exceeds a threshold, the corresponding sensor is repaired; Step S5. Take the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjust the variable parameters in the environmental data, and make the adjusted theoretical coordinates coincide with the optimal endpoint coordinates. If the distance between the actual coordinates and the optimal endpoint coordinates decreases before and after the adjustment, the adjustment process is recorded to the edge device. When the edge device collects the same environmental parameters, it actively adjusts the variable parameters. Step S2 includes: Step S21. The database of the edge device is connected to the edge server via the edge gateway, and data is transmitted bidirectionally according to the preset communication protocol. The edge server obtains the environmental parameters and dependent parameters stored in the database. The number of environmental parameters is recorded as t1, and the number of dependent parameters is recorded as t2; Step S22: Establish a t2-dimensional virtual space in the edge server, and generate actual coordinates W in the virtual space according to the current values ​​of the dependent parameters: (W1, W2, ..., W t2 ), where W1, W2, …, W t2 Represent the 1st, 2nd, …t2th dependent parameter values ​​respectively; At the same time, starting from zero, points (H1, H2, ..., H t1 ) generates a t1-dimensional pointing vector for the end point, where H1, H2, …, H t1 Represent the 1st, 2nd, …, t1th environmental parameter values ​​respectively, map the pointing vector in the t2-dimensional virtual space, and obtain the end point coordinate H of the mapping vector: (HE1, HE2, …, HE t2 ), where HE1, HE2, …, HE t2 Represent the mapping values ​​of the projection vector in the 1st, 2nd, …t2 dimensions respectively; Step S4 includes: Step S41. Generate error vector , project the error vector into the t1-dimensional space, and calculate the projection length of the error vector in each dimension, which is recorded as the set R={R1,R2,…,R t1 }, where R1, R2, …, R t1 Respectively represent the environmental parameter changes in the 1st, 2nd, …t1th dimensions of the error vector; Step S42. The elements in the set R are sequentially compared with the maximum environmental change threshold within the unit sampling period. When the threshold is exceeded, the environmental parameter corresponding to the element is marked as an abnormal parameter, and the sensor that collects the abnormal parameter in the edge device is repaired; Step S5 includes: Step S51. Taking the endpoint coordinates obtained by the initial sampling of the edge device as the optimal endpoint coordinates, adjusting the variable parameters in the environmental data, so that the projection endpoint of the pointing vector of the adjusted environmental parameters in the t2-dimensional virtual space coincides with the optimal endpoint coordinates after being processed by the environmental impact matrix B, and recording the adjustment parameters and adjustment amounts of the edge server; Step S52. In the next sampling cycle after the variable parameters are adjusted, the actual coordinates corresponding to the sampled data are obtained. If the distance between the actual coordinates and the optimal end point coordinates is less than the distance between the actual coordinates and the optimal end point coordinates in the previous cycle, the adjustment parameters, adjustment amounts and current environmental parameter values ​​of the edge server are sent to the edge device, forming a positive adjustment mode; Step S53: When the edge device collects the same environmental parameters as those recorded in the forward adjustment mode, it automatically inspects and adjusts the variable parameters according to the adjustment parameters and adjustment amounts recorded in the forward adjustment mode.

2. The method for collecting edge data of an intelligent factory based on the Internet of Things according to claim 1 is characterized in that: Step S1 includes: Step S11. The edge device is set in the production environment of the factory to collect the working parameters of the production equipment, which include variable parameters and immutable parameters. The variable parameters also include: production speed, production quality, production energy consumption, setting parameters, energy supply parameters and setting parameters of environmental conditioning facilities of the production equipment. The immutable parameters also include: temperature, humidity, sound and brightness. Step S12. The edge device partitions and stores the sampled data according to the parameter type, arranges them in chronological order, and forms a sampling sequence. For each parameter, the average correlation of the sampling sequence is calculated according to the first method. The first method is as follows; ； Where Q represents the average correlation of the sampling sequence, m represents the number of working parameters, n represents the length of the sampling sequence, Wi represents the sampling sequence corresponding to the i-th working parameter, and Wi j represents the jth element in the sequence Wi, represents the mean of the elements in the sequence Wi, represents the jth element in the sampling sequence of the current parameter, Represents the mean of the elements in the sampling sequence of the current parameter; Step S13. Preset a threshold Q0. If the average correlation Q>Q0 of the parameter, the parameter is stored as a dependent function. If Q≤Q0, the parameter is stored as an environmental parameter.

3. The method for collecting edge data of an intelligent factory based on the Internet of Things according to claim 2 is characterized in that: Step S3 includes: Step S31. Preset the unknown matrix A with t2 rows and t2 columns, and list the state transformation equation: ； Among them A (1,1) To A (t2,t2) Respectively represent the values ​​of the 1st row and 1st column to the t2th row and t2th column in the unknown matrix A; Step S32. Find history t2 in the database 2 Substitute the environmental parameters and dependent parameters obtained in the sampling into the state transformation equation, solve the value of each unknown element in the unknown matrix A, and obtain the environmental impact matrix B; Step S33: The edge server obtains the environmental parameters currently collected by the edge device, calculates the end point coordinates according to the method of step S22, and processes the end point coordinates with the environmental impact matrix B to obtain the theoretical coordinates E.

4. A smart factory edge data collection system based on the Internet of Things, wherein the system executes the smart factory edge data collection method based on the Internet of Things described in claim 1, characterized in that: It includes the following modules: edge acquisition module, local modeling module, anomaly identification module, edge communication module and local adjustment module; The edge acquisition module is used to set up edge IoT devices at factory equipment, collect the working parameters of the devices through sensors on the edge IoT devices, and the IoT devices record all collected data in the database. The database is connected to the edge server via an edge gateway, and the link can transmit data bidirectionally according to a preset communication protocol; The local modeling module is used to calculate the dependency of various types of data, and divide the collected data of the IoT device into dependent data and environmental data according to the dependency. A multidimensional space model is constructed in the edge device with the dependent data. The environmental data constitutes a pointing vector, and the end point of the pointing vector in the structure is recorded. According to the historical environmental data and the end point coordinates, a multidimensional influence function of the dependent data and the environmental data is calculated for each dependent data, and the influence function is matrixed to obtain an environmental influence matrix; The abnormal identification module is used to transmit environmental data and the end point coordinates of the pointing vector to the edge server in real time through the transmission interface in the edge device. The edge server calculates the theoretical pointing interval through the environmental impact matrix. When the end point coordinate exceeds the interval range, the impact ratio is calculated according to the components of the difference vector between the center point of the interval and the end point in each dependent variable dimension, the second communication protocol is activated, the probability distribution function of the environmental data is obtained, the sensor with a probability less than the threshold is replaced, and the acquired environmental data is corrected according to the impact ratio; The edge communication module is used to build a communication link between the edge device and the edge server, and provides protocol conversion and communication proxy paths. The supported conversion protocols include: Modbus protocol, OPC protocol, Profibus protocol, MQTT protocol and EtherNet / IP protocol. It also provides a data rollback function to restore the data status when an error or unexpected situation occurs in the database communication; The local adjustment module is used to adjust the variable parameters in the environmental data in the edge server so that the theoretical pointing point coincides with the initial pointing point, and to recover the coordinates of the end point of the pointing vector before and after the adjustment. When the coordinates of the end point after adjustment are closer to the center point of the theoretical pointing interval than before the adjustment, the adjustment behavior is recorded in the edge device as a positive adjustment mode. The edge device calculates the matching degree of the collected data and each positive adjustment mode in real time. When it is higher than the threshold, the device parameters are adjusted according to the positive adjustment mode.

5. The IoT-based smart factory edge data acquisition system according to claim 4 is characterized by: The edge acquisition module includes: an IoT device unit and a database unit; The IoT device unit is used to collect operating parameters of factory production equipment, environmental parameters in the production environment, and setting parameters of production equipment; The database unit is connected to the edge server via a network interface, and is used to store the collected data, and classify, store and analyze the data; The local modeling module includes: a data classification unit, a space pointing unit and a dependent variable influence unit; The data classification unit is used to calculate the dependent variability of each type of data, and the parameters whose dependent variability is not zero are used as dependent parameters, and the parameters whose dependent variability is zero are used as environmental parameters; The spatial pointing unit is used to establish a multi-dimensional spatial model in the edge device, the dimension of which is consistent with the number of dependent parameters, and to generate a multi-dimensional pointing vector from the environmental data, the dimension of which is consistent with the number of environmental parameters; The dependent variable influencing unit is used to project the multi-dimensional pointing vector on the multi-dimensional space model, obtain the pointing end point coordinates of the projected vector, and send them to the edge server.

6. The IoT-based smart factory edge data acquisition system according to claim 5 is characterized by: The abnormality identification module includes: an interval identification unit and a parameter troubleshooting unit; The interval identification unit is arranged in the edge server and is used to determine whether the end point coordinates transmitted back by the edge device are within the theoretical pointing interval; The parameter checking unit is used to calculate the influence ratio of the dependent variable data on each environmental data according to the components of the difference vector between the center point and the end point of the interval in each dependent variable dimension; The edge communication module includes: a protocol switching unit and a data rollback unit; The protocol switching unit activates a second communication protocol in the edge server, the second communication protocol enabling the exchange of environmental data records in the edge device database; The data rollback unit is used to detect the communication link between the edge device and the edge server, and to restore the data status when an error or unexpected situation occurs in the database communication.

7. The IoT-based smart factory edge data acquisition system according to claim 6 is characterized by: The local adjustment module includes: a central command unit, a mode recording unit and an edge adjustment unit; The central command unit is used to calculate the adjustment value of the variable parameter of the device when the end point coordinates do not exceed the interval range, so that the adjusted end point coordinates coincide with the theoretical pointing point, and determine the parameter adjustment weight according to the influence ratio; The mode recording unit is used to receive feedback data after parameter adjustment, and when the distance between the end point coordinates in the feedback data and the theoretical pointing point decreases, record the adjustment conditions and adjustment behaviors in the edge device; The edge adjustment unit is used to calculate the matching degree between the collected environmental parameters and the adjustment conditions in the edge device, and when it is higher than a threshold, it points to the adjustment behavior corresponding to the adjustment condition.

8. A computer storage medium, wherein the medium stores computer software, a network transmission protocol, and an agent communication protocol, wherein the computer software, the network transmission protocol, and the agent communication protocol are used to execute the smart factory edge data collection method based on the Internet of Things described in claim 1.

Images