Coating workshop data acquisition and analysis processing method, device, equipment and medium

Abstract

This invention discloses a method, apparatus, equipment, and medium for data acquisition and analysis in a painting workshop. The method includes: using object tools in a scripting language to transfer data from the painting workshop's systems to corresponding target tables in a preset target database via the workshop's network; analyzing and processing the data in each target table according to its characteristics and analysis requirements to obtain analysis results; and displaying or pushing the analysis results. This technical solution, by introducing object tools in a scripting language, achieves automatic data acquisition in the painting workshop and, through reasonable analysis and processing, realizes data visualization and informatization, improving the automation level of the painting workshop. This, in turn, allows maintenance personnel to utilize information technology to reduce troubleshooting and repair time and increase the mean time between failures (MTBF) of the system in the painting workshop.

Description

Technical Field

[0001] This invention relates to the technical field of vehicle body painting, specifically to a method, apparatus, equipment, and medium for data acquisition and analysis processing in a painting workshop. Background Technology

[0002] Traditional painting workshops typically feature a wide variety of equipment and relatively complex processes, resulting in long intervals between failures, unpredictable repair times, and low production efficiency. Consequently, painting workshops often contain numerous systems, each responsible for production management, vehicle body monitoring and tracking, PLC control, equipment status monitoring, fault alarms, parameter settings, and displays. This makes data acquisition and integration challenging.

[0003] Currently, the painting workshop lacks a unified data management platform. Data can only be manually retrieved from various systems such as PLCs, making real-time, automated production data collection impossible, data recording unavailable, and data analysis and processing impossible. Automating data collection and processing would require additional investment in software or hardware, with high purchase and maintenance costs, significantly impacting the overall efficiency of the painting workshop. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed to provide a method, apparatus, electronic device and storage medium for data acquisition and analysis processing in a painting workshop that overcomes or at least partially solves the above problems.

[0005] According to one aspect of the present invention, a method for data acquisition and analysis processing in a painting workshop is provided, the method comprising:

[0006] Using object tools in scripting languages, data from the painting workshop's systems is transferred to the corresponding target tables in a pre-defined target database via the network within the painting workshop.

[0007] Based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis and processing results;

[0008] Display or push the analysis and processing results.

[0009] In some implementations, data from the painting workshop's systems is transferred to corresponding target tables in a preset target database using object tools within a scripting language and the network within the painting workshop. This includes:

[0010] A connection object created using a scripting language is connected to the data source to which the data to be transferred belongs;

[0011] The connection to the data source is opened by the open function;

[0012] Perform retrieval operations in the Structured Query Language and save the retrieved data to be transferred to the record set using the created record set object;

[0013] The data to be transferred from the record set is inserted into the target table of the target database.

[0014] In some implementations, data from the painting workshop's systems is transferred to corresponding target tables in a preset target database using object tools within a scripting language and the network within the painting workshop. This further includes:

[0015] Close the recordset object and the connection object.

[0016] In some implementations, the data source is the database of the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. Then, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including:

[0017] Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value.

[0018] A dataset is constructed based on the feature values. The dataset also includes target values, which are either abnormal or normal. The number of datasets corresponds to the number of data segments.

[0019] Based on the number of feature values ​​in the dataset, the number of neurons in the input layer of the BP neural network is determined, the number of hidden layers of the BP neural network is determined according to empirical formulas, and the sigmoid function, tanh function, or logistic function is selected as the activation function of the hidden layer and the output layer, thereby constructing the BP neural network discriminative model.

[0020] The BP neural network discrimination model is trained using the dataset.

[0021] The dataset obtained from the latest data segment is input into the trained BP neural network discrimination model to determine whether the current value in the latest data segment is a normal current value, thereby determining the motor health status of the PLC system.

[0022] In some embodiments, before training the BP neural network discriminative model using the datasets, the method further includes: normalizing the feature values ​​in each of the datasets; and / or,

[0023] Training the BP neural network discriminant model using the dataset includes optimizing the weight parameters of the BP neural network discriminant model using grid search and / or cross-validation.

[0024] In some implementations, the data source is the database of the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. Then, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including:

[0025] Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value.

[0026] A dataset is constructed based on the feature values, and the dataset also includes a target value, which is either abnormal or normal. The number of datasets corresponds to the number of data segments.

[0027] A single-layer neural network is constructed, which is configured to: initialize a weight for each feature value, calculate the sum of the products of each feature value and the corresponding weight value, and then determine whether the current value corresponding to the data segment is normal based on the sum of the products using the sigmoid function or the threshold method.

[0028] The single-layer neural network is trained using the dataset to determine the weight values ​​corresponding to each feature value.

[0029] The dataset obtained from the latest data segment is input into a trained single-layer neural network to determine whether the current value in the latest data segment is normal.

[0030] In some implementations, the data source is a vehicle position coordinate storage database of the vehicle position monitoring system, the data to be transferred is vehicle position coordinate data, and the target table is a vehicle position coordinate data table. Then, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including:

[0031] Calculate the deviation between the vehicle position coordinates in the vehicle position coordinate data table and the standard vehicle position coordinates;

[0032] Determine whether the deviation is greater than a preset deviation threshold;

[0033] If the deviation is greater than or equal to the stated deviation threshold, an alarm will be issued to remind maintenance personnel to correct the position of the conveyor chain.

[0034] According to another aspect of the present invention, a data acquisition and analysis processing device for a painting workshop is provided, the device comprising:

[0035] The data acquisition module is suitable for transferring data from the painting workshop system to the corresponding target table in the preset target database using object tools in the scripting language and the network within the painting workshop.

[0036] The data processing module is suitable for analyzing and processing the data in each target table according to the data characteristics and analysis requirements of each target table, so as to obtain the analysis and processing results;

[0037] The results display module is suitable for displaying or pushing the analysis and processing results.

[0038] According to another aspect of the present invention, an electronic device is provided, comprising: a processor and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform a painting workshop data acquisition and analysis processing method according to any one of the above embodiments.

[0039] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores one or more programs, which, when executed by a processor, implement the painting workshop data acquisition and analysis processing method according to any one of the above.

[0040] In summary, the data acquisition and analysis method for painting workshops disclosed in this invention, by introducing object tools in scripting languages, realizes the automatic acquisition of data in the painting workshop, and through reasonable analysis and processing, realizes the visualization and informatization of data, improves the automation level of the painting workshop, and facilitates maintenance personnel to use information technology to reduce the time for troubleshooting and repair, and increases the mean time between failures (MTBF) of the system in the painting workshop.

[0041] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0042] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0043] Figure 1A flowchart illustrating a method for data acquisition and analysis in a painting workshop according to some embodiments of the present invention is shown;

[0044] Figure 2 A schematic diagram of the structure of a BP neural network according to some embodiments of the present invention is shown;

[0045] Figure 3 A schematic diagram of a single-layer neural network according to some embodiments of the present invention is shown;

[0046] Figure 4 A schematic diagram of the structure of a single-layer neural network according to other embodiments of the present invention is shown;

[0047] Figure 5 A schematic diagram of a data acquisition and analysis processing device for a painting workshop according to some embodiments of the present invention is shown;

[0048] Figure 6 A schematic diagram of the structure of an electronic device according to some embodiments of the present invention is shown. Detailed Implementation

[0049] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0050] Figure 1 A flowchart illustrating a data acquisition and analysis method for a painting workshop according to an embodiment of the present invention is shown. The method includes the following steps:

[0051] Step S110: Using the object tool in the scripting language, the data in the painting workshop system is transferred to the corresponding target table in the preset target database via the network in the painting workshop.

[0052] Scripting languages ​​are programming languages ​​used to write scripts, and are commonly used in practice for tasks such as automation, rapid prototyping, and simplifying complex operations. Common scripting languages ​​include Python, JavaScript, Shell, PowerShell, PHP, VBScript, AppleScript, UNIX's csh and tcsh, and Windows systems' Winbatch and 4NT. Object-oriented tools primarily include database connection tools, used to connect to databases or data sources in different systems within the painting workshop and centralize data into a target database, thus forming a unified database platform.

[0053] The target database mentioned above can be a relational database, such as MySQL, Oracle Database, Microsoft SQL Server, SQL Server DB, etc., and corresponding target tables should be created in the target database to receive data from each system within the painting workshop. The attributes of each target table are determined based on the data received from the system: field names, data types, constraints, default values, indexes, etc., and these attributes are preset before receiving the data.

[0054] The systems in the painting workshop include, but are not limited to, production management systems, industrial automation software systems such as WinCC, PLC control systems including vehicle body monitoring and tracking, motor control, and may also include equipment status monitoring, image systems, control of various equipment and fault alarm systems, data connection and display systems such as Power BI, etc.

[0055] Step S120: Based on the data characteristics and analysis requirements of each target table, analyze and process the data in each target table to obtain the analysis and processing results.

[0056] Data analysis and processing involves data mining and utilization. It involves processing, organizing, and analyzing collected data to transform it into information, which is then used in the operation of the painting workshop to improve its efficiency.

[0057] In this embodiment, data analysis tools include, but are not limited to, Pareto charts, cause-effect diagrams, stratification, questionnaires, scatter plots, histograms, control charts, correlation diagrams, matrix data diagrams, etc.

[0058] The data processing in this embodiment includes using machine learning algorithms to predict future trends based on existing datasets, or to classify data types, including identifying and classifying faults within the workshop. The aforementioned machine learning algorithms include, but are not limited to, decision trees, random forests, logistic regression, SVM, Naive Bayes, K-nearest neighbors, Markov models, and neural network algorithms. Neural network algorithms, in particular, are currently a hot topic and can be well applied in data processing within the painting workshop.

[0059] Step S130: Display or push the analysis and processing results.

[0060] In this step S130, existing display tools such as Power BI and WinCC can be used to display the image or chart results, or custom charts can be created for display. The processing results can also be pushed to relevant personnel in a timely manner so that they can take appropriate countermeasures.

[0061] In summary, this embodiment achieves automatic data collection in the painting workshop by introducing object tools in a scripting language, and realizes data visualization and informatization through reasonable analysis and processing, thereby improving the automation level and overall efficiency of the painting workshop.

[0062] In some embodiments, step S110, using an object tool in a scripting language to transfer data from the painting workshop's system to the corresponding target table in a preset target database, includes:

[0063] A connection object created using a scripting language is connected to the data source to which the data to be transferred belongs;

[0064] The connection to the data source is opened by the open function;

[0065] Perform retrieval operations in the Structured Query Language and save the retrieved data to be transferred to the record set using the created record set object;

[0066] The data to be transferred from the record set is inserted into the target table of the target database.

[0067] In one optional implementation, the VBScript scripting language is used to connect to the databases of various systems within the workshop. Specifically, two object tools from the ADODB component in the VBScript scripting language can be used: the database connection object ADODB.Connection and the database storage recordset object (ADODB.Recordset).

[0068] In practical applications, firstly, a CONN connection object can be created in advance:

[0069] Set Conn=CreateObject("ADODB.Connection")

[0070] You can also create RS storage recordset objects at the same time:

[0071] Set rsCurrents=CreateObject("ADODB.Recordset")

[0072] Source = "select *from XXXX"

[0073] Connect="Provider="sqloledb;Data Source="Data Source;"&_

[0074] "Initial Catalog = Initial Catalog; Integrated Security = SSPI; uid = Database Name; pwd = Password"

[0075] Then, open the database connection object mentioned above:

[0076] Conn.Open Connect

[0077] Set rsCurrents=Conn.Execute(Source)

[0078] Do Until rsCurrents.EOF

[0079] `msgbox rsCurrents("data")` 'Retrieves the values ​​of the field "data" and saves them to the recordset.'

[0080] rsCurrent.MoveNext

[0081] loop

[0082] Then you can also perform the action of inserting into the target table;

[0083] Finally, close the connection:

[0084] rsCurrents.Close

[0085] conn.Close

[0086] And clean up the objects:

[0087] Set rsCurrents = Nothing

[0088] Set conn = Nothing

[0089] Taking the acquisition of motor current data within a PLC system as an example, the process of acquiring PLC system data in real time and storing it in a standard format in an MS SQL database includes the following steps:

[0090] Step 1: In the MS SQL database, create a table and set the variable names and variable types for the collected data in the table;

[0091] Step 2: Use the VBScript language to read the current value in the PLC system, which includes the following steps;

[0092] Step 3: An ADODB.Connection object was created to connect to the PLC system's database;

[0093] Step 4: Then open the database connection using the "Open" method, where you need to fill in the correct data source name (DSN), username (UID), and password (PWD);

[0094] Step 5: Next, an ADODB.Recordset object is created to execute the SQL insert statement, inserting the read PLC variable value "value" into the specified MS SQL target table ("table name") and column ("column name");

[0095] Step 6: Finally, close the recordset and database connection;

[0096] Step 7: Finally, refresh the target table in the MS SQL database to see the real-time data row by row.

[0097] Therefore, according to the specific implementation described above, step S110 may further include the following steps:

[0098] Close the recordset object and the connection object.

[0099] Currently, using neural network models and large amounts of data to predict future trends or monitor and determine the operating status of equipment is a research hotspot. The embodiments of this invention use neural network models to analyze and process data from systems within the painting workshop, predict or monitor the system's status, so as to provide useful reference information or issue maintenance alarms for managers or maintenance personnel.

[0100] Specifically, in some embodiments, the data source is the database of the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. Then, step S120 analyzes and processes the data in each target table according to its characteristics and analysis requirements to obtain the analysis results, including:

[0101] Obtain historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value u. x The eigenvalues ​​are: standard deviation, root mean square, peak indicator, skewness, variance, pulse index, margin index, and peak-to-peak value. The formulas for calculating each eigenvalue are shown in Table 1.

[0102] Table 1

[0103]

[0104] A dataset is constructed based on the feature values. The dataset also includes target values, which are either abnormal or normal. For example, the target value for normal current data can be 1, and the target value for abnormal data can be 0. The number of datasets corresponds to the number of data segments.

[0105] In this embodiment, the neural network can be a BP neural network. First, the number of neurons in the input layer of the BP neural network is determined according to the number of feature values ​​in the dataset. Then, the number of hidden layers of the BP neural network is determined according to empirical formulas. The sigmoid function, tanh function, or logistic function is selected as the activation function of the hidden layer and the output layer, thereby constructing the BP neural network discriminant model of this embodiment.

[0106] It should be noted that the BP (Back-Propagation Network) neural network, also known as a reverse propagation neural network, is a supervised learning algorithm with strong adaptive, self-learning, and nonlinear mapping capabilities. It effectively addresses problems of limited data, scarce information, and uncertainty, and is not limited by nonlinear models. A typical BP network should consist of three layers: an input layer, hidden layers, and an output layer. All layers are fully connected, but there are no connections between layers within the same layer. There can be multiple hidden layers; this embodiment uses a single hidden layer. Figure 2 The BP neural network structure of this embodiment is shown.

[0107] The learning process of a BP neural network mainly consists of two key stages: forward propagation and backward propagation. During forward propagation, the input signal starts from the input layer, passes through each hidden layer, and finally reaches the output layer. Once the signal reaches the output layer, the actual response value is obtained. If there is a significant error between the actual value and the expected value, the learning process immediately enters the error backpropagation stage. The backpropagation stage is carried out using gradient descent. Specifically, it starts from the output layer and proceeds backward through each hidden layer, continuously adjusting the connection weights and thresholds of each neuron layer by layer. This iterative process is repeated until the error generated by the network output is reduced to an acceptable level, or until a pre-set number of learning iterations has been completed.

[0108] The input layer plays a crucial role in the entire neural network structure by receiving external input information. It consists of multiple neurons, each responsible for receiving specific input information and then transmitting this information to the neurons in the hidden layers. The number of neurons in the input layer is determined by the number of input variables; that is, the input layer will have the same number of neurons as there are input variables. For a task that determines the health status of a motor based on its current value, where each data sample includes a maximum of nine feature values ​​and one target value, the maximum number of neurons in the input layer can be nine.

[0109] The number of hidden layer neurons in a network is directly related to several factors, including the complexity of the actual problem, the number of neurons in the input and output layers, and the setting of the expected error. In this embodiment of the invention, the number of hidden layer neurons L is determined according to the following empirical formula:

[0110] L = n + m + a

[0111]

[0112] Where n represents the number of neurons in the input layer, m represents the number of neurons in the output layer, and a is a constant ranging from 1 to 10.

[0113] The main function of the output layer is to output the final result after the information has been processed by the entire network. Similarly, it is also composed of several neurons, and the number of nodes depends on the number of output variables. That is, the number of neurons in the output layer is determined based on the number of different variables that need to be output; in this embodiment, the number of nodes in the output layer can be selected as 1.

[0114] Then, the BP neural network discriminant model is trained using the dataset. During training, the dataset can be divided into a training set, a validation set, and a test set, and the proportions of the training set, validation set, and test set can be 70%, 15%, 15% or 60%, 20%, 20%, and finally, a trained BP neural network discriminant model is obtained.

[0115] In practical applications, the dataset sample obtained from the latest data segment is input into the trained BP neural network discrimination model to determine whether the current value in the latest data segment is a normal current value. That is, if the model output is 1, it means that the motor is normal, and if the output is 0, it means that it is abnormal, thereby determining the motor health status of the PLC system.

[0116] To improve the accuracy of identification and judgment, in some optional embodiments, before training the BP neural network discriminative model using the dataset, the following operation is further included: normalizing the feature values ​​in each dataset. This is because if the original data is used directly for model training, features with larger value ranges may dominate during the model training process, resulting in insufficient learning of features with smaller value ranges, affecting the model's accuracy and generalization ability. Through data normalization, the value ranges of all features can be adjusted to a relatively consistent interval, allowing the model to learn the information of each feature more evenly, thereby improving the model training effect.

[0117] Furthermore, training the BP neural network discriminant model using the dataset includes: further optimizing the weight parameters of the BP neural network discriminant model using grid search or cross-validation.

[0118] Grid search is a hyperparameter optimization technique. Hyperparameters are parameters that need to be manually set before model training, such as the learning rate, number of hidden layer neurons, number of training epochs, and batch size in a neural network. The values ​​of these parameters have a significant impact on the model's performance. The basic idea of ​​grid search is exhaustive search; it iterates through all possible combinations of hyperparameters within a specified range of values. For example, in the previous code, the learning rate has 3 possible values ​​([0.001, 0.01, 0.1]), the number of training epochs has 2 possible values ​​([50, 100]), and the batch size has 2 possible values ​​([10, 20]). Therefore, grid search will try a total of 3 × 2 × 2 = 12 different hyperparameter combinations.

[0119] The workflow of grid search includes: 1) For each combination of hyperparameters, the model is trained using the same training dataset. Taking a neural network as an example, it trains the model based on parameters such as the learning rate, number of training epochs, and batch size in the current combination; 2) After training, the model's performance on the validation dataset is evaluated using a specified evaluation metric (such as the negative of the mean squared error); 3) The performance evaluation results of the model under each combination of hyperparameters are recorded.

[0120] Additionally, cross-validation is a statistical method used to evaluate model performance, primarily aimed at more accurately estimating model performance on unseen data and avoiding overfitting of the training data. It divides a given dataset into multiple subsets; a common method is K-fold cross-validation, where K represents the number of folds. For example, in 3-fold cross-validation, the dataset is evenly divided into three subsets.

[0121] Workflow (using K-fold cross-validation as an example): For each round of cross-validation: Divide the dataset into K subsets, where K-1 subsets are used as training subsets and the remaining subset is used as validation subset. For example, in the first round of 3-fold cross-validation, the first subset is used as the validation set, and the second and third subsets are used as the training set. Train the model using the training subset, and then evaluate the model's performance on the validation subset, recording the values ​​of evaluation metrics (such as accuracy, mean squared error, etc.). Repeat this process K times, selecting a different subset as the validation set each time. In this way, each subset has the opportunity to be used as the validation set once. Finally, combine the results of the K evaluation metrics, for example, by averaging, to obtain a more stable evaluation result that better represents the true performance of the model.

[0122] In some embodiments, the data source is a database belonging to the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. Then, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including:

[0123] Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value.

[0124] A dataset is constructed based on the feature values, and the dataset also includes a target value, which is either abnormal or normal. The number of datasets corresponds to the number of data segments.

[0125] A single-layer neural network is constructed, which is configured to: initialize a weight for each feature value, calculate the sum of the products of each feature value and the corresponding weight value, and then determine whether the current value corresponding to the data segment is normal based on the sum of the products using the sigmoid function or the threshold method.

[0126] The single-layer neural network is trained using the dataset to determine the weight values ​​corresponding to each feature value.

[0127] The dataset obtained from the latest data segment is input into a trained single-layer neural network to determine whether the current value in the latest data segment is normal.

[0128] According to this embodiment, a single-layer neural network structure was used for training, and the judgment of the motor's health status was ultimately achieved. Figure 3 As shown, a single-layer neural network refers to a network that includes only one computational layer, and its computational formula is expressed as follows:

[0129]

[0130] Where w is the weight vector, x is the input, and b is the bias, which is a constant independent of any input value. The above f(x) judgment is either positive or negative, corresponding to the binary classification problem of judging whether the current value is normal in this embodiment.

[0131] In an alternative implementation, a simplified single-layer neural network model is used, see [link to implementation details]. Figure 4 The network structure shown can determine the health status of the motor by setting a threshold value to determine whether the current value is normal.

[0132] In some embodiments, the data source is a vehicle position coordinate storage database of a vehicle position monitoring system, the data to be transferred is vehicle position coordinate data, and the target table is a vehicle position coordinate data table. Then, step S120 analyzes and processes the data in each target table according to its data characteristics and analysis requirements to obtain analysis results, including:

[0133] Calculate the deviation between the vehicle position coordinates in the vehicle position coordinate data table and the standard vehicle position coordinates;

[0134] Determine whether the deviation is greater than a preset deviation threshold;

[0135] If the deviation is greater than or equal to the stated deviation threshold, an alarm will be issued to remind maintenance personnel to correct the position of the conveyor chain.

[0136] Specifically, the problem addressed in this embodiment is that the offset of the conveyor chain transporting the car body needs to be measured periodically in the painting workshop using a tensioner and a measuring tape, but this measurement method is neither accurate nor very labor-intensive.

[0137] Research has found that by taking pictures with a 3D camera used for vehicle positioning, the coordinates of the vehicle's stopping position can be obtained. These coordinates can be used to determine whether the offset of the aforementioned conveyor chain is within a reasonable range.

[0138] However, each 3D camera image generates corresponding coordinate values, resulting in a large data file every day. If manual data collection is used, operators would have to manually export the data file and then merge the data from the file into an Excel file, which wastes a significant amount of their time.

[0139] By adopting the method of this embodiment and combining it with the automatic data acquisition step, not only can coordinate data be automatically obtained, but the deviation between the obtained vehicle body coordinate value and the standard vehicle body coordinate value can also be further judged according to the steps disclosed in this embodiment. When the deviation is greater than the threshold, an alarm is issued to the operator, so that the operator can adjust the conveyor chain in time, thereby improving the efficiency of the painting workshop.

[0140] Combination Figure 5 As shown, another embodiment of the present invention also provides a data acquisition and analysis processing device 500 for a painting workshop, the data acquisition and analysis processing device 500 comprising:

[0141] The data acquisition module 510 is suitable for transferring data from the system within the painting workshop to the corresponding target table in the preset target database using the object tools in the scripting language and the network within the painting workshop.

[0142] The data processing module 520 is adapted to analyze and process the data in each target table according to the data characteristics and analysis requirements of each target table, so as to obtain the analysis and processing results;

[0143] The results display module 530 is suitable for displaying or pushing the analysis and processing results.

[0144] According to the data acquisition and analysis processing device for the painting workshop disclosed in this embodiment, by introducing object tools in a scripting language, the automatic acquisition of data in the painting workshop is realized, and through reasonable analysis and processing, the data visualization and informatization are realized, thereby improving the automation level and efficiency of the painting workshop.

[0145] In some embodiments, the data acquisition module 510 is adapted to:

[0146] A connection object created using a scripting language is connected to the data source to which the data to be transferred belongs;

[0147] The connection to the data source is opened by the open function;

[0148] Perform retrieval operations in the Structured Query Language and save the retrieved data to be transferred to the record set using the created record set object;

[0149] The data to be transferred from the record set is inserted into the target table of the target database.

[0150] In some embodiments, the data acquisition module 510 is further adapted to:

[0151] Close the recordset object and the connection object.

[0152] In some embodiments, the data source is a database belonging to the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. In this case, the data processing module 520 is adapted to:

[0153] Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value.

[0154] A dataset is constructed based on the feature values. The dataset also includes target values, which are either abnormal or normal. The number of datasets corresponds to the number of data segments.

[0155] Based on the number of feature values ​​in the dataset, the number of neurons in the input layer of the BP neural network is determined, the number of hidden layers of the BP neural network is determined according to empirical formulas, and the sigmoid function, tanh function, or logistic function is selected as the activation function of the hidden layer and the output layer, thereby constructing the BP neural network discriminative model.

[0156] The BP neural network discrimination model is trained using the dataset.

[0157] The dataset obtained from the latest data segment is input into the trained BP neural network discrimination model to determine whether the current value in the latest data segment is a normal current value, thereby determining the motor health status of the PLC system.

[0158] In some embodiments, before training the BP neural network discriminant model using the datasets, the method further includes: normalizing the feature values ​​in each dataset; and / or,

[0159] Training the BP neural network discriminant model using the dataset includes optimizing the weight parameters of the BP neural network discriminant model using grid search and / or cross-validation.

[0160] In some embodiments, the data source is a database belonging to the PLC system, the data to be transferred is motor operating status data, and the target table is a motor operating status data table. In this case, the data processing module 520 is adapted to:

[0161] Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value.

[0162] A dataset is constructed based on the feature values, and the dataset also includes a target value, which is either abnormal or normal. The number of datasets corresponds to the number of data segments.

[0163] A single-layer neural network is constructed, which is configured to: initialize a weight for each feature value, calculate the sum of the products of each feature value and the corresponding weight value, and then determine whether the current value corresponding to the data segment is normal based on the sum of the products using the sigmoid function or the threshold method.

[0164] The single-layer neural network is trained using the dataset to determine the weight values ​​corresponding to each feature value.

[0165] The dataset obtained from the latest data segment is input into a trained single-layer neural network to determine whether the current value in the latest data segment is normal.

[0166] In some embodiments, the data source is a vehicle position coordinate storage database of a vehicle position monitoring system, the data to be transferred is vehicle position coordinate data, and the target table is a vehicle position coordinate data table. In this case, the data processing module 520 is further adapted to:

[0167] Calculate the deviation between the vehicle position coordinates in the vehicle position coordinate data table and the standard vehicle position coordinates;

[0168] Determine whether the deviation is greater than a preset deviation threshold;

[0169] If the deviation is greater than or equal to the stated deviation threshold, an alarm will be issued to remind maintenance personnel to correct the position of the conveyor chain.

[0170] It should be noted that the specific implementation methods of the above-mentioned device embodiments can be referred to the specific implementation methods of the corresponding methods described above, and will not be repeated here.

[0171] The protection points and advantages of the embodiments of the present invention include:

[0172] First, by introducing object tools in scripting languages ​​such as VBS, this embodiment of the invention enables automatic data collection in the painting workshop, which not only improves the efficiency of data collection but also frees up labor and reduces the tedious work of operators.

[0173] Furthermore, this embodiment of the invention also employs a neural network model to identify the health status of motors in the PLC system. Specifically, it identifies and judges whether the motor is abnormal by automatically collecting the motor current value, thereby improving the automation and intelligence of identification and judgment.

[0174] Furthermore, this embodiment of the invention also obtains the coordinate values ​​of the vehicle body by acquiring data from the vehicle body 3D camera, and further analyzes the magnitude of the deviation of the vehicle body coordinates to determine whether the conveyor chain needs to be adjusted, thereby improving the automation and intelligence of the conveyor chain adjustment judgment.

[0175] Furthermore, by establishing an integrated and unified database platform, this invention enables more artificial intelligence analysis and processing, which not only realizes data visualization and informatization and improves the automation level of the painting workshop, but also makes it easier for maintenance personnel to use information technology to reduce troubleshooting and repair time, thereby improving the mean time between failures and overall efficiency of the system in the painting workshop.

[0176] It should be noted that:

[0177] The algorithms and displays provided herein are not inherently related to any particular computer, virtual device, or other equipment. Various general-purpose devices can also be used in conjunction with the teachings herein. The required structure for constructing such devices is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0178] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0179] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim.

[0180] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0181] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0182] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the painting workshop data acquisition and analysis processing apparatus according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0183] This invention provides a non-volatile computer storage medium storing at least one executable instruction that can execute the above-described painting workshop data acquisition and analysis processing method in any of the above method embodiments.

[0184] Figure 6 The diagram shows a structural schematic of an embodiment of the electronic device of the present invention. The specific embodiments of the present invention do not limit the specific structure of the electronic device.

[0185] like Figure 6As shown, the electronic device may include: a processor 602, a communications interface 604, a memory 606, and a communications bus 608.

[0186] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608. Communication interface 604 is used to communicate with other network elements, such as user terminals or other servers. Processor 602 executes program 610, specifically performing the relevant steps in the above-described embodiment of the data acquisition and analysis processing method for painting workshops used in electronic devices.

[0187] Specifically, program 610 may include program code that includes computer operation instructions.

[0188] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The electronic device may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0189] Memory 606 is used to store program 610. Memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0190] Specifically, program 610 can be used to enable processor 602 to perform the operations corresponding to the above-described embodiments of the painting workshop data acquisition and analysis processing method.

[0191] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The term "comprising" does not exclude the presence of elements or steps not listed in the claims. The term "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the terms first, second, and third, etc., does not indicate any order. These terms can be interpreted as names.

Claims

1. A method for data acquisition and analysis in a painting workshop, characterized in that, The method includes: Using object tools in scripting languages, data from the painting workshop's systems is transferred to the corresponding target tables in a pre-defined target database via the network within the painting workshop. Based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis and processing results; Display or push the analysis and processing results.

2. The method according to claim 1, characterized in that, Using object tools in a scripting language, data from the painting workshop's systems is transferred to the corresponding target tables in a pre-defined target database via the workshop's network, including: A connection object created using a scripting language is connected to the data source to which the data to be transferred belongs; The connection to the data source is opened by the open function; Perform retrieval operations in the Structured Query Language and save the retrieved data to be transferred to the record set using the created record set object; The data to be transferred from the record set is inserted into the target table of the target database.

3. The method according to claim 2, characterized in that, Using object tools in a scripting language, data from the painting workshop's systems is transferred to corresponding target tables in a pre-defined target database via the workshop's network. This further includes: Close the recordset object and the connection object.

4. The method according to any one of claims 1-3, characterized in that, The data source is the database of the PLC system, the data to be transferred is motor operating status data, and the target table is the motor operating status data table. Therefore, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including: Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value. A dataset is constructed based on the feature values. The dataset also includes target values, which are either abnormal or normal. The number of datasets corresponds to the number of data segments. Based on the number of feature values ​​in the dataset, the number of neurons in the input layer of the BP neural network is determined, the number of hidden layers of the BP neural network is determined according to empirical formulas, and the sigmoid function, tanh function, or logistic function is selected as the activation function of the hidden layer and the output layer, thereby constructing the BP neural network discriminative model. The BP neural network discrimination model is trained using the dataset. The dataset obtained from the latest data segment is input into the trained BP neural network discrimination model to determine whether the current value in the latest data segment is a normal current value, thereby determining the motor health status of the PLC system.

5. The method according to claim 4, characterized in that, Before training the BP neural network discriminant model using the datasets, the method further includes: normalizing the feature values ​​in each dataset; and / or, Training the BP neural network discriminant model using the dataset includes optimizing the weight parameters of the BP neural network discriminant model using grid search and / or cross-validation.

6. The method according to any one of claims 1-3, characterized in that, The data source is the database of the PLC system, the data to be transferred is motor operating status data, and the target table is the motor operating status data table. Therefore, based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including: Obtain the historical current values ​​from the motor operating status data table, divide the historical current values ​​into multiple data segments according to a preset number, and calculate at least one of the following characteristic values ​​for each data segment: average value, standard deviation, root mean square value, peak index, skewness, variance, ripple index, margin index, or interpeak value. A dataset is constructed based on the feature values, and the dataset also includes a target value, which is either abnormal or normal. The number of datasets corresponds to the number of data segments. A single-layer neural network is constructed, which is configured to: initialize a weight for each feature value, calculate the sum of the products of each feature value and the corresponding weight value, and then determine whether the current value corresponding to the data segment is normal based on the sum of the products using the sigmoid function or the threshold method. The single-layer neural network is trained using the dataset to determine the weight values ​​corresponding to each feature value. The dataset obtained from the latest data segment is input into a trained single-layer neural network to determine whether the current value in the latest data segment is normal.

7. The method according to any one of claims 1-3, characterized in that, The data source is the vehicle position coordinate storage database of the vehicle position monitoring system. The data to be transferred is vehicle position coordinate data. The target table is a vehicle position coordinate data table. Based on the data characteristics and analysis requirements of each target table, the data in each target table is analyzed and processed separately to obtain the analysis results, including: Calculate the deviation between the vehicle position coordinates in the vehicle position coordinate data table and the standard vehicle position coordinates; Determine whether the deviation is greater than a preset deviation threshold; If the deviation is greater than or equal to the stated deviation threshold, an alarm will be issued to remind maintenance personnel to correct the position of the conveyor chain.

8. A data acquisition and analysis processing device for a painting workshop, characterized in that, The device includes: The data acquisition module is suitable for transferring data from the painting workshop system to the corresponding target table in the preset target database using object tools in the scripting language and the network within the painting workshop. The data processing module is suitable for analyzing and processing the data in each target table according to the data characteristics and analysis requirements of each target table, so as to obtain the analysis and processing results; The results display module is suitable for displaying or pushing the analysis and processing results.

9. An electronic device, characterized in that, It includes a processor and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the painting workshop data acquisition and analysis processing method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which, when executed by a processor, implement the painting workshop data acquisition and analysis processing method according to any one of claims 1-7.

Images