A production result prediction method, device, equipment and storage medium

Abstract

The present application provides a production result prediction method, device, equipment and storage medium. The method includes obtaining multiple data sample groups, subdividing the production operation into multiple operation types according to the operation parameter characteristics in the data sample groups, collecting the pre-execution operation parameters and current state parameters in the current production process, and identifying the target operation type corresponding to the pre-execution operation parameters, so as to retrieve the target state parameters matching the target operation type in the data sample group, and predict the production result corresponding to the current pre-execution operation according to the corresponding production results of the target state parameters in the historical data; the method predicts the production result of the current pre-execution operation by integrating the historical data, thereby realizing the accurate prediction of the production result, helping to improve production efficiency, reduce production costs, and provide a scientific basis for production decisions.

Description

Technical Field

[0001] The present application relates to the field of intelligent production technology, and specifically to a production result prediction method, device, equipment and storage medium. Background Art

[0002] In the field of industrial production, especially in production processes involving complex physical and chemical reactions, the mechanism of the influence of operating parameters on control targets (such as product quality, output, energy consumption, etc.) is extremely complex. Such production processes are often accompanied by a high degree of nonlinearity, time-varying and uncertainty, making it difficult to accurately describe and predict the relationship between operating parameters and final production results through traditional mechanism control models. In addition, since these influences usually have a certain time lag, that is, the adjustment of operating parameters will not be immediately reflected in the production results, but will only appear after a period of time, which further increases the difficulty of automated precision control.

[0003] Traditionally, in order to cope with this complexity, industrial production mainly relies on experienced operators to perform manual prediction and regulation. With their long-term accumulated experience and intuition, operators predict and adjust operating parameters according to various signs and parameter changes in the production process to achieve the desired control objectives. However, this method of relying on manual prediction is difficult to achieve standardization and automation of the production process due to the subjectivity and variability of manual prediction, which is not conducive to improving production efficiency, ensuring product quality and reducing production costs. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a production result prediction method, device, equipment and storage medium to solve the above-mentioned technical problems.

[0005] The present application provides a production result prediction method, which includes: obtaining multiple data sample groups to obtain historical operation parameters, historical state parameters, and historical production result data based on the sample data groups; subdividing the production operation into multiple operation types according to the operation parameter characteristics in the data sample groups; collecting pre-execution operation parameters and current state parameters in the current production process, and identifying the operation type to which the pre-execution operation parameters belong to obtain a target operation type, wherein the pre-execution operation parameters are operation parameters corresponding to the current pre-execution operation; retrieving target state parameters matching the target operation type in the data sample group, and predicting the production result corresponding to the current pre-execution operation according to the corresponding production results of the target state parameters in the historical data.

[0006] In one embodiment of the present application, before obtaining multiple data sample groups, it also includes: obtaining historical operation parameters, historical status parameters, and historical production result data; performing correlation analysis on the historical operation parameters, historical status parameters, and historical production result data to establish a correlation relationship between the three; based on the correlation relationship, constructing multiple data sample groups, each data sample group contains correlation information between historical operation parameters, historical status parameters and their corresponding historical production result data.

[0007] In one embodiment of the present application, production operations are subdivided into multiple operation types based on operation parameter characteristics in a data sample group, including: selecting a preset number of data points from the data sample as initial cluster centroids; allocating data points based on the initial cluster centroids, the data point allocation step including calculating the distance from each data point to all initial cluster centroids, and allocating each data point to the operation category to which the nearest centroid belongs; performing a centroid update step for each operation category, the centroid update step recalculates an updated cluster centroid based on all data points in the operation category, the updated cluster centroid position is the average of all data points in the category; comparing the updated cluster centroid with the previous centroid position, and calculating the centroid position deviation; repeating the data point allocation step and the centroid update step until the centroid position deviation is less than a preset threshold or reaches a preset maximum number of iterations, so as to subdivide the production operations into multiple operation types.

[0008] In one embodiment of the present application, target state parameters matching the target operation type are retrieved in the data sample group, including: retrieving historical state parameters matching the target operation type in the data sample group, and using the retrieved historical state parameters as candidate state parameters; calculating the similarity between each candidate state parameter and the current state parameter, and determining the candidate state parameter whose similarity exceeds a preset threshold as the target state parameter.

[0009] In one embodiment of the present application, retrieving the target state parameters matching the target operation type in the data sample group also includes: obtaining the similarity between each candidate state parameter and the current state parameter; sorting the candidate state parameters in descending order of similarity to obtain a candidate state parameter sequence; and taking the first preset number of candidate state parameters in the candidate state parameter sequence as the target state parameters.

[0010] In one embodiment of the present application, the production result corresponding to the current pre-execution operation is predicted based on the corresponding production result of the target state parameter in the historical data, including: determining the production result corresponding to the target state parameter in the historical data as candidate result data; determining the prediction data type according to the preset test requirements, and calculating the prediction value corresponding to the prediction data type based on the candidate result data; comparing the prediction value with the preset standard value to obtain a prediction conclusion based on the comparison result.

[0011] In one embodiment of the present application, after obtaining a prediction conclusion based on the comparison result, it also includes: if the prediction conclusion does not meet the preset production purpose, changing the pre-execution operation to obtain changed pre-execution operation data; re-predicting based on the changed pre-execution operation data; repeating the steps of changing the pre-execution operation and re-predicting until the prediction conclusion meets the preset production purpose; when the prediction conclusion meets the preset production purpose, executing the production operation based on the changed pre-execution operation data.

[0012] The present application provides a production result prediction device, which includes: a historical data acquisition module, which is used to obtain multiple data sample groups, and the sample data groups include historical operation parameters, historical state parameters, and historical production result data; an operation classification module, which is used to subdivide the production operation into multiple operation types according to the operation parameter characteristics in the data sample group; a prediction data acquisition module, which is used to collect pre-execution operation parameters and current state parameters in the current production process, and identify the operation type to which the pre-execution operation parameters belong, so as to obtain a target operation type, and the pre-execution operation parameters are the operation parameters corresponding to the current pre-execution operation; a production prediction module, which is used to retrieve a target state parameter matching the target operation type in the data sample group, and predict the production result corresponding to the current pre-execution operation according to the corresponding production result of the target state parameter in the historical data.

[0013] The present application provides an electronic device, characterized in that it includes a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the production result prediction method as described above.

[0014] The present application provides a computer-readable storage medium, characterized in that a computer program is stored thereon, and the computer program is used to enable a computer to execute the production result prediction method as described above.

[0015] Beneficial effects of the present application: The production result prediction method in the present application obtains multiple data sample groups, and according to the operation parameter characteristics in the data sample groups, the production operation is subdivided into multiple operation types, the pre-execution operation parameters and current state parameters in the current production process are collected, and the target operation type corresponding to the pre-execution operation parameters is identified, so as to retrieve the target state parameters matching the target operation type in the data sample group, and predict the production result corresponding to the current pre-execution operation according to the corresponding production results of the target state parameters in the historical data. The method predicts the production result of the current pre-execution operation by integrating historical data, and its prediction result is not only accurate, but also based on a large amount of historical data, with a high degree of reliability and credibility, providing strong data support for production decision makers, enabling them to make more scientific and reasonable decisions; in addition, by subdividing the production operation into multiple operation types and predicting based on the data sample groups, the control over the production process is also effectively enhanced.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0018] Figure 1 is a schematic diagram of an implementation environment of a production result prediction method shown in an exemplary embodiment of the present application;

[0019] Figure 2 is a flow chart of a production result prediction method shown in an exemplary embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of the overall steps of a production result prediction method shown in an exemplary embodiment of the present application;

[0021] Figure 4 is a block diagram of a production result prediction device shown in an exemplary embodiment of the present application;

[0022] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0023] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0025] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0026] Figure 1 It is a schematic diagram of an implementation environment of a production result prediction method shown in an exemplary embodiment of the present application.

[0027] like Figure 1 As shown, the implementation environment of the production result prediction method proposed in this embodiment includes a data acquisition module 101 and a computer module 102. Among them, the data acquisition module 101 is responsible for collecting various types of data sample groups from the production environment in real time or periodically. These data sample groups not only contain historical operating parameters, historical state parameters and corresponding historical production result data, but may also cover other relevant information, such as the status information of production equipment, environmental conditions, etc. The data acquisition module can be a sensor or interface (such as a temperature sensor, a pressure sensor, etc.) integrated in the production equipment, or it can be an external device connected to the production equipment by wireless or wired means. This application does not impose any restrictions on the specific implementation form of the data acquisition module, as long as the required production-related data can be effectively collected. In addition, during the data acquisition process, the module 101 will perform preliminary processing on the collected data, such as formatting, cleaning, etc., to ensure the quality and consistency of the data, and store the pre-processed data in a temporary buffer, and then pass it to the computer module 102 for further processing and analysis.

[0028] The computer module 102 is used to process the raw data collected by the data acquisition module 101. According to the characteristics of the operating parameters in the data sample group, the production operation is subdivided into multiple operation types using a machine learning algorithm (such as a clustering algorithm) to more accurately match the operation mode under the current production conditions. In addition, the computer module 102 is also responsible for collecting the pre-execution operation parameters and current state parameters in the current production process, and identifies the operation type to which these pre-execution operation parameters belong through a pre-established classification model to determine the target operation type. Based on the determined target operation type, the computer module 102 retrieves the target state parameters that match it in the data sample group, and then predicts the production results that may be generated by the current pre-execution operation based on the corresponding production results of these target state parameters in the historical data. The computer module 102 can also provide feedback suggestions based on the prediction results to help adjust the production parameters and optimize the production process to achieve the expected production goals. Therefore, the computer module 102 can be a high-performance processor, a dedicated production data analysis platform, or an embedded system integrated on the production equipment. This application also does not impose any restrictions on the specific type of the computer module, as long as it can efficiently process and analyze data and support the needs of production result prediction.

[0029] Figure 2 It is a flowchart of a production result prediction method shown by an exemplary embodiment of the present application.

[0030] like Figure 2 As shown, in an exemplary embodiment, the production result prediction method includes at least steps S210 to S240, which are described in detail as follows:

[0031] Step S210, acquiring multiple data sample groups to obtain historical operation parameters, historical state parameters, and historical production result data based on the sample data groups.

[0032] In one embodiment of the present application, before obtaining multiple data sample groups, it also includes: obtaining historical operation parameters, historical status parameters, and historical production result data; performing correlation analysis on the historical operation parameters, historical status parameters, and historical production result data to establish a correlation relationship between the three; based on the correlation relationship, constructing multiple data sample groups, each data sample group contains correlation information between the historical operation parameters, historical status parameters and their corresponding historical production result data.

[0033] In a specific embodiment of the present application, three types of key data are first collected in real time or periodically according to the production time sequence: adjustable historical operating parameters (such as temperature setting values, pressure setting values, etc.), non-adjustable historical state parameters (such as ambient humidity, raw material characteristics, etc.), and historical production result data (such as product quality indicators, output, etc.). Then, all these data are recorded according to the timestamp and immediately entered into the message queue to ensure the timeliness and integrity of the data. Next, these data are extracted from the message queue, and the operating parameters and state parameters are correlated and analyzed according to the control objectives (such as improving product quality or reducing costs). Then, the multiple data samples after correlation are stored in the database, each of which contains not only the corresponding timestamp, but also the operating parameters, state parameters and production result data.

[0034] In addition, in order to establish the correlation between operating parameters, state parameters and production results, statistical analysis or other machine learning methods are used to understand which combinations of operating parameters and state parameters have a significant impact on the final production results. Based on this correlation, data samples with similar characteristics are grouped together to form multiple data sample groups. The data points within each sample group have a strong correlation between operating parameters, state parameters and production results. The formed sample group contains not only the original historical operating parameters, historical state parameters and their corresponding historical production result data, but also the correlation information between these data.

[0035] It is understandable that this structured sample group can provide strong support for subsequent production optimization, prediction model training, etc., and can be used for decision support in the production process. For example, when planning to adjust operating parameters, the adjustment effect can be estimated by comparing the current conditions with similar situations in the historical sample group, so as to make more scientific and reasonable decisions. Therefore, based on the data processing method proposed in this embodiment, it is possible to obtain and process data from the actual production environment, and then form a valuable sample group for further analysis and decision support. This not only helps to improve the efficiency and quality of production, but also provides data-driven support for future production planning.

[0036] Step S220: subdividing the production operation into multiple operation types according to the operation parameter characteristics in the data sample group.

[0037] In one embodiment of the present application, production operations are subdivided into multiple operation types based on operation parameter characteristics in a data sample group, including: selecting a preset number of data points from the data sample as initial cluster centroids; allocating data points based on the initial cluster centroids, the data point allocation step including calculating the distance from each data point to all initial cluster centroids, and allocating each data point to the operation category to which the nearest centroid belongs; performing a centroid update step for each operation category, the centroid update step recalculates an updated cluster centroid based on all data points in the operation category, the updated cluster centroid position is the average of all data points in the category; compares the updated cluster centroid with the previous centroid position, and calculates the centroid position deviation; and repeatedly executing the data point allocation step and the centroid update step until the centroid position deviation is less than a preset threshold or reaches a preset maximum number of iterations, so as to subdivide the production operations into multiple operation types.

[0038] In order to subdivide production operations into multiple operation types according to the operating parameter characteristics in the data sample group, a clustering algorithm is usually used to classify them, such as K-means, DBSCAN and other classic clustering algorithms, etc. This application does not impose any restrictions on its specific clustering algorithm.

[0039] In a specific embodiment of the present application, the classification of multidimensional operating parameters using the K-means clustering algorithm is taken as an example. First, the system obtains a sample group of historical production data from the database, each sample group containing historical operating parameters, historical state parameters, and corresponding historical production result data. These operating parameters are usually multidimensional and can be one-dimensional or multi-dimensional data. Next, a preset number (K) of data points are randomly selected from these data samples as the initial clustering centroids, where the K value can be pre-set according to actual needs, such as determined by experience or domain knowledge.

[0040] Then, based on the selected initial cluster centroids, the system calculates the distance of each data point to all initial cluster centroids and assigns each data point to the operation category to which the nearest centroid belongs. This step ensures that each data point is classified into the most similar operation category. For each operation category, the system recalculates the updated cluster centroid based on all data points in that category, and the updated centroid position is the average of all data points in that category. At the same time, the updated cluster centroid is compared with the previous centroid position, and the centroid position deviation is calculated to assess whether the centroid has changed significantly.

[0041] Then, the data point allocation step and the centroid update step are repeated until the centroid position deviation is less than the preset threshold or the preset maximum number of iterations is reached. At this point, the algorithm converges and the subdivision of the operation type is completed. The preset threshold and the maximum number of iterations can be flexibly adjusted according to the specific application scenario to ensure the efficiency and accuracy of the algorithm. Once the clustering process is completed and converged, the system will store the generated clustering model. This model can not only be used to classify data samples in the existing database according to the operation parameters, but also to quickly determine the operation parameter categories of newly added samples.

[0042] Later, during the production process, when the operating parameters need to be adjusted, the operator can input the planned operating parameters into the trained clustering model to quickly obtain the operation type to which the current parameters to be adjusted belong. In addition, the model also supports subsequent predictive analysis and decision support, such as estimating the adjustment effect by comparing the current conditions with similar situations in the historical sample group, so as to make more scientific and reasonable decisions. This method combines classic clustering algorithms (such as K-means), which not only improves the flexibility and adaptability of production, but also provides data-driven support for future production planning, ensuring the scientificity and reliability of the classification process while maintaining high computational efficiency.

[0043] It can be understood that the method proposed in this embodiment can effectively subdivide production operations into multiple operation types, thereby enhancing the flexibility and adaptability of production, and also providing strong data support for future production planning.

[0044] Step S230, collecting pre-execution operation parameters and current state parameters in the current production process, and identifying the operation type to which the pre-execution operation parameters belong, so as to obtain the target operation type, wherein the pre-execution operation parameters are operation parameters corresponding to the current pre-execution operation.

[0045] In one embodiment of the present application, in order to collect the pre-execution operation parameters and current state parameters in the current production process, and identify the operation type to which the pre-execution operation parameters belong to obtain the target operation type, the following steps are adopted: First, according to the production time sequence, the system collects the pre-execution operation parameters (such as temperature setting value, pressure setting value, etc.) and current state parameters (such as ambient humidity, raw material characteristics, etc.) in the current production process in real time or periodically. These data are recorded according to the timestamp and immediately input into the message queue to ensure the timeliness and integrity of the data. When the operator needs to adjust the operation parameters, they input the planned adjustment pre-execution operation parameters into the classification model trained by the clustering algorithm before. The model has been fully trained on the basis of historical data and can effectively classify multidimensional operation parameters. Based on the input pre-execution operation parameters, the model quickly identifies the operation type to which it belongs, that is, the target operation type. Next, the identified target operation type is further associated with the analysis, and the possible impact on the production results is evaluated in combination with the current state parameters.

[0046] In actual applications, when an operator is ready to adjust an operating parameter, he only needs to enter the planned adjustment pre-execution operating parameter into the system, and the system will immediately return the target operation type to which the parameter belongs and provide corresponding adjustment suggestions. This not only improves the flexibility and adaptability of production, but also provides data-driven support for future production planning. As more data accumulates, the system can continuously optimize the clustering model, improve the accuracy of classification and the reliability of prediction, and ensure the efficient operation of the production process.

[0047] It can be understood that the method proposed in this embodiment can effectively collect the pre-execution operation parameters and current state parameters in the current production process, and identify the operation type to which the pre-execution operation parameters belong to obtain the target operation type. The combination of real-time data acquisition, advanced clustering algorithms and model applications not only improves the efficiency and quality of production, but also provides powerful decision-making support for operators, ensuring the efficiency and stability of the production process. It not only enhances the flexibility and adaptability of production, but also provides a solid data foundation and technical support for future production planning. By continuously optimizing the clustering model, the system can continuously improve the accuracy of classification and the reliability of prediction, thereby ensuring the efficient operation of the production process.

[0048] Step S240, searching the data sample group for target state parameters that match the target operation type, and predicting the production result corresponding to the current pre-execution operation based on the corresponding production results of the target state parameters in the historical data.

[0049] In one embodiment of the present application, a target state parameter that matches a target operation type is retrieved in a data sample group, including: retrieving historical state parameters that match the target operation type in the data sample group, and using the retrieved historical state parameters as candidate state parameters; calculating the similarity between each candidate state parameter and the current state parameter, and determining the candidate state parameter whose similarity exceeds a preset threshold as the target state parameter; or, sorting the candidate state parameters in descending order of similarity to obtain a candidate state parameter sequence, and taking the first preset number of candidate state parameters in the candidate state parameter sequence as the target state parameters.

[0050] In one embodiment of the present application, in order to retrieve the target state parameters that match the target operation type in the data sample group, the system first collects the pre-execution operation parameters (such as temperature setting value, pressure setting value, etc.) and current state parameters (such as ambient humidity, raw material characteristics, etc.) in the current production process in real time or periodically according to the production time sequence, and inputs these data into the message queue after recording them according to the timestamp to ensure the timeliness and integrity of the data. When the operators need to adjust the operation parameters, they input the planned pre-execution operation parameters into the classification model trained by the clustering algorithm before, and the model quickly identifies the operation type to which it belongs, that is, the target operation type.

[0051] Next, all historical state parameters that match the target operation type are searched in the historical data sample group, and the retrieved historical state parameters are used as candidate state parameters. This step aims to filter out samples similar to the current planned adjustment operation from a large amount of historical data to eliminate the impact of different state parameters on the control target. For further screening, the system calculates the similarity between each candidate state parameter and the current state parameter, using a variety of multidimensional similarity measurement methods, such as Euclidean distance, Manhattan distance, etc. Here, the Euclidean distance is used as an example to explain in detail: For the target sample (current state parameter) and each candidate state parameter in the sample library, the Euclidean distance formula is used to calculate the distance between the two. The calculation formula is as follows:

[0052]

[0053] Among them, x i and i are the values ​​of the target sample and the sample in the library on the i-th feature, respectively.

[0054] Then, the calculated distance and its corresponding candidate state parameters are stored to form a list or array, each element of which contains the distance value and the corresponding sample information. The distance values ​​in the stored distance list are compared, and the candidate state parameters whose similarity exceeds the preset threshold are selected and finally determined as the target state parameters. The target state parameters will be used to predict and optimize the adjustment effect in subsequent production decisions.

[0055] In a specific embodiment of the present application, after calculating the similarity distance between each candidate state parameter and the current state parameter, the distance values ​​in the stored distance list are compared, and the candidate state parameter whose similarity exceeds the preset threshold is selected. The preset threshold can be flexibly set according to the specific application scenario to ensure the accuracy of the screening result. The present application does not impose any restrictions on the setting method of the preset threshold and its specific value.

[0056] In another specific embodiment of the present application, after calculating the similarity distance between each candidate state parameter and the current state parameter, the candidate state parameters are sorted in descending order of similarity to obtain a candidate state parameter sequence. Then, the system takes the candidate state parameters of the first preset number of items in the candidate state parameter sequence as the target state parameters. The preset number of items can be pre-set according to specific needs, such as the first N most similar samples. Similarly, the present application does not impose any restrictions on the determination method and value of the specific item value of the preset number of items.

[0057] In addition, two groups of target state parameters can be obtained based on threshold comparison and sorting, and the result parameters corresponding to the two groups of parameters can be weighted or verified for deduplication. Specifically, first, the system determines the first group of target state parameters based on the candidate state parameters whose similarity exceeds the preset threshold, ensuring that the screened samples have a high similarity, thereby improving the reliability of the prediction results. At the same time, the system sorts all candidate state parameters in descending order of similarity, and takes the candidate state parameters of the first preset number of items as the second group of target state parameters to ensure that even if some high-similarity samples fail to pass the threshold screening, they can be taken into consideration through sorting interception. Then, different weights can be assigned to the corresponding result parameters (such as historical production result data). For example, the first group of target state parameters based on threshold screening may be considered more reliable and therefore assigned a higher weight; while the second group of target state parameters based on sorting interception is assigned a lower weight; and the result parameters corresponding to the two groups of target state parameters are weighted averaged to generate the final prediction results. This weighted calculation method can combine the advantages of the two screening mechanisms to provide more accurate and robust predictions. In addition, in order to eliminate the impact of duplicate samples, the two sets of target state parameters can also be verified and deduplicated. If a sample appears in both groups, only one copy is retained, and the weight is adjusted according to its source to ensure the uniqueness and accuracy of the final result. After deduplication, the system integrates and analyzes the two sets of target state parameters and their corresponding result parameters, including but not limited to calculating statistics such as mean, median, upper and lower quartiles, to comprehensively evaluate the expected effects of different operating parameter adjustments.

[0058] In one embodiment of the present application, the production result corresponding to the current pre-execution operation is predicted based on the corresponding production results of the target state parameters in the historical data, including: determining the production result corresponding to the target state parameters in the historical data as candidate result data; determining the prediction data type according to the preset test requirements, and calculating the prediction value corresponding to the prediction data type based on the candidate result data; comparing the prediction value with the preset standard value to obtain a prediction conclusion based on the comparison result.

[0059] At the production site, when an operator is preparing to adjust a certain operating parameter, he will not only obtain candidate result data through the target state parameter screening, but also determine the prediction data type according to the preset test requirements, and calculate the prediction value based on the candidate result data. Finally, the prediction value is compared with the preset standard value to draw a prediction conclusion.

[0060] In a specific embodiment of the present application, based on the target state parameters obtained by the similarity exceeding the preset threshold or the sorting interception, the production results corresponding to these target state parameters are searched in the historical data, and the historical production results are determined as candidate result data. Then, based on the actual production needs, it is necessary to evaluate the calculation situation after the operation is executed, and it is necessary to calculate the maximum value of the predicted result, so as to determine the extreme situation based on the maximum value. Therefore, after obtaining each candidate result data, the corresponding maximum value is calculated, and the maximum and minimum values ​​of the candidate result data are found to evaluate the extreme situation.

[0061] In addition, you can also calculate the mean of the candidate result data as the average level of the expected adjustment results; calculate the median of the candidate result data to evaluate the typical performance after adjustment; calculate the upper and lower quartiles to understand the distribution range of the adjusted results; draw a distribution histogram of the candidate result data to intuitively display the overall distribution after adjustment.

[0062] In general, the type of data that needs to be displayed will vary depending on different production requirements. Therefore, based on the candidate result data obtained through screening, the specific data values ​​corresponding to these requirements are calculated to accurately predict the final production situation. Specifically, first identify the specific production requirements (such as increasing output, optimizing quality, reducing costs, etc.), and then determine the specific data type related to the requirement. For each requirement, the corresponding statistical indicators are calculated based on the candidate result data, such as the mean is used to evaluate the average level of the expected adjustment results, the median reflects the typical performance after adjustment, the upper and lower quartiles show the distribution range of the adjusted results, the extreme value evaluates the performance in extreme cases, and the distribution histogram intuitively displays the overall distribution after adjustment. Through the above calculations, detailed forecast data is generated to provide a comprehensive estimate and analysis of the final production situation.

[0063] It can be understood that the method proposed based on this embodiment ensures that the prediction results can accurately match specific production needs, and provides more customized and targeted decision support, so that operators can make more scientific and reasonable adjustment decisions based on these prediction data, thereby optimizing the production process and improving efficiency and quality.

[0064] In one embodiment of the present application, after obtaining a prediction conclusion based on the comparison result, it also includes: if the prediction conclusion does not meet the preset production purpose, changing the pre-execution operation to obtain changed pre-execution operation data; re-predicting based on the changed pre-execution operation data; repeating the steps of changing the pre-execution operation and re-predicting until the prediction conclusion meets the preset production purpose; when the prediction conclusion meets the preset production purpose, executing the production operation based on the changed pre-execution operation data.

[0065] In a specific embodiment of the present application, after the prediction conclusion is obtained according to the comparison result, the pre-execution operation is further optimized to meet the preset production purpose. The specific process is as follows: First, the system makes an initial prediction based on the current pre-execution operation parameters and state parameters, and evaluates the expected production situation after adjustment by screening the candidate result data in the historical data and calculating the corresponding statistical indicators (such as mean, median, upper and lower quartiles, maximum, distribution histogram, etc.). Compare these predicted values ​​with the preset standard values ​​to generate a prediction conclusion. If the prediction conclusion is "meeting the preset production purpose", go directly to the next step; otherwise, enter the pre-execution operation change process. If the prediction conclusion is "not meeting the preset production purpose", the system will prompt the operator or automatically change the pre-execution operation, such as adjusting the temperature setting value, changing the pressure setting value, etc., to obtain the changed pre-execution operation data, and record the data of each change. Subsequently, the system uses the changed pre-execution operation data to find similar samples in the historical data, re-predict, including recalculating the above statistical indicators, evaluating the new expected adjustment results, and updating the prediction conclusion. If the new prediction conclusion is still "does not meet the preset production purpose", the system will continue to execute the steps of changing the pre-execution operation and re-prediction, and iterate continuously until the prediction conclusion becomes "meets the preset production purpose". When the final prediction conclusion meets the preset production purpose, the system stops iterating and prepares to perform actual production operations. At this time, the system guides the operator to perform actual production operations based on the finalized changed pre-execution operation data, ensuring the flexibility and adaptability of production, while providing scientific and reasonable decision support.

[0066] In addition, during the actual production process, the system continuously monitors production parameters and results in order to adjust and optimize production operations in a timely manner to ensure the achievement of production goals. At the production site, when the operator is ready to adjust a certain operating parameter, the system will not only make an initial prediction and generate a conclusion, but also automatically or manually change the pre-execution operation if the prediction conclusion does not meet the preset production purpose, and repeatedly predict until the demand is met. Finally, the system executes production operations based on the optimal pre-execution operation data to ensure the efficiency and stability of the production process. As more data accumulates, the system can continuously optimize the similarity measurement method, threshold setting, and weight distribution strategy to improve the accuracy of classification and the reliability of prediction, ensuring the efficient operation of the production process.

[0067] Figure 3 It is a schematic diagram of the overall steps of a production result prediction method shown in an exemplary embodiment of the present application.

[0068] like Figure 3 As shown, firstly, historical production data is obtained, and the historical production data at least includes historical operation parameters, historical state parameters, and historical production result data, and the association relationship between each production data is established, and a data sample group is generated based on the association relationship and stored in the database. Then, the operation types in the production process are classified based on the operation parameters, which can be achieved by constructing a classification model. Then, the pre-execution operation parameters of the current pre-executed production operation are obtained, and the target operation type corresponding to the pre-execution operation parameters is identified through the aforementioned constructed classification model. Next, the candidate state parameters corresponding to the target operation type are retrieved in the database, and then the current state parameters corresponding to the pre-execution operation parameters and the candidate state parameters are compared to obtain the target state parameters. Finally, the prediction results corresponding to the pre-execution operation parameters are obtained by calculating the target state parameters.

[0069] In a specific embodiment of the present application, historical production data including historical operating parameters, historical state parameters and historical production result data are first collected, and the association relationship between these data is established, and a data sample group is generated and stored in a database. Next, the operation types in the production process are classified based on the operating parameters, and this is achieved by building and training a classification model, which can identify the operation types corresponding to different operating parameters. Subsequently, the current pre-executed production operation parameters are obtained, and the target operation types corresponding to these parameters are identified using the aforementioned classification model. After determining the target operation type, candidate state parameters matching the type are retrieved from the database, and these candidate state parameters are compared with the current state parameters to determine the most similar target state parameters. Finally, based on the target state parameters, the prediction results corresponding to the pre-execution operation parameters are calculated to provide scientific and reasonable adjustment suggestions and support future production decisions.

[0070] It can be understood that the method proposed in this embodiment ensures the flexibility and adaptability of the production process, while providing strong decision support for operators and improving the efficiency and quality of production. Through continuous data collection, application of classification models and precise matching of target state parameters, not only the accuracy and reliability of prediction are improved, but also the adjustment process of production operations is optimized, ensuring the efficiency and stability of the production process.

[0071] Figure 4 is a block diagram of a production result prediction device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1 The device may also be applicable to other exemplary implementation environments and specifically configured in other devices, and this embodiment does not limit the implementation environment to which the device is applicable.

[0072] like Figure 4 As shown, the exemplary production result prediction device includes: a historical data collection module 410, an operation classification module 420, a prediction data collection module 430, and a production prediction module 440.

[0073] Among them, the historical data collection module 410 is used to obtain multiple data sample groups, and the sample data groups include historical operation parameters, historical state parameters, and historical production result data; the operation classification module 420 is used to subdivide the production operation into multiple operation types according to the operation parameter characteristics in the data sample group; the prediction data collection module 430 is used to collect the pre-execution operation parameters and current state parameters in the current production process, and identify the operation type to which the pre-execution operation parameters belong, so as to obtain the target operation type, and the pre-execution operation parameters are the operation parameters corresponding to the current pre-execution operation; the production prediction module 440 is used to retrieve the target state parameters matching the target operation type in the data sample group, and predict the production result corresponding to the current pre-execution operation according to the corresponding production result of the target state parameters in the historical data.

[0074] It should be noted that the production result prediction device provided in the above embodiment and the production result prediction method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In practical applications, the production result prediction device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0075] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device implements the production result prediction method provided in the above-mentioned embodiments.

[0076] Figure 5 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 5 The computer system 500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0077] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503, such as executing the method described in the above embodiment. In the RAM 503, various programs and data required for system operation are also stored. The CPU 501, the ROM 502 and the RAM 503 are connected to each other through the bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0078] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read therefrom is installed into the storage section 508 as needed.

[0079] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 509, and / or installed from a removable medium 511. When the computer program is executed by a central processing unit (CPU) 501, various functions defined in the system of the present application are executed.

[0080] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0081] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0082] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0083] Another aspect of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor of a computer, the computer executes the production result prediction method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0084] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the production result prediction method provided in each of the above embodiments.

[0085] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A production result prediction method, characterized in that: The method comprises: Acquire a plurality of data sample groups to obtain historical operation parameters, historical state parameters, and historical production result data based on the sample data groups; According to the characteristics of the operating parameters in the data sample group, the production operations are subdivided into multiple operation types; Collecting pre-execution operation parameters and current state parameters in the current production process, and identifying the operation type to which the pre-execution operation parameters belong, so as to obtain a target operation type, wherein the pre-execution operation parameters are operation parameters corresponding to the current pre-execution operation; The target state parameters matching the target operation type are retrieved from the data sample group, and the production result corresponding to the current pre-execution operation is predicted based on the corresponding production results of the target state parameters in the historical data.

2. The production result prediction method according to claim 1, characterized in that: Before getting multiple data sample groups, also include: Obtain historical operation parameters, historical status parameters, and historical production result data; Conduct correlation analysis on historical operation parameters, historical status parameters and historical production result data to establish the correlation relationship between the three; Based on the association relationship, multiple data sample groups are constructed, each data sample group contains association information between historical operation parameters, historical status parameters and their corresponding historical production result data.

3. The production result prediction method according to claim 1, characterized in that: According to the characteristics of the operating parameters in the data sample group, the production operations are subdivided into multiple operation types, including: Select a preset number of data points from the data sample as initial cluster centroids; Performing data point allocation based on the initial cluster centroids, the data point allocation step comprising calculating the distance of each data point to all initial cluster centroids, and allocating each data point to the operation category to which the nearest centroid belongs; Performing a centroid updating step for each operation category, wherein the centroid updating step recalculates an updated cluster centroid based on all data points in the operation category, wherein the updated cluster centroid position is the average value of all data points in the category; comparing the updated cluster centroid position with the previous centroid position, and calculating the centroid position deviation; The data point allocation step and the centroid updating step are repeatedly performed until the centroid position deviation is less than a preset threshold or a preset maximum number of iterations is reached, so as to subdivide the production operation into multiple operation types.

4. The production result prediction method according to claim 1, characterized in that: Retrieving a target state parameter matching the target operation type from the data sample group includes: Retrieving historical state parameters matching the target operation type in the data sample group, and using the retrieved historical state parameters as candidate state parameters; The similarity between each candidate state parameter and the current state parameter is calculated, and the candidate state parameter whose similarity exceeds a preset threshold is determined as the target state parameter.

5. The production result prediction method according to claim 4, characterized in that: Retrieving a target state parameter matching the target operation type from the data sample group further includes: Obtaining the similarity between each candidate state parameter and the current state parameter; Sorting the candidate state parameters in descending order of similarity to obtain a candidate state parameter sequence; The first preset number of candidate state parameters in the candidate state parameter sequence are taken as the target state parameters.

6. The production result prediction method according to claim 1, characterized in that: According to the corresponding production results of the target state parameters in the historical data, the production results corresponding to the current pre-execution operation are predicted, including: Determine the production results corresponding to the target state parameters in the historical data as candidate result data; Determine the prediction data type according to the preset test requirements, and calculate the prediction value corresponding to the prediction data type based on the candidate result data; The predicted value is compared with a preset standard value to obtain a prediction conclusion based on the comparison result.

7. The production result prediction method according to claim 6, characterized in that: After the prediction conclusion is obtained based on the comparison results, it also includes: If the prediction conclusion is that the preset production purpose is not met, the pre-execution operation is changed to obtain the changed pre-execution operation data; Re-forecast based on the changed pre-execution operation data; Repeat the steps of changing the pre-execution operation and re-forecasting until the forecast conclusion meets the preset production purpose; When the prediction conclusion meets the preset production purpose, the production operation is performed based on the changed pre-execution operation data.

8. A production result prediction device, characterized in that: The device comprises: A historical data acquisition module, used to obtain a plurality of data sample groups, so as to obtain historical operation parameters, historical state parameters, and historical production result data based on the sample data groups; An operation classification module is used to subdivide production operations into multiple operation types according to operation parameter characteristics in a data sample group; A prediction data acquisition module, used to collect pre-execution operation parameters and current state parameters in the current production process, and identify the operation type to which the pre-execution operation parameters belong, so as to obtain a target operation type, wherein the pre-execution operation parameters are operation parameters corresponding to the current pre-execution operation; The production prediction module is used to retrieve the target state parameters matching the target operation type in the data sample group, and predict the production result corresponding to the current pre-execution operation based on the corresponding production results of the target state parameters in the historical data.

9. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute the computer program stored in the memory to implement the production result prediction method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to enable a computer to execute the production result prediction method as described in any one of claims 1-7.

Images