Liquid valve die casting production line abnormality early warning method, device and medium

Abstract

The application discloses a liquid valve die-casting production line abnormality early warning method and device and a medium, and relates to the technical field of data processing.The method comprises the following steps: collecting aluminum alloy die-casting production line information of a fire-fighting liquid valve, and obtaining E liquid valve die-casting production lines; obtaining E die-casting production line monitoring data sets according to a production line sensing monitoring module; extracting an e-th die-casting production line monitoring data set; reconstructing and optimizing the e-th die-casting production line monitoring data set, and obtaining an e-th optimized die-casting production line monitoring data set; performing deep abnormality detection on the e-th optimized die-casting production line monitoring data set, and obtaining an e-th die-casting production line abnormality coefficient; and performing abnormality early warning on an e-th liquid valve die-casting production line according to a liquid valve die-casting production line grading early warning module.The technical problem that the fire-fighting liquid valve production line monitoring data is inaccurate, which leads to difficulty in accurately judging the production line running state and thus failing to timely early warn potential faults is solved, and the technical effect of improving the production stability and safety of the fire-fighting liquid valve is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an abnormality early warning method, device and medium for a liquid valve die casting production line. BACKGROUND

[0002] With the rapid development of urban construction, fire safety has become an important link to protect people's life and property safety. As an indispensable part of the fire fighting system, the quality and reliability of the fire fighting liquid valve are directly related to the emergency response capability in the event of a fire. In the production process of the fire fighting liquid valve, aluminum alloy die casting is a common production process, and the running state and product quality of the production line have an important influence on the performance of the final product. However, in the actual operation process of the aluminum alloy die casting production line, due to various factors such as equipment aging, improper operation, environmental factors, etc., the production line monitoring data may have problems such as noise and distortion, making it difficult for traditional monitoring methods to accurately determine the running state of the production line and predict potential failures. This not only affects the production efficiency and product quality of the fire fighting liquid valve, but also may pose a potential risk to fire safety. SUMMARY

[0003] The embodiments of the present application provide an abnormality early warning method, device and medium for a liquid valve die casting production line, which solves the technical problem that the inaccurate monitoring data of the fire fighting liquid valve production line makes it difficult to accurately determine the running state of the production line, so that potential failures cannot be timely warned.

[0004] In a first aspect, the present application provides an abnormality early warning method for a liquid valve die casting production line, which comprises:

[0005] Collecting aluminum alloy die casting production line information of a fire fighting liquid valve to obtain E liquid valve die casting production lines, wherein each liquid valve die casting production line corresponds to a liquid valve aluminum alloy die casting node, and E is a positive integer greater than 1; obtaining E die casting production line monitoring data sets corresponding to the E liquid valve die casting production lines according to a production line sensor monitoring module; extracting an e-th die casting production line monitoring data set corresponding to an e-th liquid valve die casting production line according to the E die casting production line monitoring data sets, wherein e is a positive integer and 1≤e≤E; reconstructing and optimizing the e-th die casting production line monitoring data set according to a pre-processing loss function and a sensor monitoring compensation model to obtain an e-th optimized die casting production line monitoring data set; performing deep anomaly detection on the e-th optimized die casting production line monitoring data set according to a die casting production line anomaly detection module to obtain an e-th die casting production line anomaly coefficient; and performing abnormality early warning on the e-th liquid valve die casting production line in combination with the e-th die casting production line anomaly coefficient according to a liquid valve die casting production line grading early warning module.

[0006] In a second aspect, the application provides an electronic device, comprising a memory configured to store executable instructions, and a processor configured to execute the executable instructions stored in the memory to implement the method for early warning of abnormality of a liquid valve die casting production line.

[0007] In a third aspect, the application provides a computer readable storage medium storing a computer program, which, when executed by a processor, implements the method for early warning of abnormality of a liquid valve die casting production line.

[0008] The one or more technical solutions provided in the application have at least the following technical effects or advantages:

[0009] First, information of an aluminum alloy die casting production line of a fire-fighting liquid valve is collected to obtain E liquid valve die casting production lines, where each liquid valve die casting production line corresponds to a liquid valve aluminum alloy die casting node, and E is a positive integer greater than 1. Then, E die casting production line monitoring data sets corresponding to the E liquid valve die casting production lines are obtained according to a production line sensor monitoring module. Next, an e-th die casting production line monitoring data set corresponding to an e-th liquid valve die casting production line is extracted from the E die casting production line monitoring data sets, where e is a positive integer and 1≤e≤E. Subsequently, the e-th die casting production line monitoring data set is reconstructed and optimized according to a pre-processing loss function and a sensor monitoring compensation model to obtain an e-th optimized die casting production line monitoring data set. Then, a deep abnormality detection is performed on the e-th optimized die casting production line monitoring data set according to a die casting production line abnormality detection module to obtain an e-th die casting production line abnormality coefficient. Finally, an early warning of abnormality is performed on the e-th liquid valve die casting production line in combination with the e-th die casting production line abnormality coefficient according to a liquid valve die casting production line grading early warning module. The technical problem that the fire-fighting liquid valve production line monitoring data is inaccurate, which makes it difficult to accurately determine the running state of the production line and thus cannot timely warn potential faults is solved, and the technical effect of improving the production stability and safety of the fire-fighting liquid valve is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative effort.

[0011] Figure 1 The flowchart of the method for early warning of abnormality of a liquid valve die casting production line provided by the embodiments of the application is shown.

[0012] Figure 2 The flowchart of the method for early warning of abnormality of a liquid valve die casting production line provided by the embodiments of the application is shown.

[0013] Figure 3 Fig. 1 is a schematic diagram of an exemplary electronic device of the present application.

[0014] Reference numerals: processor 31, memory 32, input device 33, output device 34. DETAILED DESCRIPTION

[0015] The embodiment of the present application provides a liquid valve die casting production line abnormality early warning method, device and medium, and solves the technical problem that monitoring data of a fire-fighting liquid valve production line is inaccurate, so that it is difficult to accurately determine the running state of the production line, and potential faults cannot be timely warned.

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0017] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to the process, method, product or device.

[0018] Embodiment one, as shown in the present application, provides a liquid valve die casting production line abnormality early warning method, wherein the method comprises the following steps: Figure 1

[0019] Collecting aluminum alloy die casting production line information of the fire-fighting liquid valve to obtain E liquid valve die casting production lines, wherein each liquid valve die casting production line corresponds to a liquid valve aluminum alloy die casting node, and E is a positive integer greater than 1.

[0020] In the production process of the fire-fighting liquid valve, the aluminum alloy die casting is a crucial link, which directly determines the performance and quality of the liquid valve. In order to ensure the smooth progress of the production process and the stability of the product quality, the aluminum alloy die casting production line must be strictly monitored and managed. By collecting the aluminum alloy die casting production line information of the fire-fighting liquid valve, E liquid valve die casting production lines are obtained, and E is a positive integer greater than 1. Each liquid valve die casting production line corresponds to a liquid valve aluminum alloy die casting node, and these nodes are key control points in the die casting process, which have a direct impact on the final quality of the product. By monitoring the running state and key parameters of these die casting production lines in real time, potential problems can be found and handled in time, so as to ensure the stability of the production process and the reliability of the product quality.

[0021] ​According to the production line sensing monitoring module, E pressure casting production line monitoring data sets corresponding to the E liquid valve pressure casting production lines are obtained.

[0022] The production line sensing monitoring module is used to collect various data on the production line in real time, ensuring comprehensive monitoring of the production process. When the production line sensing monitoring module starts, data collection will be performed for the E liquid valve pressure casting production lines. Each liquid valve pressure casting production line is equipped with corresponding sensors and monitoring equipment that can capture and record key parameters and state information on the production line in real time. Through the production line sensing monitoring module, E pressure casting production line monitoring data sets corresponding to the E liquid valve pressure casting production lines can be obtained. Each data set contains various monitoring data of a specific production line over a period of time, such as temperature, pressure, flow, speed, etc., which are important indicators reflecting the running status of the production line and product quality.

[0023] According to the E pressure casting production line monitoring data sets, the e pressure casting production line monitoring data set corresponding to the e liquid valve pressure casting production line is extracted, where e is a positive integer and 1≤e≤E.

[0024] From the E monitoring data sets, the e pressure casting production line monitoring data set corresponding to the e liquid valve pressure casting production line is extracted, where e is a positive integer and 1≤e≤E, and e represents a specific liquid valve pressure casting production line.

[0025] According to the pre-processing loss function and the sensing monitoring compensation model, the e pressure casting production line monitoring data set is reconstructed and optimized to obtain the e optimized pressure casting production line monitoring data set.

[0026] After obtaining the e liquid valve pressure casting production line monitoring data set, in order to improve the quality of the data and remove potential noise or distortion, a pre-processing loss function and a sensing monitoring compensation model can be used to reconstruct and optimize the monitoring data set. The pre-processing loss function is used to preprocess the pressure casting production line monitoring data set to remove noise, outliers, duplicates or unnecessary redundant information in the data. Due to the influence of various factors such as environment, aging, calibration error, etc., the monitoring data of the sensor may have certain inaccuracies. The sensing monitoring compensation model is used to compensate for errors or deviations that may occur in the monitoring process of the sensor to improve the accuracy and reliability of the data. After optimization and reconstruction by the pre-processing loss function and the sensing monitoring compensation model, a more accurate and reliable e optimized pressure casting production line monitoring data set will be obtained.

[0027] Further, as shown in Figure 2 According to the pre-processing loss function and the sensing monitoring compensation model, the e pressure casting production line monitoring data set is reconstructed and optimized to obtain the e optimized pressure casting production line monitoring data set, including:

[0028] According to the preprocessing loss function, the e-th die casting production line monitoring data set is preprocessed to obtain an e-th reconstructed die casting production line monitoring data set; and according to the sensor monitoring compensation model, the e-th reconstructed die casting production line monitoring data set is corrected to generate an e-th optimized die casting production line monitoring data set.

[0029] Preferably, the e-th die casting production line monitoring data set is input into a preprocessing loss function for preprocessing, so as to remove noise, outliers, duplicates or unnecessary redundant information in the data, so as to obtain an e-th reconstructed die casting production line monitoring data set, which has higher data quality and more accurate parameter values than the e-th die casting production line monitoring data set. The sensor monitoring compensation model can be trained based on historical data, calibration data or machine learning algorithms to accurately predict and compensate for sensor errors. According to the sensor monitoring compensation model, the e-th reconstructed die casting production line monitoring data set is corrected to obtain an e-th optimized die casting production line monitoring data set. Through the above process, the quality of the monitoring data can be improved to more accurately reflect the actual running state of the die casting production line, thereby providing strong data support for the production of fire-fighting liquid valves.

[0030] Further, the preprocessing loss function is:

[0031] wherein n represents the subscript of the current data sample, and N represents the total number of data samples, represents the e-th die casting production line monitoring data set, represents the e-th reconstructed die casting production line monitoring data set, represents the preprocessing loss function of the e-th die casting production line monitoring data set and the e-th reconstructed die casting production line monitoring data set, so as to tends to 0 as a constraint for preprocessing the e-th die casting production line monitoring data set to obtain the e-th reconstructed die casting production line monitoring data set, is the feature similarity of the e-th die casting production line monitoring data set and the e-th reconstructed die casting production line monitoring data set, is the real-time similarity of the e-th die casting production line monitoring data set and the e-th reconstructed die casting production line monitoring data set, is the preset similarity of the e-th die casting production line monitoring data set and the e-th reconstructed die casting production line monitoring data set.

[0032] is the preprocessing loss function of the e-th die casting production line monitoring data set and the e-th reconstructed die casting production line monitoring data set, and the preprocessing loss function is obtained by tends to 0 as a constraint for preprocessing the e-th die casting production line monitoring data set. In the preprocessing loss function, represents the e-th die casting production line monitoring data set, characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset; characterizing the e-th reconstructed die casting production line monitoring dataset;

[0033] Further, according to the sensor monitoring compensation model, the e-th reconstructed die casting production line monitoring dataset is corrected for sensor monitoring to generate the e-th optimized die casting production line monitoring dataset, including:

[0034] According to the e-th reconstructed die casting production line monitoring dataset, a first production line monitoring data is randomly extracted; according to the production line sensor monitoring module, a first monitoring sensor equipment characteristic data corresponding to the first production line monitoring data is called; according to the sensor monitoring compensation model, a first sensor monitoring compensation data is obtained in combination with the first monitoring sensor equipment characteristic data; according to the first sensor monitoring compensation data, the first production line monitoring data is optimized to generate a first optimized production line monitoring data; the first optimized production line monitoring data is added to the e-th optimized die casting production line monitoring dataset; according to the sensor monitoring compensation model, the e-th reconstructed die casting production line monitoring dataset is iteratively corrected for sensor monitoring to obtain the e-th optimized die casting production line monitoring dataset.

[0035] A portion of data is randomly selected from the e-th reconstructed die casting production line monitoring data set as first production line monitoring data to ensure that different data points are covered in each iteration, thereby improving the accuracy of the overall data set. Using the production line sensor monitoring module, find the sensor devices associated with the first production line monitoring data, and call the feature data of these sensor devices, i.e. the first monitoring sensor device feature data, which includes sensor device status, sensor device environment, etc. According to the sensor monitoring compensation model, combined with the first monitoring sensor device feature data, the first sensor monitoring compensation data is calculated. According to the first sensor monitoring compensation data, the first production line monitoring data is optimized to generate the first optimized production line monitoring data. Add the first optimized production line monitoring data to the e-th optimized die casting production line monitoring data set. Repeat the above process, continue to extract data from the e-th reconstructed die casting production line monitoring data set, perform sensor monitoring correction, and add the optimized data to the e-th optimized die casting production line monitoring data set. After multiple iterations, the e-th optimized die casting production line monitoring data set finally obtained will contain a large amount of monitoring data that has been corrected by sensor monitoring and has higher quality.

[0036] Further, according to the sensor monitoring compensation model, combined with the first monitoring sensor device feature data, the first sensor monitoring compensation data is obtained, including:

[0037] According to the first monitoring sensor device feature data, abnormal detection is performed to obtain first sensor device abnormal feature data; according to the first sensor device abnormal feature data, monitoring loss depth evaluation is performed to obtain a first monitoring loss depth index; it is judged whether the first monitoring loss depth index is greater than / equal to a preset monitoring loss depth index; if the first monitoring loss depth index is greater than / equal to the preset monitoring loss depth index, the sensor monitoring compensation model is activated; the first sensor device abnormal feature data is input into the sensor monitoring compensation model to generate the first sensor monitoring compensation data.

[0038] The first monitoring sensor device feature data is subjected to anomaly detection by checking whether the data in the first monitoring sensor device feature data meets a preset threshold value, which is set based on historical data and professional knowledge, to determine whether there is abnormal feature data. Through anomaly detection, abnormal feature data that does not conform to the normal mode or exceeds the preset threshold value is identified, and these data are marked as first sensor device abnormal feature data. The monitoring loss depth evaluation is used to quantify the potential impact of abnormal feature data on the overall monitoring data quality and the operation of the production line. By analyzing factors such as the number, duration, and fluctuation range of abnormal feature data, combined with historical data of the production line operation and expert knowledge, a first monitoring loss depth index is calculated. The monitoring loss depth index is a quantitative indicator for evaluating the severity of abnormal feature data and the potential impact on the operation of the production line. The preset monitoring loss depth index is a threshold value set according to the operation of the production line and the quality requirements of the monitoring data. The first monitoring loss depth index is compared with the preset monitoring loss depth index to determine whether the severity of the abnormal feature data exceeds the preset threshold value. When the first monitoring loss depth index exceeds the preset threshold value, it is considered that the impact of abnormal feature data on the operation of the production line is relatively serious, and compensation measures need to be taken to activate the sensor monitoring compensation model to process abnormal feature data. The first sensor device abnormal feature data is input into the sensor monitoring compensation model. The sensor monitoring compensation model calculates the first sensor monitoring compensation data based on the input abnormal feature data, combined with the actual situation of the production line operation and prior knowledge. The first sensor monitoring compensation data is used to correct or modify the abnormal part of the original monitoring data to improve the accuracy and reliability of the monitoring data.

[0039] According to the die casting line anomaly detection module, the e-th optimized die casting line monitoring data set is subjected to deep anomaly detection to obtain an e-th die casting line anomaly coefficient.

[0040] The die casting line anomaly detection module is used to perform deep anomaly detection on the e-th optimized die casting line monitoring data set to obtain an e-th die casting line anomaly coefficient.

[0041] Further, according to the die casting line anomaly detection module, the e-th optimized die casting line monitoring data set is subjected to deep anomaly detection to obtain an e-th die casting line anomaly coefficient, which includes:

[0042] The die casting line anomaly detection module includes a monitoring backtracking module and a die casting line anomaly evaluation module; based on the monitoring backtracking module, a first e die casting line anomaly detection matrix is calculated according to the first e optimized die casting line monitoring data set; a first e die casting line anomaly evaluation unit in the die casting line anomaly evaluation module is activated, wherein the die casting line anomaly evaluation module includes E die casting line anomaly evaluation units corresponding to the E liquid valve die casting lines; the first e die casting line anomaly detection matrix is input into the first e die casting line anomaly evaluation unit, and J first e die casting line anomaly evaluation coefficients are output, wherein the first e die casting line anomaly evaluation unit includes J first e die casting line anomaly evaluation models corresponding to the first e liquid valve die casting line, and J is a positive integer greater than 1; a central value of the J first e die casting line anomaly evaluation coefficients is calculated to generate the first e die casting line anomaly coefficient.

[0043] The die casting line anomaly detection module includes a monitoring backtracking module and a die casting line anomaly evaluation module, the monitoring backtracking module is responsible for calculating an anomaly detection matrix according to an optimized monitoring data set, and the die casting line anomaly evaluation module includes a plurality of anomaly evaluation units for anomaly evaluation of specific die casting lines. The first e optimized die casting line monitoring data set is input into the monitoring backtracking module for calculation to obtain a first e die casting line anomaly detection matrix. The first e die casting line anomaly detection matrix is a matrix including a plurality of dimension information, used to describe the anomaly state of the line at different times, different positions or different parameters. The die casting line anomaly evaluation module includes E die casting line anomaly evaluation units, each unit corresponding to a liquid valve die casting line. The first e die casting line anomaly evaluation unit corresponding to the first e liquid valve die casting line is activated, and the first e die casting line anomaly detection matrix is input into the first e die casting line anomaly evaluation unit. The first e die casting line anomaly evaluation unit includes J anomaly evaluation models related to the first e liquid valve die casting line, and the anomaly evaluation models are constructed based on different evaluation standards. Each anomaly evaluation model outputs an anomaly evaluation coefficient according to the input anomaly detection matrix. Therefore, a total of J first e die casting line anomaly evaluation coefficients are output. The J first e die casting line anomaly evaluation coefficients are averaged to obtain a central value. The calculated central value is taken as the first e die casting line anomaly coefficient, which comprehensively reflects the anomaly state of the first e liquid valve die casting line.

[0044] Further, based on the monitoring backtracking module, a first e die casting line anomaly detection matrix is calculated according to the first e optimized die casting line monitoring data set, including:

[0045] obtain the e-th production line control scheme corresponding to the e-th liquid valve die casting production line; use the e-th production line control scheme as a data retrieval characteristic constraint; perform normal monitoring record retrieval on the e-th liquid valve die casting production line according to the data retrieval characteristic constraint and the monitoring backtracking module to obtain an e-th production line normal monitoring record retrieval set; integrate the e-th production line normal monitoring record retrieval set to construct an e-th die casting production line monitoring benchmark matrix; construct an e-th die casting production line monitoring matrix according to the e-th optimized die casting production line monitoring data set; and perform deviation analysis on the e-th die casting production line monitoring matrix based on the e-th die casting production line monitoring benchmark matrix to generate an e-th die casting production line anomaly detection matrix.

[0046] Preferably, the e-th production line control scheme corresponding to the e-th liquid valve die casting production line is obtained from the production line, and the e-th production line control scheme includes a series of operation processes, parameter settings, process requirements, etc. Key parameters and operation conditions in the e-th production line control scheme are used as data retrieval characteristic constraints. The monitoring backtracking module is a data retrieval center. According to the data retrieval characteristic constraints, normal monitoring record retrieval is performed on the e-th liquid valve die casting production line by using the monitoring backtracking module. During the retrieval process, normal monitoring records meeting the requirements of the e-th production line control scheme are screened out. The retrieved normal monitoring records are integrated into an e-th production line normal monitoring record retrieval set. The data in the e-th production line normal monitoring record retrieval set are arranged and analyzed, and an e-th die casting production line monitoring benchmark matrix is constructed according to the characteristics of the data (such as time, parameter value, monitoring point, etc.). The e-th die casting production line monitoring benchmark matrix represents the monitoring data distribution and characteristics of the production line in the normal running state. The e-th optimized die casting production line monitoring data set is used to construct an e-th die casting production line monitoring matrix according to the same data characteristics (such as time, parameter value, monitoring point, etc.). The e-th die casting production line monitoring matrix contains the monitoring data of the production line in actual operation. The e-th die casting production line monitoring matrix is compared and analyzed with the e-th die casting production line monitoring benchmark matrix, and the abnormal parts in the monitoring data are identified by calculating the differences, fluctuations and trends between the data, and the abnormal parts are integrated into an e-th die casting production line anomaly detection matrix. The e-th die casting production line anomaly detection matrix contains the abnormal conditions occurring in the running process of the production line.

[0047] According to the liquid valve die casting production line grading early warning module, the e-th die casting production line anomaly coefficient is used to perform abnormal early warning on the e-th liquid valve die casting production line.

[0048] When the e-th liquid valve die casting production line is abnormally warned by the liquid valve die casting production line grading warning module, the level of warning and the corresponding response measures are usually determined in combination with the value of the abnormal coefficient and the grading strategy of the warning module. Specifically, different levels of warning standards are pre-set in the liquid valve die casting production line grading warning module. These standards are usually based on the size of the abnormal coefficient, historical abnormal data, the importance of the production line operation and other factors. The warning levels can be divided into several grades from low to high, such as first-level warning (slight abnormality), second-level warning (moderate abnormality), third-level warning (serious abnormality), etc. According to the e-th die casting production line abnormal coefficient, the value size and the abnormal degree represented by it are evaluated. The e-th die casting production line abnormal coefficient is compared with the warning module set at each level of warning standard to determine the current warning level of the production line. If the abnormal coefficient is lower than the standard of first-level warning, the production line may be in a normal or slight abnormal state, at which time no warning or only low-level warning can be performed. If the abnormal coefficient reaches or exceeds the standard of first-level warning but is lower than the standard of second-level warning, the production line may be in a moderate abnormal state, at which time first-level warning should be triggered and corresponding preventive measures should be taken. If the abnormal coefficient reaches or exceeds the standard of second-level warning, the production line may be in a serious abnormal state, at which time higher-level warning (such as second-level or third-level warning) should be triggered, and the emergency response mechanism should be immediately started to ensure the safe and stable operation of the production line. According to the determined warning level, warning information is sent to relevant personnel through system notification, short message, email and other ways. The warning information should include the production line name, abnormal coefficient, warning level, suggested response measures and other contents, so that relevant personnel can quickly understand the production line state and take corresponding measures.

[0049] In summary, the embodiments of the present application have at least the following technical effects:

[0050] First, the aluminum alloy die casting production line information of the fire-fighting liquid valve is collected to obtain E liquid valve die casting production lines, wherein each liquid valve die casting production line corresponds to a liquid valve aluminum alloy die casting node, and E is a positive integer greater than 1. Next, according to the production line sensor monitoring module, E die casting production line monitoring data sets corresponding to the E liquid valve die casting production lines are obtained. Then, according to the E die casting production line monitoring data sets, the e th die casting production line monitoring data set corresponding to the e th liquid valve die casting production line is extracted, wherein e is a positive integer and 1≤e≤E. Next, the e th die casting production line monitoring data set is reconstructed and optimized according to the pre-processing loss function and the sensor monitoring compensation model to obtain the e th optimized die casting production line monitoring data set. Then, the e th optimized die casting production line monitoring data set is subjected to deep anomaly detection according to the die casting production line anomaly detection module to obtain the e th die casting production line anomaly coefficient. Finally, the e th liquid valve die casting production line is subjected to anomaly early warning according to the liquid valve die casting production line grading early warning module in combination with the e th die casting production line anomaly coefficient. The technical problem that the fire-fighting liquid valve production line monitoring data is inaccurate in the prior art, which leads to difficulty in accurately judging the production line running state and thus cannot timely early warn potential faults is solved, and the technical effect of improving the production stability and safety of the fire-fighting liquid valve is achieved.

[0051] Embodiment two, Figure 3 The structural schematic diagram of the electronic device provided for the embodiment two of the present application shows a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present application. Figure 3 The electronic device shown is only an example and should not impose any limitation on the functions and use range of the embodiment of the present application. As Figure 3 As shown, the electronic device includes a processor 31, a memory 32, an input device 33 and an output device 34; the number of processors 31 in the electronic device can be one or more, Figure 3 In an embodiment, the processor 31 in the electronic device, the memory 32, the input device 33 and the output device 34 can be connected through a bus or other means, Figure 3 In an embodiment, the connection through the bus is taken as an example.

[0052] The memory 32 as a kind of computer readable storage medium can be used to store software programs, computer executable programs and modules, such as the program instructions / modules of the anomaly early warning method of the liquid valve die casting production line in the embodiment of the present application. The processor 31 executes the software programs, instructions and modules stored in the memory 32 to perform various functional applications and data processing of the electronic device, i.e. to implement the anomaly early warning method of the liquid valve die casting production line described above.

[0053] It should be noted that the above-mentioned embodiment sequences of the present application are merely for description only, but not for representing the advantages and disadvantages of the embodiments. And the above-mentioned embodiments of the present specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0054] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0055] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that are within the scope of the present application should be included. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. An abnormality early warning method for a liquid valve die-casting production line, characterized in that, The method includes: Collect information on aluminum alloy die-casting production lines for fire-fighting liquid valves to obtain E liquid valve die-casting production lines, where each liquid valve die-casting production line corresponds to one liquid valve aluminum alloy die-casting node, and E is a positive integer greater than 1; According to the production line sensing and monitoring module, the monitoring datasets of the E die-casting production lines corresponding to the E liquid valve die-casting production lines are obtained. Based on the E die-casting production line monitoring datasets, extract the e-th die-casting production line monitoring dataset corresponding to the e-th liquid valve die-casting production line, where e is a positive integer, 1≤e≤E; The e-th die-casting production line monitoring dataset is reconstructed and optimized based on the preprocessing loss function and the sensor monitoring compensation model to obtain the e-th optimized die-casting production line monitoring dataset. The die casting production line anomaly detection module performs deep anomaly detection on the monitoring dataset of the e-th optimized die casting production line to obtain the anomaly coefficient of the e-th die casting production line. Based on the graded early warning module for liquid valve die casting production line, and combined with the abnormality coefficient of the eth die casting production line, an abnormality early warning is given for the eth liquid valve die casting production line. Specifically, the e-th die-casting production line monitoring dataset is reconstructed and optimized based on the preprocessing loss function and the sensor monitoring compensation model to obtain the e-th optimized die-casting production line monitoring dataset, including: The e-th die-casting production line monitoring dataset is preprocessed according to the preprocessing loss function to obtain the e-th reconstructed die-casting production line monitoring dataset. Based on the sensor monitoring compensation model, the eth reconstructed die casting production line monitoring dataset is corrected by sensor monitoring to generate the eth optimized die casting production line monitoring dataset. The preprocessing loss function is: ； Where n represents the index of the current data sample, and N represents the total number of data samples. Characterizing the monitoring dataset of the e-th die-casting production line, Characterizes the e-th reconstructed die-casting production line monitoring dataset. The preprocessing loss function characterizes the e-th die-casting production line monitoring dataset and the e-th reconstructed die-casting production line monitoring dataset, in order to The e-th die-casting production line monitoring dataset is preprocessed using the constraint of approaching 0 to obtain the e-th reconstructed die-casting production line monitoring dataset. The feature similarity between the e-th die-casting production line monitoring dataset and the e-th reconstructed die-casting production line monitoring dataset. The real-time similarity between the e-th die-casting production line monitoring dataset and the e-th reconstructed die-casting production line monitoring dataset is given. The preset similarity between the e-th die-casting production line monitoring dataset and the e-th reconstructed die-casting production line monitoring dataset; Specifically, the die-casting production line anomaly detection module performs deep anomaly detection on the monitoring dataset of the e-th optimized die-casting production line to obtain the anomaly coefficient of the e-th die-casting production line, including: The die-casting production line anomaly detection module includes a monitoring and backtracking module and a die-casting production line anomaly assessment module. Based on the monitoring backtracking module, the anomaly detection matrix of the e-th die-casting production line is calculated according to the monitoring dataset of the e-th optimized die-casting production line. Activate the eth die-casting production line anomaly assessment unit within the die-casting production line anomaly assessment module, wherein the die-casting production line anomaly assessment module includes E die-casting production line anomaly assessment units corresponding to the E liquid valve die-casting production lines; The abnormality detection matrix of the e-th die-casting production line is input into the abnormality evaluation unit of the e-th die-casting production line, and J abnormality evaluation coefficients of the e-th die-casting production line are output. The abnormality evaluation unit of the e-th die-casting production line includes J abnormality evaluation models of the e-th die-casting production line corresponding to the e-th liquid valve die-casting production line, where J is a positive integer greater than 1. Calculate the concentrated value of the Jth anomaly evaluation coefficient of the eth die-casting production line to generate the anomaly coefficient of the eth die-casting production line.

2. The abnormal early warning method for a liquid valve die-casting production line as described in claim 1, characterized in that, Based on the aforementioned sensor monitoring compensation model, the e-th reconstructed die-casting production line monitoring dataset is calibrated using sensor monitoring to generate the e-th optimized die-casting production line monitoring dataset, including: Randomly extract the first production line monitoring data based on the e-th reconstructed die-casting production line monitoring dataset; According to the production line sensing and monitoring module, retrieve the first monitoring sensing device feature data corresponding to the first production line monitoring data; Based on the aforementioned sensing and monitoring compensation model, and combined with the characteristic data of the first monitoring and sensing device, the first sensing and monitoring compensation data is obtained. The first production line monitoring data is optimized based on the first sensor monitoring compensation data to generate the first optimized production line monitoring data. Add the first optimized production line monitoring data to the e-th optimized die-casting production line monitoring dataset; Based on the sensor monitoring compensation model, iterative sensor monitoring correction is performed on the e-th reconstructed die-casting production line monitoring dataset to obtain the e-th optimized die-casting production line monitoring dataset.

3. The abnormal early warning method for a liquid valve die-casting production line as described in claim 2, characterized in that, Based on the aforementioned sensing and monitoring compensation model, and combined with the characteristic data of the first monitoring and sensing device, the first sensing and monitoring compensation data is obtained, including: Anomaly detection is performed based on the characteristic data of the first monitoring sensor to obtain the abnormal characteristic data of the first sensor. Based on the abnormal feature data of the first sensing device, the monitoring loss depth is assessed to obtain the first monitoring loss depth index; Determine whether the first monitoring loss depth index is greater than or equal to the preset monitoring loss depth index; If the first monitoring loss depth index is greater than or equal to the preset monitoring loss depth index, the sensor monitoring compensation model is activated. The abnormal feature data of the first sensing device is input into the sensing monitoring compensation model to generate the first sensing monitoring compensation data.

4. The abnormal early warning method for a liquid valve die-casting production line as described in claim 1, characterized in that, Based on the monitoring backtracking module, and according to the monitoring dataset of the e-th optimized die-casting production line, the anomaly detection matrix of the e-th die-casting production line is calculated, including: Obtain the control scheme for the eth production line corresponding to the eth liquid valve die-casting production line; The control scheme for the e-th production line is used as the data retrieval feature constraint; Based on the data retrieval feature constraints and the monitoring backtracking module, the normal monitoring records of the eth liquid valve die-casting production line are retrieved to obtain the normal monitoring record retrieval set of the eth production line. By integrating the normal monitoring record retrieval set of the e-th production line, a monitoring benchmark matrix for the e-th die-casting production line is constructed; Based on the e-th optimized die-casting production line monitoring dataset, construct the e-th die-casting production line monitoring matrix; Based on the monitoring benchmark matrix of the e-th die-casting production line, a deviation analysis is performed on the monitoring matrix of the e-th die-casting production line to generate the anomaly detection matrix of the e-th die-casting production line.

5. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the abnormal early warning method for the liquid valve die-casting production line according to any one of claims 1-4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the abnormal early warning method for the liquid valve die-casting production line as described in any one of claims 1-4.

Images