A data optimization method for a foundry plant monitoring system

Abstract

This invention relates to a data optimization method for a foundry monitoring system, comprising a monitoring center and a data optimization module. The data optimization module utilizes a comprehensive, end-to-end optimization approach covering data acquisition, preprocessing, transmission, storage, and display to reduce data redundancy. The data optimization module is electrically connected to the monitoring center's cache module, data extraction module, and data conversion module, and also communicates bidirectionally with the human-machine interface display screen. The end-to-end optimization approach specifically includes: reducing the amount of source data through differentiated sampling strategies; streamlining invalid data through multi-level dynamic filtering; balancing real-time performance and bandwidth usage through a tiered transmission mechanism; separating hot and cold data storage to avoid storage accumulation; and finally, reducing hardware load through dynamic refresh at the display end. This invention, through its end-to-end optimization scheme, reduces data redundancy at the source, streamlines invalid information during transmission and processing, and matches data characteristics during storage and display.

Description

Technical Field

[0001] This invention relates to the field of foundry monitoring technology, and more specifically to a data optimization method for a foundry monitoring system. Background Technology

[0002] As a fundamental process in the equipment manufacturing industry, casting is a core means of producing key components for automobiles, aerospace, and construction machinery. The quality of its products directly determines the performance and reliability of downstream equipment. In the casting production process, the metal melting process in the melting furnace and the workpiece tempering process in the heat treatment furnace are two core stages. The melting furnace requires precise control of parameters such as melting temperature and holding time to ensure uniform metal composition and the absence of impurities. The heat treatment furnace, on the other hand, needs to set stepped temperature curves according to different casting materials, optimizing the mechanical properties (such as hardness and toughness) of the castings through precise control of heating, holding, and cooling. Deviations in the operating parameters of these two stages can lead to minor issues such as dimensional inaccuracies and surface defects, or even serious problems like cracking and scrapping, resulting in significant economic losses. Therefore, real-time monitoring and precise control of their operating status are crucial.

[0003] The applicant disclosed an intelligent monitoring system for a foundry in Chinese patent CN121115650A. During its application, the temperature of the smelting furnace and heat treatment furnace changes in real time, resulting in a large amount of temperature data being transmitted to the monitoring system. Although the patented solution processes this data before displaying it on the screen, the still large volume of temperature data causes data congestion during transmission, leading to prolonged periods of stagnation and stuttering on the display. Summary of the Invention

[0004] The main objective of this invention is to provide a data optimization method for a foundry monitoring system. This method designs a full-link optimization scheme to reduce data redundancy at the source, streamline invalid information during transmission and processing, and match data characteristics during storage and display, thereby ensuring the stable operation of the monitoring system.

[0005] To achieve the above objectives, this invention provides a data optimization method for a foundry monitoring system. The monitoring system includes a data acquisition terminal, a PLC control system, a monitoring center, and a human-machine interface. The data acquisition terminal includes a smelting furnace control system and a heat treatment furnace control system. The smelting furnace control system can acquire and control the temperature of the smelting furnace, while the heat treatment furnace control system is used to control and acquire the temperature of the heat treatment furnace. Each heat treatment furnace has at least two temperature detection devices inside its furnace cavity, and each furnace cavity is equipped with an independent data acquisition module. The data acquisition module is used to acquire signals from the temperature detection devices and transmit them to the monitoring center. The monitoring center also includes a cache module, a pre-stored data module, a data extraction module, a data conversion module, a graph comparison module, and a database. The cache module stores the data uploaded by the data acquisition terminal in real time. The data extraction module extracts temperature data. The data conversion module converts the temperature data into a temperature curve graph. The graph comparison module compares whether the temperature curve is within a preset range. The monitoring center also includes a data optimization module, which is electrically connected to the monitoring center's caching module, data extraction module, and data conversion module, and communicates bidirectionally with the display screen of the human-machine interface. The data optimization module optimizes the entire data acquisition, preprocessing, transmission, storage, and display process sequentially using technologies that interconnect each link. Each link forms a closed-loop collaboration through parameter linkage. Simultaneously, it quantifies and controls the full-link data redundancy using a full-link data redundancy rate calculation formula, which is:

[0006] in, The end-to-end data redundancy rate represents the proportion of invalid data that ultimately enters the display after sampling, filtering, transmission, and storage. The sampling redundancy rate is determined by the technical characteristics of the data acquisition process. The redundancy rate at the filtering end is determined by the technical characteristics of the data preprocessing stage. The redundancy rate at the transmission end is determined by the technical characteristics of the data transmission process. The redundancy rate of the storage end is determined by the technical characteristics of the data storage process; The specific technical measures across the entire supply chain include: The system reduces the amount of source data through a differentiated sampling strategy, simplifies invalid data through multi-level dynamic filtering, balances real-time performance and bandwidth usage through a hierarchical transmission mechanism, avoids storage accumulation by separating hot and cold data, and finally reduces hardware load through dynamic refresh at the display end.

[0007] Preferably, the differentiated sampling strategy uses a dynamic sampling frequency formula to adjust the sampling frequency, collecting only valid temperature data. The dynamic sampling frequency formula is:

[0008] in, This is the real-time sampling frequency, in units of ; This is the upper limit of the sampling frequency. ; The base sampling frequency is ; This is the proportionality coefficient, and its value range is... ; The absolute rate of change of temperature, in units of ; The valid data collection rules are: The data acquisition module has a signal filtering function to remove abnormal sensor fluctuation values ​​that change by more than 5°C in a single instance and then return within 1 second. Only valid temperature data is transmitted.

[0009] A further preferred approach is to use multi-level dynamic filtering, which includes local moving average filtering at the data acquisition end and pre-selection of trend feature points at the monitoring center. The moving average filtering formula is as follows:

[0010] in, For the first The filtered data, For the first One original sampled data, The window size is dynamically sliding, and hour, ,otherwise ; The trend feature point selection rules are as follows: The data after three consecutive filters satisfy At that time, only the first data is retained, and the rest are marked as redundant and discarded. The data extraction module has a temperature data targeted filtering function to optimize data processing efficiency.

[0011] Furthermore, the hierarchical transmission mechanism divides the preprocessed data into critical data and general data. Critical data is transmitted in real time, while general data is transmitted in batches. The batch transmission cycle formula is as follows:

[0012] in, This refers to the batch transmission cycle, in units of... ; The base cycle, for ; ; The key data judgment criteria are: The temperature exceeds the range of the standard process temperature curve processed by the pre-stored data module, or the temperature change rate... The graphical comparison module determines data priority based on the temperature curve comparison results.

[0013] In a further optimized approach, in the cold and hot data separation storage, hot data is stored in a high-speed cache, retaining a certain amount of data. For cold data, differential compression is used for storage, and the storage formula is as follows:

[0014] in, The stored temperature difference value; The cache cleanup mechanism is as follows: The cache data volume has reached the threshold When processing a data entry, the earliest entry in the ordinary data is cleaned up first. The caching module works in conjunction with the database to store the collected data and comparison results in the database in real time.

[0015] Furthermore, the dynamic refresh of the display adopts a dual-screen split-screen display logic. Display screen one refreshes only when the standard curve is updated or an anomaly occurs, with a refresh rate ≤0.1Hz. Display screen two dynamically adjusts its refresh rate, using the following formula:

[0016] in, For dynamic refresh rate, ; The rule for prioritizing the display of anomalies is as follows: Upon receiving an abnormal signal, display screen 1 stops displaying standard values ​​and switches to real-time updates of abnormal temperatures. Display screen 2 highlights the abnormal segment curve. The pre-stored data module and the graphic comparison module are both electrically connected to the display screens, and the displayed content is switched according to the operating status.

[0017] More preferably, the data optimization module includes a sampling control unit, a filtering processing unit, a transmission scheduling unit, a storage management unit, and a display control unit, which respectively execute differentiated sampling strategies, multi-level dynamic filtering, hierarchical transmission mechanisms, separate storage of hot and cold data, and dynamic refresh of the display. The units are electrically connected to each other. The sampling control unit communicates bidirectionally with the data acquisition module of the acquisition end to receive raw data from the temperature detection device. The display control unit is electrically connected to the display screen of the human-machine interface to control the split-screen display and refresh rate. The display screen is used to display temperature values ​​and temperature curves.

[0018] Furthermore, the PLC control system can remotely modify the operating parameters of the smelting furnace control system and the heat treatment furnace control system, and the modified parameters are synchronously fed back to the databases of the corresponding control systems and the monitoring center; the pre-stored data module of the monitoring center stores the standard process temperature curves of the smelting furnace and the heat treatment furnace, and the steps for processing the standard process temperature curves include: extracting the standard temperature of each step segment. ,right conduct Correct the temperature profile and draw the process temperature range diagram based on the corrected temperature.

[0019] The beneficial effects of this invention are as follows: This invention employs a differentiated sampling strategy with dynamic sampling frequency to achieve precise matching between sampling frequency and temperature change rate. This avoids data explosion caused by fixed high-frequency sampling during the heating and cooling stages, and maintains basic low-frequency sampling during the heat preservation stage, thereby reducing the proportion of invalid data from the source.

[0020] The multi-level dynamic filtering mechanism smooths high-frequency fluctuating data through local moving average filtering at the acquisition end, and then removes duplicate trend data through trend feature point screening at the monitoring center. After dual optimization, the proportion of effective data is increased, avoiding the waste of computing power caused by the monitoring center to process massive redundant data, and ensuring the real-time response of core functions such as temperature curve conversion and comparison.

[0021] The hierarchical transmission mechanism distinguishes between critical data and ordinary data, enabling differentiated transmission logic for real-time transmission of critical data and batch transmission of ordinary data. The batch transmission cycle formula adapts to the characteristics of data changes, ensuring low-delay transmission of critical information such as out-of-range temperature and abnormal rate of change, while avoiding bandwidth congestion caused by high-frequency transmission of full data, thereby improving transmission efficiency.

[0022] In the hot and cold data separation storage mode, hot data is limited to high-frequency real-time data, while cold data significantly reduces storage usage through differential compression formulas. Combined with a cache cleanup mechanism, ordinary data resources are released first. Compared with the existing mixed storage mode, this reduces storage space, improves data reading speed, avoids end-to-end response lag caused by storage accumulation, and ensures the long-term stable operation of the monitoring system.

[0023] The dual-display dynamic refresh solution adapts to data characteristics through differentiated refresh logic. Display screen one (static screen) refreshes at a low frequency to avoid unnecessary load, while display screen two (dynamic screen) dynamically adjusts the refresh rate to ensure smooth display of temperature curves during the heating and cooling phases, while avoiding hardware overload caused by fixed high-frequency refresh. This reduces the CPU usage of the display screen, reduces blue screen occurrences, and ensures that operators can monitor the equipment's operating status in real time. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] This embodiment is based on an existing intelligent monitoring system for foundries, specifically an improvement upon patent CN121115650A. This intelligent monitoring system includes a data acquisition terminal, a PLC control system, a monitoring center, and a human-machine interface. This invention achieves end-to-end data optimization by adding a data optimization module to the monitoring center and electrically connecting it to the existing module. It is suitable for production monitoring scenarios involving various smelting furnaces and multi-cavity heat treatment furnaces. Furthermore, this embodiment, with appropriate adaptation, can also be applied to other systems and is not limited to patent CN121115650A.

[0026] Specifically, the data acquisition end includes a smelting furnace control system and a heat treatment furnace control system, which are responsible for temperature acquisition and control of the smelting furnace and heat treatment furnace, respectively. The heat treatment furnace in this embodiment is the one disclosed in patent CN120591538A. Because this furnace is equipped with numerous temperature detection devices, it generates a large amount of temperature data that needs to be processed. Specifically, each furnace cavity of the heat treatment furnace is equipped with at least two temperature detection devices to ensure the comprehensiveness and accuracy of temperature data acquisition. Each furnace cavity is equipped with an independent data acquisition module. This data acquisition module establishes bidirectional communication with the temperature detection devices and the data optimization module of the monitoring center, undertaking the functions of temperature signal acquisition, preliminary screening, and transmission.

[0027] The monitoring center includes a caching module, a pre-stored data module, a data extraction module, a data conversion module, a graphical comparison module, a database, and a newly added data optimization module. The pre-stored data module stores standard process temperature curves for melting furnaces and heat treatment furnaces adapted to different casting materials, and performs a processing flow of extracting and correcting standard temperature steps and drawing process temperature curve range diagrams on the standard process temperature curves; the caching module is used to store the raw data uploaded by the acquisition terminal in real time, and the database is responsible for long-term storage of processed data and comparison results.

[0028] The data optimization module consists of a sampling control unit, a filtering processing unit, a transmission scheduling unit, a storage management unit, and a display control unit. These units are electrically connected to achieve continuous data transmission and processing. The sampling control unit communicates with the data acquisition module at the acquisition end to receive raw temperature data; the display control unit is electrically connected to the dual displays at the human-machine interface to adjust the display mode and refresh rate.

[0029] The human-computer interaction terminal adopts a dual-screen design. Screen 1 is used for static data display, and screen 2 is used for dynamic temperature curve display. Both screens support display mode switching based on system operating status and maintain electrical connection with the pre-stored data module and graphic comparison module of the monitoring center.

[0030] The monitoring center also includes a data optimization module. This module optimizes the entire data acquisition, preprocessing, transmission, storage, and display process sequentially using technologies that reduce data redundancy during the heating and cooling phases, preventing display lag and blue screens. The data optimization module is electrically connected to the monitoring center's cache module, data extraction module, and data conversion module, and communicates bidirectionally with the human-machine interface display. The specific technologies used in the entire process include: reducing the amount of source data through differentiated sampling strategies, simplifying invalid data through multi-level dynamic filtering, balancing real-time performance and bandwidth usage through a tiered transmission mechanism, separating hot and cold data storage to avoid storage accumulation, and finally reducing hardware load through dynamic refresh at the display end.

[0031] In this embodiment, the data optimization module quantifies and adjusts the proportion of invalid data in each stage through the end-to-end data redundancy rate calculation formula, realizing parameter linkage and closed loop in the sampling, filtering, transmission, and storage stages. At the same time, it quantifies the source of display terminal CPU utilization through the display terminal hardware load correlation formula, realizing the load distribution of the display terminal by the preceding stages. The two formulas together constitute the quantitative core of end-to-end collaborative optimization. The specific formulas and correlation explanations are as follows:

[0032] in, The end-to-end data redundancy rate represents the proportion of invalid data that ultimately enters the display after sampling, filtering, transmission, and storage.

[0033] The sampling redundancy rate is determined by the technical characteristics of the data acquisition process. Specifically, it is determined by the dynamic sampling frequency formula of the differentiated sampling strategy and the effective data filtering rules, including the real-time sampling frequency. and They are negatively correlated. The sampling redundancy rate refers to the proportion of invalid sampled data to the total sampled data after dynamic sampling and outlier filtering at the acquisition end. Invalid data includes two categories: abnormal fluctuation values ​​of the sensor and redundant sampled data.

[0034] The redundancy rate at the filtering end is determined by the technical characteristics of the data preprocessing stage; specifically, it is determined by the moving average and moving window of the multi-stage dynamic filtering. The sliding window is determined by the trend feature point filtering rules. and They are negatively correlated. The redundancy rate at the filter end refers to the proportion of data marked as redundant and discarded after the data acquisition end's moving average filtering and the monitoring center's trend feature point screening, out of the total filtered data.

[0035] The redundancy rate at the transmission end is determined by the technical characteristics of the data transmission link. Specifically, it is determined by the batch transmission cycle formula of the hierarchical transmission mechanism and the key data judgment criteria. and They are negatively correlated. The transmission redundancy rate refers to the proportion of invalid data in the batch transmission of ordinary data after being filtered by the hierarchical transmission mechanism (critical data is transmitted in real time, has no redundancy, and is not included in the calculation).

[0036] Storage redundancy is determined by the technical characteristics of the data storage process, specifically by the cache threshold for separating hot and cold data. Decision, cache threshold and They are positively correlated. Storage redundancy rate refers to the proportion of invalid data in the total stored data in the storage system after hot and cold data separation and cache cleanup mechanisms. Invalid data includes ordinary data that exceeds the cache threshold and cold data fragments that have no traceability value.

[0037] In this embodiment, each unit of the data optimization module collects its own operational parameters in real time, and the storage management unit automatically calculates the data redundancy rate across the entire link. When detected At this time, the system will automatically fine-tune the parameters of the preceding stages, such as reducing the proportional coefficient k to lower the sampling frequency and increasing the sliding window. Enhance the filtering effect to ensure that the data redundancy rate of the entire link is stably controlled below 12%, and achieve dynamic collaborative optimization of each link from a quantitative perspective.

[0038] Specifically, the differentiated sampling strategy adjusts the sampling frequency through a dynamic sampling frequency formula while simultaneously implementing effective data filtering rules to reduce data redundancy at the source. Details are as follows: 1. Dynamic adjustment of sampling frequency: Based on the current absolute temperature change rate of the melting furnace and heat treatment furnace Substitute into the dynamic sampling frequency formula Determine the real-time sampling frequency. When the temperature change rate is high during the heating and cooling phases, the sampling frequency should be adjusted synchronously with the change rate, but should not exceed the set upper limit. To avoid data explosion caused by fixed high-frequency sampling; the temperature change rate is low during the heat preservation stage, and the basic sampling frequency is used. Sampling is performed to balance data integrity and redundancy control. In the above formula, This is the real-time sampling frequency, in units of ; This is the upper limit of the sampling frequency. ; The base sampling frequency is ; This is the proportionality coefficient, and its value range is... ; The absolute rate of change of temperature, in units of .

[0039] Valid data filtering: The data acquisition module identifies abnormal fluctuations in sensor values ​​in real time, and discards data that meet the characteristic of "a single sudden change exceeding 5°C and subsequent return within 1 second". Only the effective temperature data that meets the process monitoring requirements is transmitted to the monitoring center, further reducing the amount of invalid data transmission.

[0040] 2. Implementation of multi-stage dynamic filtering Multi-level dynamic filtering includes two levels: local moving average filtering at the acquisition end and pre-processing trend feature point filtering at the monitoring center, achieving dual data simplification.

[0041] Local moving average filtering at the acquisition end: Based on the rate of change of absolute temperature Dynamically adjust the size of the sliding window ,when When using a smaller window It balances data trend preservation with volatility suppression; when When using a larger window This enhances the data smoothing effect. The continuously collected raw data is substituted into the moving average filtering formula. This yields smoothed data after filtering.

[0042] Pre-monitoring trend feature point filtering in the monitoring center: Perform continuous trend analysis on the filtered data. When three consecutive data points meet the following conditions... When data is identified as having a recurring trend, only the first data point is retained, while the remaining data is marked as redundant and discarded. The targeted filtering function of the data extraction module improves the efficiency of subsequent data processing.

[0043] 3. Implementation of hierarchical transmission mechanism The hierarchical transmission mechanism executes differentiated transmission strategies based on data priority differences, balancing transmission real-time performance with bandwidth utilization efficiency.

[0044] Data priority determination: The graphical comparison module, combined with the standard process temperature curve range processed by the pre-stored data module, prioritizes the filtered data. Data exceeding the standard curve range or with a high temperature change rate is prioritized. The data in question is classified as critical data, while the rest is classified as ordinary data.

[0045] Differential transmission execution: Critical data is transmitted in real-time to ensure that important information such as temperature anomalies is transmitted without delay; general data is transmitted in batches according to the rate of temperature change. Substitute into the batch transfer cycle formula

[0046] Dynamically adjust the transmission cycle to avoid bandwidth congestion caused by high-frequency transmission of full data.

[0047] 4. Implementation of separate storage for hot and cold data Separating hot and cold data storage optimizes storage resource allocation through differentiated storage strategies, avoiding storage accumulation and read latency.

[0048] Hot data storage: The cache only retains frequently used real-time hot data, and the amount of data retained is calculated according to the formula. This ensures that real-time data from the current equipment operation can be quickly accessed, guaranteeing timely monitoring responses. hour, .

[0049] Cold data storage: Historical data exceeding real-time usage requirements is treated as cold data and stored using differential compression. The continuously filtered data is then substituted into the storage formula.

[0050] It stores only temperature difference values ​​and baseline data, significantly reducing storage space usage.

[0051] Cache cleanup mechanism: When the cache data volume reaches a set threshold When a record is retrieved, the system automatically performs cache cleanup, prioritizing the release of the earliest entry in ordinary data while retaining critical data and recent real-time data to ensure sufficient cache space and maintain data reading efficiency.

[0052] 5. Dynamic refresh implementation on the display end Dynamic refresh on the display is based on a dual-screen split-screen design. It reduces hardware load through differentiated refresh logic, avoiding lag and blue screen issues.

[0053] Normal operating status refresh: Display screen one (static screen) refreshes only when the standard process temperature profile is updated, and the refresh rate is strictly controlled within [specific parameters]. To avoid unnecessary refreshes consuming resources; Display screen two (dynamic screen) adjusts according to temperature change rate. Substitute into the refresh rate formula

[0054] The refresh rate is dynamically adjusted to balance the temperature curve display of smoothness and hardware load. In the above formula, For dynamic refresh rate, .

[0055] Abnormal running status refresh: When the monitoring center sends an abnormal signal, display screen one stops displaying standard values ​​and switches to real-time abnormal temperature update mode; display screen two automatically increases the refresh rate to the base value. It also highlights the temperature curve of abnormal segments to ensure that operators can quickly detect abnormal information and take timely control measures.

[0056] The collaborative workflow of each unit in the data optimization module of this embodiment is as follows: The sampling control unit receives the raw temperature signal transmitted by the data acquisition module, filters valid data according to the differentiated sampling strategy, removes abnormal fluctuation values, and sends the valid data to the filtering processing unit. The filtering unit first smooths the data using a moving average filter, then simplifies repetitive trend data according to the trend feature point filtering rules, and synchronously distributes the optimized data to the transmission scheduling unit and the storage management unit. The transmission scheduling unit combines the curve comparison results from the graphic comparison module to determine the data priority. Key data is transmitted to the display control unit in real time, while ordinary data is transmitted in batches according to the dynamically adjusted cycle. The storage management unit stores real-time data in a high-speed cache as hot data, periodically compresses historical data and transfers it to the database as cold data, and executes a cleanup mechanism based on cache usage. The display control unit adjusts the refresh rate and display content of the dual displays according to the data type and system operating status, and triggers display mode switching in abnormal conditions to ensure the monitoring visualization effect and system stability.

[0057] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A data optimization method for a foundry monitoring system, the monitoring system comprising a data acquisition terminal, a PLC control system, a monitoring center, and a human-machine interface terminal, wherein the data acquisition terminal includes a smelting furnace control system and a heat treatment furnace control system, the smelting furnace control system being able to acquire and control the temperature of the smelting furnace, the heat treatment furnace control system being used to control and acquire the temperature of the heat treatment furnace, each heat treatment furnace having at least two temperature detection devices inside its furnace cavity, and each furnace cavity having an independent data acquisition module outside, the data acquisition module being used to acquire signals from the temperature detection devices and transmit them to the monitoring center; the monitoring center further comprising a cache module, a pre-stored data module, a data extraction module, a data conversion module, a graphic comparison module, and a database, the cache module being used to store data uploaded by the data acquisition terminal in real time, the data extraction module being used to extract temperature data, the data conversion module being used to convert the temperature data into a temperature curve graph, and the graphic comparison module being used to compare whether the temperature curve is within a preset range, characterized in that, The monitoring center also includes a data optimization module, which is electrically connected to the monitoring center's caching module, data extraction module, and data conversion module, and communicates bidirectionally with the display screen of the human-machine interface. The data optimization module optimizes the entire data acquisition, preprocessing, transmission, storage, and display process sequentially using technologies that interconnect each link. Each link forms a closed-loop collaboration through parameter linkage. Simultaneously, it quantifies and controls the full-link data redundancy using a full-link data redundancy rate calculation formula, which is: ， In the formula, The end-to-end data redundancy rate represents the proportion of invalid data that ultimately enters the display after sampling, filtering, transmission, and storage. The sampling redundancy rate is determined by the technical characteristics of the data acquisition process. The redundancy rate at the filtering end is determined by the technical characteristics of the data preprocessing stage. The redundancy rate at the transmission end is determined by the technical characteristics of the data transmission process. The redundancy rate of the storage end is determined by the technical characteristics of the data storage process; The specific technical measures across the entire supply chain include: The system reduces the amount of source data through a differentiated sampling strategy, simplifies invalid data through multi-level dynamic filtering, balances real-time performance and bandwidth usage through a hierarchical transmission mechanism, avoids storage accumulation by separating hot and cold data, and finally reduces hardware load through dynamic refresh at the display end.

2. A data optimization method for a foundry monitoring system according to claim 1, characterized in that, The differentiated sampling strategy uses a dynamic sampling frequency formula to adjust the sampling frequency, collecting only valid temperature data. The dynamic sampling frequency formula is: ， In the formula, This is the real-time sampling frequency, in units of ; This is the upper limit of the sampling frequency. ; The base sampling frequency is ; This is the proportionality coefficient, and its value range is... ; The absolute rate of change of temperature, in units of ; The valid data collection rules are: The data acquisition module has a signal filtering function to remove abnormal sensor fluctuation values ​​that change by more than 5°C in a single instance and then return within 1 second. Only valid temperature data is transmitted.

3. A data optimization method for a foundry monitoring system according to claim 2, characterized in that, Multi-level dynamic filtering includes local moving average filtering at the data acquisition end and pre-processing trend feature point screening at the monitoring center. The moving average filtering formula is: ， In the formula, For the first The filtered data, For the first One original sampled data, The window size is dynamically sliding, and hour, ,otherwise ; The trend feature point selection rules are as follows: The data after three consecutive filters satisfy At that time, only the first data is retained, and the rest are marked as redundant and discarded. The data extraction module has a temperature data targeted filtering function to optimize data processing efficiency.

4. A data optimization method for a foundry monitoring system according to claim 3, characterized in that, The hierarchical transmission mechanism divides the preprocessed data into critical data and general data. Critical data is transmitted in real time, while general data is transmitted in batches. The formula for the batch transmission cycle is: ， In the formula, This refers to the batch transmission cycle, in units of... ; The base cycle, for ; ； The key data judgment criteria are: The temperature exceeds the range of the standard process temperature curve processed by the pre-stored data module, or the temperature change rate... The graphical comparison module determines data priority based on the temperature curve comparison results.

5. A data optimization method for a foundry monitoring system according to claim 4, characterized in that, In hot and cold data separation storage, hot data is stored in a cache, retaining a certain amount of data. For cold data, differential compression is used for storage, and the storage formula is as follows: ， In the formula, The stored temperature difference value; The cache cleanup mechanism is as follows: The cache data volume has reached the threshold When processing a data entry, the earliest entry in the ordinary data is cleaned up first. The caching module works in conjunction with the database to store the collected data and comparison results in the database in real time.

6. A data optimization method for a foundry monitoring system according to claim 5, characterized in that, The display dynamically refreshes using a dual-screen split-screen logic. Screen one refreshes only when the standard curve is updated or an anomaly occurs, with a refresh rate ≤0.1Hz. Screen two dynamically adjusts its refresh rate using the following formula: ， In the formula, For dynamic refresh rate, ; The rule for prioritizing the display of anomalies is as follows: Upon receiving an abnormal signal, display screen 1 stops displaying standard values ​​and switches to real-time updates of abnormal temperatures. Display screen 2 highlights the abnormal segment curve. The pre-stored data module and the graphic comparison module are both electrically connected to the display screens, and the displayed content is switched according to the operating status.

7. A data optimization method for a foundry monitoring system according to claim 6, characterized in that, The data optimization module includes a sampling control unit, a filtering processing unit, a transmission scheduling unit, a storage management unit, and a display control unit. These units respectively execute differentiated sampling strategies, multi-level dynamic filtering, hierarchical transmission mechanisms, separate storage of hot and cold data, and dynamic refresh of the display. All units are electrically connected. The sampling control unit communicates bidirectionally with the data acquisition module at the acquisition end to receive raw data from the temperature detection device. The display control unit is electrically connected to the display screen at the human-machine interface, controlling the split-screen display and refresh rate. The display screen is used to display temperature values ​​and temperature curves.

8. A data optimization method for a foundry monitoring system according to any one of claims 1-7, characterized in that, The PLC control system can remotely modify the operating parameters of the smelting furnace control system and the heat treatment furnace control system. The modified parameters are synchronously fed back to the databases of the corresponding control systems and the monitoring center. The pre-stored data module of the monitoring center stores the standard process temperature curves of the smelting furnace and the heat treatment furnace. The steps for processing the standard process temperature curves include: extracting the standard temperature of each step segment. ,right conduct Correct the temperature profile and draw the process temperature range diagram based on the corrected temperature.

Images