A pure benzene safe production information collection method and system

Abstract

This invention relates to the field of data storage structure technology, specifically to a method and system for collecting safety production information for pure benzene. Considering that abnormal production cycles require higher attention and are therefore of greater importance, the method first preliminarily identifies abnormal cycles based on production deviations to narrow the scope of analysis. Further, for all abnormal cycles, its continuity is analyzed, and influencing factors are analyzed based on this continuity. Furthermore, based on the temperature differences between different cycles, an importance index characterizing the degree of influence of different factors is determined. Based on this importance index, the corresponding data compression ratio is adaptively determined. Finally, safety production information for different production cycles is adaptively stored according to the data compression ratio. This approach reduces storage space while ensuring data accuracy, improves storage efficiency, and enhances the effectiveness of safety production information collection.

Description

Technical Field

[0001] This invention relates to the field of data storage structure technology, specifically to a method and system for collecting information on the safe production of pure benzene. Background Technology

[0002] As an important organic chemical raw material, benzene requires strict control over raw material quality, reaction conditions, and product quality during its production. For example, it is necessary to accurately control safety production information data such as reaction temperature, pressure, space velocity, and hydrogen-to-oil ratio, as well as indicators such as raw material composition and impurity content, to ensure that the raw materials meet the requirements of the production process and that the reaction takes place under optimal conditions. Furthermore, it is necessary to ensure the safety of the production process because benzene is highly toxic and a flammable liquid. Therefore, it is essential to strictly monitor various safety production information data during the benzene production process to ensure production safety and continuity.

[0003] For safety production information data generated during the production process, existing technologies typically use Discrete Cosine Transform (DCT) to compress and store all data at the same compression ratio. However, because these process parameter data need to be monitored for a long time during production, a large amount of monitoring data is generated. This means that using only DCT to compress all data at the same ratio results in excessive storage space consumption, reduced storage efficiency, and poor effectiveness in collecting safety production information. Summary of the Invention

[0004] To address the technical problem that using discrete cosine transform to compress all data at the same ratio results in excessive storage space consumption, reduced storage efficiency, and poor effectiveness of safety production information collection, this application aims to provide a method and system for collecting pure benzene safety production information. The specific technical solution adopted is as follows: The first aspect of this application provides a method for collecting information on the safe production of pure benzene, including: During the benzene production period, the actual benzene output for each production cycle and the corresponding furnace temperature data sequence in the catalytic reforming reaction stage were collected. Based on the deviation of the actual benzene production from the prior theoretical production, an anomaly index for each production cycle is determined; based on the continuous temporal distribution of the anomaly cycles selected by the anomaly index, cycles are merged to obtain all cycle anomaly groups; based on the changing trend of the anomaly index in the cycle anomaly groups and the number of anomaly cycles, the corresponding initial importance index is determined. Based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic anomaly group, all abnormal cycles are divided into at least two periodic anomaly segments; within the periodic anomaly group, the corresponding optimized importance index is determined based on the number of periodic anomaly segments, the differences in the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index. Based on the optimized importance index, the differences in furnace temperature data sequences between abnormal periods, and the duration of each abnormal period, the data compression ratio of each abnormal period is determined; and the safety production information data collected in each abnormal period is compressed and stored according to the data compression ratio.

[0005] Furthermore, the process of obtaining the anomaly index includes: The product of the preset error coefficient and the prior theoretical output for each production cycle is taken as the minimum error output. The difference between the minimum error output and the actual benzene output produced in each production cycle is used as the anomaly index for each production cycle.

[0006] Furthermore, the process of obtaining the periodic anomaly group includes: Adjacent abnormal cycles are merged to obtain all abnormal cycle groups; each abnormal cycle group consists of abnormal cycles, and the first production cycle before and the first production cycle after the abnormal cycle group are not abnormal cycles.

[0007] Furthermore, the process of obtaining the initial importance index includes: After arranging the abnormal indices of all abnormal periods in each abnormal period group in order, curve fitting is performed to obtain the abnormal index fitting curve; the slope of the tangent line at the coordinate point corresponding to the abnormal index fitting curve for each abnormal period is taken as the corresponding slope of change; the mean of the slopes of change of all abnormal periods is normalized to determine the trend growth reference value of each abnormal period group. The initial importance index for each periodic anomaly group is determined by multiplying the number of anomalous periods, the maximum anomalous index, and the trend growth reference value in each periodic anomaly group.

[0008] Furthermore, the process of obtaining the periodic abnormal segment includes: Obtain the adjacent periods that are adjacent to each abnormal period in each abnormal period group; The Dynamic Time Warping (DTW) distance between the furnace temperature data sequence of each abnormal cycle and the corresponding furnace temperature data sequence of each adjacent cycle is negatively correlated and normalized to determine the adjacent temperature similarity corresponding to each adjacent cycle. The adjacent cycles with adjacent temperature similarity greater than a preset similarity threshold are taken as the similar cycles of each abnormal cycle. Each abnormal cycle is merged with its corresponding similar cycle to obtain all the abnormal cycle segments.

[0009] Furthermore, the process of obtaining the optimization importance index includes: Calculate the DTW distance between the two furnace temperature data sequences corresponding to any two abnormal periods within the periodic anomaly group; take the mean of all DTW distances corresponding to the periodic anomaly group as the temperature sequence difference; The optimized importance index of the periodic anomaly group is obtained by multiplying the number of periodic anomaly segments in the periodic anomaly group, the temperature sequence difference, and the initial importance index.

[0010] Furthermore, the process of obtaining the data compression ratio includes: Each periodic anomaly segment in all periodic anomaly groups is taken as the target anomaly segment; each periodic anomaly segment other than the target anomaly segment is taken as the corresponding comparison anomaly segment. The DTW distance between the furnace temperature data sequence of each abnormal period in the target abnormal segment and the furnace temperature data sequence of each abnormal period in each comparison abnormal segment is taken as the corresponding comparison sequence difference; the mean of all comparison sequence differences between the target abnormal segment and each comparison abnormal segment is taken as the cluster distance between the target abnormal segment and each comparison abnormal segment; the cluster distance between each period abnormal segment and all other period abnormal segments is calculated; cluster analysis is performed based on the cluster distances between all period abnormal segments to obtain at least two period clusters; The sequence of periodic anomalies is obtained by arranging all periodic anomalies in chronological order; the range of index values ​​of all periodic anomalies in each periodic cluster in the sequence of periodic anomalies is used as the corresponding distribution range reference value; the product between the number of anomalies in each periodic anomaly segment and the optimization importance index of the periodic anomaly group in which it belongs is used as the corresponding local influence degree. By performing a negative correlation mapping between the cumulative value of the local impact of all periodic outlier segments and the product of the distribution range reference value, the data compression ratio of all periodic outlier segments in each periodic cluster is determined.

[0011] Furthermore, the process of compressing and storing the safety production data collected in each periodic anomaly segment according to the data compression ratio includes: Based on the data compression ratio, the safety production information data collected in each abnormal period of each abnormal segment is compressed and stored using the discrete cosine transform function.

[0012] Furthermore, the process of obtaining the abnormal period includes: Production cycles with an abnormality index greater than 0 are considered abnormal cycles.

[0013] Secondly, this application provides a pure benzene safety production information collection system, the system comprising: The data acquisition module is used to collect the actual benzene output and the corresponding furnace temperature data sequence in each production cycle during the benzene production period. The first determining module is used to determine the anomaly index of each production cycle based on the deviation of the actual pure benzene production from the prior theoretical production; to merge the anomaly cycles selected by the anomaly index in the time sequence to obtain all the cycle anomaly groups; and to determine the corresponding initial importance index based on the changing trend of the anomaly index in the cycle anomaly group and the number of anomaly cycles. The second determining module is used to divide all abnormal cycles into at least two abnormal segments based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic abnormality group; and to determine the corresponding optimized importance index in the periodic abnormality group based on the number of periodic abnormal segments, the differences in the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index. The safety production information collection module is used to determine the data compression ratio of each cycle anomaly segment based on the optimized importance index, the difference in heating furnace temperature data sequence between cycle anomaly segments, and the time length of each cycle anomaly segment; and to compress and store the safety production information data collected in each cycle anomaly segment according to the data compression ratio.

[0014] Thirdly, this application provides a computer device including a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to perform the method as described in the first aspect of this application or any embodiment of the first aspect.

[0015] Fourthly, this application provides a computer program product comprising computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0016] Fifthly, this application provides a computer-readable storage medium that stores computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0017] This application has the following beneficial effects: Considering the varying importance of safety production information data across different production batches or cycles in the benzene production process, when using discrete cosine transform (DCT) for data compression, the data compression ratio can be determined based on the importance of data from different production cycles. This allows for different levels of compression for different production cycles, reducing storage space while maintaining data accuracy. For importance calculation, production cycles exhibiting anomalies require higher attention and are therefore more important. Thus, abnormal cycles are initially identified based on production deviations to narrow the analysis scope. Further analysis of the continuity of all abnormal cycles is conducted, influencing factors are analyzed based on this continuity, and importance indices characterizing the impact of different factors are determined based on temperature differences between cycles. The corresponding data compression ratio is then adaptively determined based on this importance index. Finally, safety production information from different production cycles is adaptively stored according to the data compression ratio, ensuring data accuracy while reducing storage space, improving storage efficiency, and enhancing the effectiveness of safety production information collection. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for collecting safety production information for pure benzene, provided in one embodiment of the present invention. Figure 2 This is a structural diagram of a pure benzene safety production information collection system provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of a computer device structure provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method and system for collecting information on safe production of pure benzene according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] The following description, in conjunction with the accompanying drawings, details the specific scheme of the pure benzene safety production information collection method and system provided by the present invention.

[0023] This application provides a method for collecting information on the safe production of pure benzene. Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of a method for collecting safety production information for pure benzene according to an embodiment of the present invention. The method includes: Step S101: During the benzene production period, collect the actual benzene output for each production cycle and the corresponding furnace temperature data sequence in the catalytic reforming reaction stage.

[0024] There are various processes for producing pure benzene, including catalytic reforming, hydrocracking, and coking crude benzene refining. The principles and processes of different methods are different, and the relevant key parameters will also vary. This invention uses catalytic reforming as an example for data collection. When producing pure benzene using catalytic reforming, it mainly includes three stages: feedstock pretreatment, catalytic reforming reaction, and aromatic separation. Among them, the catalytic reforming reaction stage needs to be carried out under specific high temperature and high pressure conditions to carry out reactions such as cycloalkane dehydrogenation and alkane cyclization dehydrogenation to generate reformate oil rich in aromatics. The temperature of this stage needs to be 480-520℃ and the pressure 1.5-2.5 MPa. The entire process generally takes 2-3 days or longer to complete.

[0025] In one specific implementation of this invention, the benzene production time period is set to two months prior to the current time. During the benzene production time period, the process method for producing benzene remains unchanged, and the purity of raw materials, raw material weight, operation, and equipment remain unchanged for each production cycle. Therefore, the time required for each production cycle is also approximately consistent.

[0026] When monitoring temperature, it is first necessary to determine the key locations for temperature monitoring based on the characteristics of the catalytic reforming process and equipment layout. These include, for example, different axial and radial positions of the reactor, the inlet and outlet of the heater, the hot and cold sides of the heat exchanger, and the bearings of pumps and compressors. Temperature changes at these locations directly reflect the operating status of the production process. Temperature monitoring equipment can include thermocouples, resistance temperature detectors (RTDs), thermometers, and temperature sensors. The specific temperature monitoring locations and equipment can be selected according to the specific implementation environment. In one specific implementation of this invention, a thermocouple is installed at the heater outlet to collect the heater temperature at each sampling moment during the catalytic reforming reaction stage. Due to the high temperature in the monitoring environment of this invention, thermocouples are used for temperature measurement and monitoring. The sampling frequency is set to once every 10 minutes, which can be adjusted. The heater temperature data collected in each production cycle are arranged in chronological order to obtain the required new sequence of heater temperature data. The actual benzene production is detected using a high-precision digital weighing scale. Furthermore, in one specific implementation of this invention, the safety production information data for each production cycle includes reaction temperature, pressure, space velocity, and hydrogen-to-oil ratio, all collected every 10 minutes, which will not be elaborated further here.

[0027] Step S102: Determine the anomaly index for each production cycle based on the deviation of the actual benzene production from the theoretical prior production; merge the cycles based on the continuous temporal distribution of the anomaly cycles selected by the anomaly index to obtain all the cycle anomaly groups; determine the corresponding initial importance index based on the changing trend of the anomaly index in the cycle anomaly group and the number of anomaly cycles.

[0028] Based on the stoichiometric relationships of chemical reactions and the law of conservation of mass, under a specific production process, there is a definite stoichiometric relationship between a certain mass and purity of raw materials and the yield of pure benzene. Therefore, when the production process, raw material purity, operation, and equipment conditions are the same, the yield of pure benzene obtained for a certain mass of raw materials is also constant. Ideally, based on the amount of hydrocarbons in the raw materials that can be converted into benzene, the theoretically producible mass of pure benzene can be precisely calculated. As long as the production process conditions are stable, the raw materials react completely, and there are no other side reactions or losses, the corresponding mass of pure benzene should be obtained, i.e., the a priori theoretical yield. However, in actual production, due to some factors that are difficult to completely control, the actual pure benzene yield is usually lower than the a priori theoretical yield corresponding to this theoretical value, and the error is usually between 5% and 15%. It should be noted that since the purity and mass of the raw materials are the same in each production cycle of this application, the a priori theoretical mass is the same for all production cycles. The magnitude of the a priori theoretical mass is related to the specific purity and mass of the raw materials in the specific implementation environment, and the calculation of the a priori theoretical mass is a technique well-known to those skilled in the art, and will not be further limited or elaborated upon here. Furthermore, it should be noted that when the prior theoretical quality differs in different production cycles within the specific implementation environment, the prior theoretical quality needs to be recalculated based on the purity and quality of the raw materials for each production cycle. This will not be elaborated further here.

[0029] Given that the error rate is typically between 5% and 15%, the actual benzene production yield is usually 85% to 95% of the theoretical yield. Excessive error indicates a potential anomaly in the benzene production process, thus requiring close monitoring of the corresponding production cycle. Preferably, in some possible implementations of this invention, the process for obtaining the anomaly index includes: The product of a preset error coefficient and the theoretical output for each production cycle is taken as the minimum error output. The difference between the minimum error output and the actual benzene output for each production cycle is taken as the anomaly index for each production cycle. In one specific implementation of this invention, the preset error coefficient is set to 0.9. The larger the anomaly index, the lower the actual benzene output, and the more likely a production anomaly is to occur. In another specific implementation of this invention, production cycles with an anomaly index greater than 0 are considered abnormal cycles. That is, this application considers an anomaly in the corresponding production cycle to have occurred when the actual benzene output is less than 90% of the theoretical output, requiring close attention. Output below 85% is usually considered a more serious production problem, and production is typically stopped and adjusted when this occurs. Therefore, it is usually impossible to collect data on actual benzene output below 85% of the theoretical output.

[0030] In one specific implementation of this invention, the process of obtaining the anomaly index is expressed by the formula: ;in, For the first Anomaly index for each production cycle; For the first Actual pure benzene production per production cycle; The preset error coefficient; For the first The prior theoretical output per production cycle; For the first Minimum error output per production cycle.

[0031] When producing pure benzene using catalytic reforming, even if the purity and quality of raw materials, as well as operating and equipment conditions remain constant, other factors may still lead to lower yields in certain reaction cycles under the same conditions. These factors include catalyst performance degradation, deviations from optimal reaction conditions such as temperature and pressure, temporary fluctuations in external energy supply, ambient temperature and humidity, and potential equipment problems. The distribution of various abnormal cycles within the total data collection time varies. They may be scattered individual abnormal cycles or multiple abnormal cycles may be continuously distributed. Scattered abnormal cycles may be caused by a single, incidental factor, while continuously distributed abnormal cycles may be caused by the sustained influence of one or more common factors. Therefore, continuously distributed abnormal cycles should be given greater importance during data collection.

[0032] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the periodic anomaly group includes: Adjacent abnormal cycles are merged to obtain all abnormal cycle groups; each abnormal cycle group consists entirely of abnormal cycles, and neither the first production cycle before nor the first production cycle after the abnormal cycle group is an abnormal cycle; that is, continuously distributed abnormal cycles are merged into one abnormal cycle group, and independently distributed abnormal cycles are also considered as one abnormal cycle group for analysis.

[0033] Different periodic abnormality groups have different continuity, that is, they contain different numbers of abnormal periods. The abnormality index of each abnormal period and the changing trend of the abnormality index within the group may also be different. Therefore, the importance of each continuous abnormality group is determined based on the continuity of each periodic abnormality group and the changing trend of the abnormality index within the group.

[0034] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the initial importance index includes: After sequentially arranging the abnormal indices of all abnormal periods in each abnormal period group, curve fitting is performed to obtain the abnormal index fitting curve; the slope of the tangent line at the corresponding coordinate point on the abnormal index fitting curve for each abnormal period is taken as the corresponding slope of change; the mean of the slopes of change of all abnormal periods is normalized to determine the trend growth reference value for each abnormal period group; the initial importance index of each abnormal period group is determined based on the product of the number of abnormal periods, the maximum value of the abnormal index, and the trend growth reference value in each abnormal period group.

[0035] On the anomaly index fitting curve, when the mean slope of all abnormal cycles is greater than 0, it indicates that the anomaly index of all abnormal cycles in the abnormal cycle group shows an overall upward trend. The larger the trend growth reference value, the more obvious the overall upward trend of the anomaly index, reflecting that the anomalies in each abnormal cycle are becoming more and more serious, and the degree of influence from the abnormal factors is gradually increasing. Therefore, the more attention needs to be paid to this abnormal cycle group, the greater its importance will be. The more abnormal cycles there are, the greater the continuity of this abnormal cycle group, and the longer the production anomaly lasts. Therefore, greater importance is needed to pay attention to these abnormal cycles. The maximum value of the anomaly index can numerically characterize the degree of abnormality of this abnormal cycle group affected by abnormal factors. Therefore, the maximum value of the anomaly index is also used to characterize the initial importance index.

[0036] In one specific implementation of this invention, the process of obtaining the initial importance index is expressed by the formula: ;in, For the first The initial importance index of each abnormal cycle group; For the first The number of abnormal cycles in each abnormal cycle group; For the first The maximum value of the abnormal index of each abnormal cycle group; For the first In the abnormal index fitting curve of the abnormal period group, the first... The slope of the abnormal cycle; For the first Reference values ​​for the trend growth of each abnormal cycle group; It is a linear normalization function.

[0037] Step S103: Based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic anomaly group, divide all abnormal cycles into at least two periodic anomaly segments; within the periodic anomaly group, determine the corresponding optimized importance index based on the number of periodic anomaly segments, the differences in the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index.

[0038] For each abnormal cycle group, there are various non-accidental factors causing abnormal production, so the causes of abnormalities in each abnormal cycle group are not entirely the same. That is, each abnormal cycle group may be caused by a single factor persisting for a certain period, such as reduced catalyst performance, temperature or pressure deviating from the normal range, etc., or it may be caused by the combined influence of multiple factors, such as both temperature or pressure deviating from the normal range and potential equipment failure. Furthermore, the occurrence time and duration of different factors vary; different factors may occur individually or one or more factors may occur consecutively. Therefore, it is necessary to analyze the specific impact within each abnormal cycle group. Because the catalytic reforming reaction for producing pure benzene requires specific high-temperature and high-pressure conditions, the temperature will be affected to varying degrees by different factors, thus affecting the reaction conditions of the catalytic reforming reaction and leading to abnormalities. Therefore, the influencing factors can be determined by analyzing the temperature of each abnormal cycle.

[0039] Since one or more factors typically persist for a period of time, their impact manifests in the periodic anomaly group as similar temperature changes within consecutive anomaly cycles. Therefore, further analysis of the similarity in furnace temperature data sequences between adjacent anomaly cycles is performed to divide periodic anomaly segments based on different temperature influencing factors. Preferably, in some possible implementations of this invention, the process of obtaining periodic anomaly segments includes: The process involves obtaining adjacent cycles within each abnormal cycle group; using a dynamic time warping algorithm, negatively normalizing the DTW distance between the furnace temperature data sequence of each abnormal cycle and the corresponding adjacent cycle's furnace temperature data sequence to determine the adjacent temperature similarity for each adjacent cycle; identifying adjacent cycles with an adjacent temperature similarity greater than a preset similarity threshold as similar cycles for each abnormal cycle; and merging each abnormal cycle with its corresponding similar cycle to obtain all abnormal cycle segments. In one specific implementation of this invention, the preset similarity threshold is set to 0.6, which can be adjusted according to the specific implementation environment. For each abnormal cycle, the smaller the DTW distance between its furnace temperature data sequence and its corresponding adjacent cycle, the more similar the furnace temperature changes between the corresponding abnormal cycles are, and the more likely they belong to production cycles influenced by the same temperature influencing factor. Therefore, this characteristic is used to merge abnormal cycles, ensuring that each obtained abnormal cycle segment corresponds to the same temperature influencing factor. It should be noted that the merging of abnormal cycles here is a diffusion-type process. For example, when four abnormal cycles, A, B, C, and D, are distributed sequentially, if A and C are similar cycles to B, and D is a similar cycle to C, then A, B, C, and D are considered as one abnormal cycle segment. In addition, the dynamic time warping algorithm is a well-known technique in the art, and will not be further limited or elaborated here.

[0040] In one specific implementation of this invention, the process of obtaining adjacent temperature similarity is expressed by the formula: ;in, For the first In the abnormal period group, the first Each abnormal cycle and its corresponding Adjacent temperature similarity between adjacent cycles; For the first In the abnormal period group, the first The furnace temperature data sequence of each abnormal cycle and its corresponding The DTW distance between adjacent furnace temperature data sequences; because abnormal cycles that are not at the beginning or end correspond to two adjacent abnormal cycles, each abnormal cycle usually corresponds to two adjacent cycles, while those at the beginning or end of a cycle abnormal group correspond to only one; it should be noted that for a cycle abnormal group with only one abnormal cycle, it is itself considered as a cycle abnormal segment.

[0041] During the production process, there are many factors that have a long-term and continuous impact on the output of pure benzene, such as heating equipment failure, pressure control equipment failure, parameter adjustment errors, and catalyst performance degradation. The number of influencing factors contained in the periodic anomaly group varies, and the degree of influence of different factors on the output also varies. Therefore, the initial importance index of each group is updated based on the number of periodic anomaly segments within the periodic anomaly group and the differences between periodic anomaly segments.

[0042] Preferably, in some possible implementations of the embodiments of the present invention, the process of optimizing the acquisition of the importance index includes: Calculate the DTW distance between the two furnace temperature data sequences corresponding to any two abnormal periods within the periodic anomaly group; take the mean of all DTW distances corresponding to the periodic anomaly group as the temperature sequence difference; obtain the optimized importance index of the periodic anomaly group based on the product of the number of periodic anomaly segments in the periodic anomaly group, the temperature sequence difference, and the initial importance index.

[0043] For each periodic anomaly group, the more periodic anomaly segments it contains, the more types of factors are affected by temperature, and the more attention needs to be paid to the anomalies in the data within that group, thus increasing its importance. Conversely, the greater the DTW distance between the furnace temperature data sequences of different anomaly cycles within a periodic anomaly group, the more diverse the temperature distribution across different production cycles within that group, and the greater the variety of influencing factors, thus requiring greater attention to the data anomalies. Therefore, the initial importance index is weighted by combining the number of periodic anomaly segments and the temperature sequence differences within the periodic anomaly group, resulting in a more accurate optimized importance index.

[0044] In one specific implementation of this invention, the process of obtaining the importance index is expressed by the following formula: ;in, For the first The importance index for optimizing the abnormal groups in each cycle; For the first The initial importance index of each periodic anomaly group; For the first The number of periodic abnormal segments in each periodic abnormality group; For the first The mean DTW distance between any two pairs of furnace temperature data sequences in all anomalous periods within the anomalous period group, i.e., the nth period... Temperature sequence differences in a periodic anomaly group.

[0045] Step S104: Determine the data compression ratio of each cycle anomaly segment based on the optimization importance index, the difference in heating furnace temperature data sequence between cycle anomaly segments, and the time length of each cycle anomaly segment; compress and store the safety production information data collected in each cycle anomaly segment according to the data compression ratio.

[0046] In all the periodic anomaly groups, the lengths of different periodic anomaly segments are different, which means that the duration of the influence of different influencing factors is different, and the distribution of anomaly segments within the group is also different. It is possible that anomalies caused by different influencing factors appear interspersed, or that the same influencing factor appears repeatedly. Therefore, it is necessary to determine the importance of each influencing factor or combination of influencing factors based on the length of the periodic anomaly segments and the distribution range between periodic anomaly segments of the same influencing factor, so as to determine the data compression ratio of each periodic anomaly segment according to the importance.

[0047] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the data compression ratio includes: Each periodic anomaly segment in all periodic anomaly groups is sequentially designated as the target anomaly segment; each periodic anomaly segment outside the target anomaly segment is designated as the corresponding comparison anomaly segment; the DTW distance between the furnace temperature data sequence of each anomaly period in the target anomaly segment and the furnace temperature data sequence of each anomaly period in each comparison anomaly segment is used as the corresponding comparison sequence difference; the mean of all comparison sequence differences between the target anomaly segment and each comparison anomaly segment is used as the cluster distance between the target anomaly segment and each comparison anomaly segment; the cluster distance between each periodic anomaly segment and all other periodic anomaly segments is calculated; cluster analysis is performed based on the cluster distances between all periodic anomaly segments to obtain at least two periodic clusters. For both the target anomaly segment and the comparison anomaly segment, if the overall DTW distance of the furnace temperature data sequences for each anomaly period is small, it indicates that the temperature change trends between the target anomaly segment and the comparison anomaly segment are more similar, meaning they are more likely to be affected by the same temperature influencing factor or a combination of temperature influencing factors. Therefore, a smaller clustering distance is preferred to group them into the same period cluster. Based on this clustering distance calculation principle, each period cluster corresponds to each period anomaly segment under the influence of one temperature influencing factor or a combination of temperature influencing factors. In this embodiment of the invention, k-means clustering analysis is used, and the number of clusters is set to 6, which can be adjusted according to the specific implementation environment.

[0048] In one specific implementation of this invention, the process of obtaining the clustering distance is expressed by the formula: ;in, For the target abnormal segment Corresponding anomaly segment Cluster distance between them; For the target abnormal segment Corresponding anomaly segment The number of corresponding sequence differences; For the target abnormal segment Corresponding anomaly segment The corresponding number Differences in the compared sequences.

[0049] The periodic anomaly segments are arranged in chronological order to obtain the periodic anomaly segment sequence. The range of the index values ​​of all periodic anomaly segments in the periodic anomaly segment sequence in each periodic cluster is used as the corresponding distribution range reference value. The product between the number of anomaly periods in each periodic anomaly segment and the optimization importance index of its periodic anomaly group is used as the corresponding local influence degree. The product between the cumulative value of the local influence degree of all periodic anomaly segments and the distribution range reference value is negatively correlated to determine the data compression ratio of all periodic anomaly segments in each periodic cluster.

[0050] For the temperature-influencing factors of each periodic cluster, the wider the temporal distribution of all corresponding periodic anomalies, the more widespread the temporal influence of that temperature-influencing factor. Therefore, the more important and influential the temperature-influencing factor of the corresponding periodic cluster is. Furthermore, the more anomalous periods there are in each periodic anomaly segment within a periodic cluster, the longer the duration of that periodic anomaly segment's influence. Combining this with the calculated optimization importance index, the local influence degree of the influencing factor representing each periodic anomaly segment is determined. Therefore, for each periodic cluster, the greater the overall local influence degree and the larger the reference value of the distribution range of its corresponding periodic anomalies, the greater the influence of the temperature-influencing factor corresponding to that periodic cluster. The more important the anomalous periods within these periodic anomalies are, the more important they are. Thus, a smaller data compression ratio is needed to preserve the authenticity of this important data.

[0051] In one specific implementation of this invention, the process of obtaining the data compression ratio is expressed by the formula: ;in, For the first The data compression ratio of all periodic outliers in each periodic cluster; For the first The range of index values ​​of all periodic outliers in a periodic cluster in the periodic outlier sequence, i.e., the reference value of the distribution range; For the first The number of periodic outliers in each periodic cluster; For the first In the periodic cluster, the first The number of abnormal cycles in each abnormal segment; For the first In the periodic cluster, the first The importance index for optimizing abnormal segments in each cycle; For the first In the periodic cluster, the first The degree of local impact of each periodic abnormal segment; The preset correction factor is set to 9 in this embodiment of the invention, so that the compression ratio of this embodiment is set to between 1:1 and 10:1. The magnitude of the preset correction factor can be adjusted according to the specific implementation environment, which will not be elaborated further here. The data compression ratio obtained is expressed as the ratio between the number and 1.

[0052] Preferably, in some possible implementations of the embodiments of the present invention, the process of compressing and storing the safety production data collected in each periodic anomaly segment includes: The safety production information data collected in each abnormal period within each periodic abnormal segment is compressed and stored using the Discrete Cosine Transform (DCT) function according to the data compression ratio. When using the DCT function for data compression, the compression ratio is a crucial indicator that directly reflects the data compression effect. Determining the compression ratio is paramount; a higher compression ratio means saving more storage space and transmission bandwidth, but usually comes with some information loss, leading to a decrease in the quality of the reconstructed data. A lower compression ratio retains more original data information, resulting in higher reconstruction quality, but with relatively higher storage space and transmission costs. Therefore, compression using an adaptively calculated compression ratio allows the collected data to reduce storage space while maintaining authenticity, improving storage efficiency and enhancing the effectiveness of safety production information collection. It should be noted that safety production information data from production periods outside of abnormal periods are compressed and stored at a compression ratio of 10:1, which can be adjusted according to the specific implementation environment. It should also be noted that using the DCT function for data compression is a technique well-known to those skilled in the art, and will not be further limited or elaborated upon here.

[0053] In summary, this method for collecting safety production information in pure benzene takes into account the higher attention and importance required for abnormal production cycles. Therefore, it first identifies abnormal cycles based on production deviations to narrow the analysis scope. Further, for all abnormal cycles, its continuity is analyzed, and influencing factors are identified based on this continuity. Furthermore, based on temperature differences between different cycles, an importance index characterizing the degree of influence of different factors is determined. The corresponding data compression ratio is then adaptively determined based on this importance index. Finally, safety production information for different production cycles is adaptively stored according to the data compression ratio. This approach reduces storage space while ensuring data accuracy, improves storage efficiency, and enhances the effectiveness of safety production information collection.

[0054] This application also provides a pure benzene safety production information collection system; please refer to [link / reference]. Figure 2 The diagram shows a structural diagram of a benzene safety production information collection system according to an embodiment of the present invention. The system includes: a data collection module 201, a first determination module 202, a second determination module 203, and a safety production information collection module 204.

[0055] The data acquisition module 201 is used to collect the actual pure benzene output and the corresponding furnace temperature data sequence in the catalytic reforming reaction stage for each production cycle during the pure benzene production period. The first determining module 202 is used to determine the anomaly index of each production cycle based on the deviation of the actual pure benzene production from the prior theoretical production; to merge the cycles based on the continuous distribution of the anomaly cycles selected by the anomaly index in the time series, and to obtain all the cycle anomaly groups; and to determine the corresponding initial importance index based on the changing trend of the anomaly index in the cycle anomaly group and the number of anomaly cycles. The second determining module 203 is used to divide all abnormal cycles into at least two abnormal segments based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic abnormality group; and to determine the corresponding optimized importance index in the periodic abnormality group based on the number of periodic abnormal segments, the differences between the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index. The safety production information collection module 204 is used to determine the data compression ratio of each cycle anomaly segment based on the optimization importance index, the difference in heating furnace temperature data sequence between cycle anomaly segments, and the time length of each cycle anomaly segment; and to compress and store the safety production information data collected in each cycle anomaly segment according to the data compression ratio.

[0056] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the pure benzene safety production information collection system and the pure benzene safety production information collection method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0057] This application also provides a computer device; please refer to [link / reference]. Figure 3 The diagram illustrates a computer device structure according to an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned methods for collecting information on the safe production of pure benzene.

[0058] This application also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned methods for collecting information on safe production of pure benzene.

[0059] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned methods for collecting information on safe production of pure benzene.

[0060] In the embodiments provided in this application, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to perform the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0061] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for collecting safety production information for pure benzene, characterized in that, The method includes: During the benzene production period, the actual benzene output for each production cycle and the corresponding furnace temperature data sequence in the catalytic reforming reaction stage were collected. Based on the deviation of the actual benzene production from the prior theoretical production, an anomaly index for each production cycle is determined; based on the continuous temporal distribution of the anomaly cycles selected by the anomaly index, cycles are merged to obtain all cycle anomaly groups; based on the changing trend of the anomaly index in the cycle anomaly groups and the number of anomaly cycles, the corresponding initial importance index is determined. Based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic anomaly group, all abnormal cycles are divided into at least two periodic anomaly segments; within the periodic anomaly group, the corresponding optimized importance index is determined based on the number of periodic anomaly segments, the differences in the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index. Based on the optimized importance index, the differences in furnace temperature data sequences between abnormal periods, and the duration of each abnormal period, the data compression ratio of each abnormal period is determined; and the safety production information data collected in each abnormal period is compressed and stored according to the data compression ratio.

2. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the abnormal index includes: The product of the preset error coefficient and the prior theoretical output for each production cycle is taken as the minimum error output. The difference between the minimum error output and the actual benzene output produced in each production cycle is used as the anomaly index for each production cycle.

3. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the periodic anomaly group includes: Adjacent abnormal cycles are merged to obtain all abnormal cycle groups; each abnormal cycle group consists of abnormal cycles, and the first production cycle before and the first production cycle after the abnormal cycle group are not abnormal cycles.

4. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the initial importance index includes: After arranging the abnormal indices of all abnormal periods in each abnormal period group in order, curve fitting is performed to obtain the abnormal index fitting curve; the slope of the tangent line at the coordinate point corresponding to the abnormal index fitting curve for each abnormal period is taken as the corresponding slope of change; the mean of the slopes of change of all abnormal periods is normalized to determine the trend growth reference value of each abnormal period group. The initial importance index for each periodic anomaly group is determined by multiplying the number of anomalous periods, the maximum anomalous index, and the trend growth reference value in each periodic anomaly group.

5. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the periodic abnormal segment includes: Obtain the adjacent periods that are adjacent to each abnormal period in each abnormal period group; The Dynamic Time Warping (DTW) distance between the furnace temperature data sequence of each abnormal cycle and the corresponding furnace temperature data sequence of each adjacent cycle is negatively correlated and normalized to determine the adjacent temperature similarity corresponding to each adjacent cycle. The adjacent cycles with adjacent temperature similarity greater than a preset similarity threshold are taken as the similar cycles of each abnormal cycle. Each abnormal cycle is merged with its corresponding similar cycle to obtain all the abnormal cycle segments.

6. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the optimization importance index includes: Calculate the DTW distance between the two furnace temperature data sequences corresponding to any two abnormal periods within the periodic anomaly group; take the mean of all DTW distances corresponding to the periodic anomaly group as the temperature sequence difference; The optimized importance index of the periodic anomaly group is obtained by multiplying the number of periodic anomaly segments in the periodic anomaly group, the temperature sequence difference, and the initial importance index.

7. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the data compression ratio includes: Each periodic anomaly segment in all periodic anomaly groups is taken as the target anomaly segment; each periodic anomaly segment other than the target anomaly segment is taken as the corresponding comparison anomaly segment. The DTW distance between the furnace temperature data sequence of each abnormal period in the target abnormal segment and the furnace temperature data sequence of each abnormal period in each comparison abnormal segment is taken as the corresponding comparison sequence difference; the mean of all comparison sequence differences between the target abnormal segment and each comparison abnormal segment is taken as the cluster distance between the target abnormal segment and each comparison abnormal segment; the cluster distance between each period abnormal segment and all other period abnormal segments is calculated; cluster analysis is performed based on the cluster distances between all period abnormal segments to obtain at least two period clusters; The sequence of periodic anomalies is obtained by arranging all periodic anomalies in chronological order; the range of index values ​​of all periodic anomalies in each periodic cluster in the sequence of periodic anomalies is used as the corresponding distribution range reference value; the product between the number of anomalies in each periodic anomaly segment and the optimization importance index of the periodic anomaly group in which it belongs is used as the corresponding local influence degree. By performing a negative correlation mapping between the cumulative value of the local impact of all periodic outlier segments and the product of the distribution range reference value, the data compression ratio of all periodic outlier segments in each periodic cluster is determined.

8. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of compressing and storing the safety production data collected in each periodic anomaly segment according to the data compression ratio includes: Based on the data compression ratio, the safety production information data collected in each abnormal period of each abnormal segment is compressed and stored using the discrete cosine transform function.

9. The method for collecting safety production information for pure benzene according to claim 1, characterized in that, The process of obtaining the abnormal period includes: Production cycles with an abnormality index greater than 0 are considered abnormal cycles.

10. A pure benzene safety production information collection system, characterized in that, The system includes: The data acquisition module is used to collect the actual benzene output and the corresponding furnace temperature data sequence in each production cycle during the benzene production period. The first determining module is used to determine the anomaly index of each production cycle based on the deviation of the actual pure benzene production from the prior theoretical production; to merge the anomaly cycles selected by the anomaly index in the time sequence to obtain all the cycle anomaly groups; and to determine the corresponding initial importance index based on the changing trend of the anomaly index in the cycle anomaly group and the number of anomaly cycles. The second determining module is used to divide all abnormal cycles into at least two abnormal segments based on the similarity of the furnace temperature data sequences between adjacent abnormal cycles in the periodic abnormality group; and to determine the corresponding optimized importance index in the periodic abnormality group based on the number of periodic abnormal segments, the differences in the corresponding furnace temperature data sequences between each abnormal cycle, and the initial importance index. The safety production information collection module is used to determine the data compression ratio of each cycle anomaly segment based on the optimized importance index, the difference in heating furnace temperature data sequence between cycle anomaly segments, and the time length of each cycle anomaly segment; and to compress and store the safety production information data collected in each cycle anomaly segment according to the data compression ratio.

Images