An intelligent monitoring method and system for abnormal reaction parameters of antioxidants

Abstract

The present invention relates to the field of data processing technology, and more specifically, to an antioxidant reaction parameter abnormality intelligent monitoring method and system, the method comprising: constructing a curve graph of the inhibition efficiency of the antioxidant in a historical period before the current moment, calculating the proportion of data with a negative slope in the statistical curve graph, calculating the accuracy of the change trend of the inhibition efficiency in the historical period according to the data proportion and the standard deviation of the inhibition efficiency in the historical period, calculating the abnormal degree of the inhibition efficiency of the antioxidant at the current moment in combination with the accuracy and the difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, and monitoring the abnormality of the reaction parameters of the antioxidant according to the normalized value of the product of the abnormal degree and the abnormal score of the inhibition efficiency at the current moment and the comparison result of the preset threshold value. The present invention can improve the accuracy of monitoring the abnormal situation of the reaction parameters of the antioxidant.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to an abnormal intelligent monitoring method and system for reaction parameters of an antioxidant. Background Art

[0002] In the modern industrial field, polymer materials are widely used in various products due to their unique properties. During the production process of polymer materials, oxidation reactions will significantly reduce the strength, heat resistance and weather resistance of the materials. Therefore, antioxidants are usually added during the production process to inhibit the oxidation reaction of the materials. However, the complexity of the production process will affect the reaction parameters of the antioxidants, and then affect their activity and production efficiency. Therefore, in order to ensure production efficiency, it is very important to conduct abnormal intelligent monitoring of the reaction parameters of antioxidants.

[0003] Robust Random Cut Forest (RRCF) is an algorithm used for anomaly detection and unsupervised learning. It is suitable for processing large-scale data, has strong noise resistance and real-time detection capabilities, and is suitable for anomaly detection of reaction parameters of antioxidants.

[0004] In the related art, for example, an application document with publication number CN113779103A discloses a method and device for detecting abnormal data, the method comprising: constructing a sequence sample set of the commodity to be detected within a preset time length; using a preset RRCF model, based on the sequence sample set, determining the abnormality index of the target sequence sample and the reference sequence sample, and determining the abnormality score of the target sequence sample based on the abnormality index of the target sequence sample and the reference sequence sample; in response to determining that the abnormality score is greater than a preset score threshold, determining that abnormal data exists in the data stream.

[0005] However, the above scheme does not take into account the impact of data change trends on anomaly detection results, resulting in certain misjudgments of anomaly results identified by the algorithm, and may fail to accurately monitor data anomalies. Summary of the invention

[0006] In order to solve the problem of abnormal situations in which data cannot be accurately monitored, the present invention provides an abnormal intelligent monitoring method and system for reaction parameters of an antioxidant.

[0007] According to a first aspect of the present invention, there is provided a method for intelligently monitoring abnormal reaction parameters of an antioxidant, comprising:

[0008] Obtaining the initial content of oxides in the reactor, calculating the difference between the initial content and the oxide content at any time after the addition of the antioxidant, and taking the ratio of the difference to the initial content as the inhibition efficiency of the antioxidant at any time;

[0009] Construct a curve graph of the inhibition efficiency of the antioxidant in the historical period before the current moment, calculate the percentage of data with a negative slope in the curve graph, calculate the normalized value of the product of the percentage of data and the standard deviation of the inhibition efficiency in the historical period, and use the difference between 1 and the normalized value as the accuracy of the variation trend of the inhibition efficiency in the historical period;

[0010] Calculate the abnormal degree of the inhibition efficiency of the antioxidant at the current moment, and the abnormal degree is positively correlated with the difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, as well as the accuracy;

[0011] The RRCF algorithm is used to detect anomalies in the inhibition efficiency at the current moment, and the anomaly score of the inhibition efficiency at the current moment is obtained. The anomaly of the reaction parameters of the antioxidant is monitored based on the normalized value of the product of the anomaly degree and the anomaly score and the comparison result with the preset threshold.

[0012] The present invention determines the abnormal degree of the inhibition efficiency at the current moment in combination with the accuracy of the changing trend of the inhibition efficiency in the historical period, thereby ensuring the accuracy of the abnormal degree, and uses the abnormal degree to correct the abnormal score of the inhibition efficiency at the current moment determined based on the RRCF algorithm, which can reduce the situation where the algorithm cannot pay attention to the changing trend of the data and causes misjudgment of the data, thereby improving the accuracy of abnormal detection of the inhibition efficiency, and further accurately monitoring the abnormal conditions of the reaction parameters of the antioxidant.

[0013] Preferably, the accuracy of the variation trend of the suppression efficiency in the historical period satisfies the following relationship:

[0014] ;

[0015] In the formula, The accuracy of the change trend of the inhibition efficiency of the antioxidant in the historical period before the current moment; It is the first antioxidant in history. the efficiency of suppression at each historical moment; is the average value of the inhibition efficiency of the antioxidant in the historical period; is the number of data points with negative slope in the curve graph of inhibition efficiency in the historical period; is the number of historical moments in the historical period; is a hyperparameter; is the normalization function.

[0016] The present invention limits the range of accuracy to 0-1, which can reduce the difficulty of calculation.

[0017] Preferably, the method for obtaining the curve graph includes:

[0018] A curve graph of the inhibition efficiency of the oxidant in the historical period before the current moment is constructed by taking several historical moments before the current moment as the horizontal coordinate and taking the inhibition efficiency of the oxidant at each historical moment as the vertical coordinate.

[0019] The present invention can accurately analyze the variation trend of the inhibition efficiency in a historical period by constructing a curve graph.

[0020] Preferably, the abnormal degree of the inhibition efficiency of the antioxidant at the current moment satisfies the following relationship:

[0021] ;

[0022] In the formula, is the abnormal degree of the inhibitory efficiency of the antioxidant at the current moment; is the inhibitory efficiency of the antioxidant at the current moment; is the inhibition efficiency of the antioxidant at the moment before the current moment; The accuracy of the change trend of the inhibition efficiency of the antioxidant in the historical period before the current moment; is the absolute value symbol.

[0023] The present invention combines accuracy to determine the degree of abnormality, and can ensure the accuracy of the determined degree of abnormality.

[0024] Preferably, the RRCF algorithm is used to perform anomaly detection on the suppression efficiency at the current moment to obtain an anomaly score of the suppression efficiency at the current moment, including:

[0025] Using the RRCF algorithm, multiple random segmentation trees are constructed based on the inhibition efficiency at historical moments, and the inhibition efficiency of the antioxidant at the current moment is inserted into each random segmentation tree;

[0026] The average of the complexity changes of all random split trees before and after inserting the new suppression efficiency is calculated to obtain the anomaly score.

[0027] Preferably, the normalized value of the product of the abnormality degree and the abnormality score is used as the final abnormality score, and the final abnormality score satisfies the following relationship:

[0028] ;

[0029] In the formula, is the final abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal degree of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the normalization function.

[0030] Preferably, monitoring the abnormality of the reaction parameters of the antioxidant comprises:

[0031] A preset threshold is obtained, and if the final abnormality score is greater than the threshold, it is determined that the reaction parameters of the antioxidant are abnormal at the current moment.

[0032] The present invention can timely discover the abnormality of reaction parameters, thereby ensuring production efficiency.

[0033] According to a second aspect of the present invention, there is provided an antioxidant reaction parameter abnormality intelligent monitoring system, the antioxidant reaction parameter abnormality intelligent monitoring system comprising a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the first aspect of the present invention.

[0034] The present invention has the following effects:

[0035] The present invention calculates the accuracy of the inhibition efficiency in the historical period based on the overall situation of the inhibition efficiency in the historical period, and can effectively distinguish the normal change trend and the abnormal change trend of the inhibition efficiency in the historical period before any moment, and calculates the abnormal degree of the inhibition efficiency at the current moment based on the accuracy, which can ensure the accuracy of the determined abnormal degree, so that when the abnormal degree is used to correct the abnormal score of the inhibition efficiency at the current moment, the influence of the change trend of the data can be fully considered, so that the abnormal situation of the inhibition efficiency at the current moment can be accurately monitored, and the intelligent monitoring of the abnormal situation of the reaction parameters of the antioxidant can be realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0037] Figure 1 It is a schematic diagram of the steps of a method for intelligently monitoring abnormal reaction parameters of an antioxidant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0039] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0040] Reference Figure 1, a method for intelligently monitoring abnormal reaction parameters of an antioxidant, comprising steps S1 to S4, specifically as follows:

[0041] S1: Obtain the initial content of oxides in the reactor, calculate the difference between the initial content and the oxide content at any time after the addition of the antioxidant, and use the ratio of the difference to the initial content as the inhibition efficiency of the antioxidant at any time.

[0042] In an exemplary embodiment of the present invention, the initial content refers to the content of oxides in the reactor before the antioxidant is added. Specifically, the spectral analysis method can be used to measure the content of oxides in the reactor before the antioxidant is added to obtain the initial content of oxides in the reactor, and then the content of oxides in the reactor at each moment after the antioxidant is added can be measured at a certain frequency, such as once per minute.

[0043] It should be noted that in the production process of polymer materials, adding antioxidants can effectively prevent the performance of materials from deteriorating due to oxidation, and the effectiveness of antioxidants mainly depends on their anti-oxidation ability. If the content of oxides in the reactor after adding antioxidants is significantly lower than the content of oxides in the reactor without adding antioxidants, it means that the antioxidant can effectively inhibit the oxidation reaction in the production process of polymer materials under the current conditions. Therefore, the inhibition efficiency of the antioxidant at any time can be evaluated by combining the difference between the content of oxides in the reactor at each moment after adding antioxidants and the content of oxides in the reactor before adding antioxidants.

[0044] Specifically, the inhibition efficiency of the antioxidant at any time satisfies the following relationship:

[0045] ;

[0046] In the formula, For antioxidants at this time The inhibition efficiency of is the initial content of oxides in the reactor; is the current time in the reactor after adding the antioxidant of oxide content.

[0047] in, The larger it is, the greater the difference between the oxide content in the reactor at the current moment and the initial content after the addition of the antioxidant, the better the inhibition effect of the antioxidant at the current moment, and the greater the corresponding inhibition efficiency of the antioxidant at the current moment.

[0048] S2: Construct a curve graph of the inhibition efficiency of antioxidants in the historical period before the current moment, count the proportion of data with negative slope in the curve graph, calculate the normalized value of the product of the data proportion and the standard deviation of the inhibition efficiency in the historical period, and use the difference between 1 and the normalized value as the accuracy of the changing trend of the inhibition efficiency in the historical period.

[0049] In an exemplary embodiment of the present invention, accuracy refers to data that measures the possibility that the changing trend of the inhibition efficiency within a historical period is a normal changing trend. If the changing trend of the inhibition efficiency within the historical period is closer to the normal changing trend of the inhibition efficiency of the antioxidant, it means that the probability that the changing trend of the inhibition efficiency within the historical period is a normal changing trend is relatively high, and the corresponding accuracy of the changing trend of the inhibition efficiency within the historical period is relatively high.

[0050] It should be noted that in the production process of polymer materials, sudden changes in the environment or changes in boundary conditions may cause the reaction parameters in the reactor to approach the activity threshold of the antioxidant, thereby failing to effectively inhibit the oxidation reaction in the production process of the polymer material, resulting in a decrease in the inhibition efficiency at the corresponding moment; in general, after the antioxidant is added to the reactor, as the reaction proceeds, the antioxidant will gradually disperse and react with potential oxidizing substances, thereby gradually reducing the oxide content in the reactor. Correspondingly, the inhibition efficiency of the antioxidant will gradually increase over time until it reaches a stable state. Therefore, this feature can be used to evaluate the possibility that the change trend of the inhibition efficiency of the antioxidant is a normal change trend.

[0051] In an exemplary embodiment of the present invention, the construction of a curve graph of the inhibition efficiency of the antioxidant in a historical period before the current moment can be achieved by the following steps:

[0052] A curve graph of the inhibition efficiency of the oxidant in the historical period before the current moment is constructed by taking several historical moments before the current moment as the horizontal coordinate and taking the inhibition efficiency of the oxidant at each historical moment as the vertical coordinate.

[0053] Furthermore, the slope of any data point in the graph can be calculated, and then the number of data points with negative slopes in the graph can be counted, and the ratio of the number to the number of all data points in the graph can be calculated to obtain the proportion of data with negative slopes in the graph.

[0054] Furthermore, the normalized value of the product of the data proportion and the standard deviation of the suppression efficiency in the historical period can be calculated, and then the difference between 1 and the normalized value can be calculated to obtain the accuracy of the change trend of the suppression efficiency in the historical period.

[0055] Specifically, the accuracy of the variation trend of the inhibition efficiency of the antioxidant in the historical period before any moment satisfies the following relationship:

[0056] ;

[0057] In the formula, The accuracy of the change trend of the inhibition efficiency of the antioxidant in the historical period before the current moment; It is the first antioxidant in history. the efficiency of suppression at each historical moment; is the average value of the inhibition efficiency of the antioxidant in the historical period; is the number of data points with negative slope in the curve graph of inhibition efficiency in the historical period; is the number of historical moments in the historical period; To preset hyper parameters, in this embodiment ; is the normalization function.

[0058] in, is the standard deviation of the inhibition efficiency in the historical period before the current moment. The smaller the value is, the better the stability of the inhibition efficiency in the historical period is, and the greater the possibility that the change trend of the inhibition efficiency in the corresponding historical period is a normal change trend.

[0059] In the curve chart constructed based on the inhibition efficiency in the historical period before the current moment, the data with a negative slope accounts for a smaller proportion. The smaller the value, the more the changing trend of the inhibition efficiency of the antioxidant in the historical period conforms to the normal changing trend of the inhibition efficiency of the antioxidant, and the corresponding antioxidant is more likely to have a normal changing trend in the historical period.

[0060] S3: Calculate the abnormality of the inhibition efficiency of the antioxidant at the current moment. The abnormality is positively correlated with the difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, as well as the accuracy.

[0061] In an exemplary embodiment of the present invention, the degree of abnormality refers to data that measures the possibility that the inhibition efficiency of the antioxidant at the current moment is abnormal data. It should be noted that the antioxidant can continuously inhibit the oxidation reaction during the production process of the polymer material, so that the change of the inhibition efficiency of the antioxidant at each moment is relatively smooth. If there is a significant difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, it means that the reaction parameters at the current moment are more likely to be abnormal data. In addition, if the accuracy of the change trend of the inhibition efficiency in the historical time period before the current moment is high, it means that the reliability of the inhibition efficiency of the antioxidant in the historical time period is high. If there is a large difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, the inhibition efficiency at the current moment is more likely to be abnormal data. Therefore, this feature can be used to calculate the degree of abnormality of the inhibition efficiency of the antioxidant at the current moment.

[0062] Specifically, the abnormal degree of the inhibition efficiency of the antioxidant at any time satisfies the following relationship:

[0063] ;

[0064] In the formula, is the abnormal degree of the inhibitory efficiency of the antioxidant at the current moment; is the inhibitory efficiency of the antioxidant at the current moment; is the inhibition efficiency of the antioxidant at the moment before the current moment; The accuracy of the change trend of the inhibition efficiency of the antioxidant in the historical period before the current moment; is the absolute value symbol.

[0065] in, It reflects the difference between the inhibition efficiency of the antioxidant at the current moment and the inhibition efficiency at the previous moment. The larger the value is, the more likely the inhibition efficiency at the current moment is to be abnormal data. When the accuracy of the changing trend of the inhibition efficiency of the antioxidant in the historical period before the current moment is high, the possibility that the inhibition efficiency at the current moment is abnormal data is relatively greater, and the corresponding degree of abnormality of the inhibition efficiency of the antioxidant at the current moment is greater.

[0066] In another embodiment, other formulas may be used to calculate the abnormal degree of the inhibition efficiency of the antioxidant at any time, such as using a difference calculation formula instead of a proportional relationship calculation formula.

[0067] S4: Use the RRCF algorithm to perform anomaly detection on the inhibition efficiency at the current moment, obtain the anomaly score of the inhibition efficiency at the current moment, and monitor the anomaly of the reaction parameters of the antioxidant based on the normalized value of the product of the anomaly degree and the anomaly score and the comparison result of the preset threshold.

[0068] It should be noted that the Robust Random Cut Forest (RRCF) can determine the abnormality of new samples compared to the overall data based on the changes in the complexity of the random cut tree caused by the insertion or deletion of new samples. The inhibition efficiency of the antioxidant will gradually increase over time until it reaches a stable state, but because the algorithm will regard changes inconsistent with the overall data as abnormalities when processing data, when the inhibition efficiency of the antioxidant is at a normal change node of the change trend, but different from the historical change trend, the algorithm may misjudge it as abnormal data, which may reduce the accuracy of abnormality detection to a certain extent. Therefore, the present invention combines the abnormal degree of the inhibition efficiency of the antioxidant at the current moment, and corrects the abnormal score of the inhibition efficiency determined by the algorithm at the current moment, which can take into account the impact of the change trend of the data on the data abnormality, so that the reaction parameters of the antioxidant can be accurately detected for abnormality.

[0069] In an exemplary embodiment of the present invention, the determination of the abnormal score of the suppression efficiency at the current moment can be achieved by the following steps:

[0070] Using the RRCF algorithm, multiple random segmentation trees are constructed based on the inhibition efficiency at historical moments, and the inhibition efficiency of the antioxidant at the current moment is inserted into each random segmentation tree; the average value of the complexity change of all random segmentation trees before and after the insertion of the new inhibition efficiency is calculated to obtain the anomaly score.

[0071] Among them, the process of constructing multiple random partitioning trees using RRCF and the process of inserting new data into the constructed random partitioning trees are existing technologies and will not be described in detail in this embodiment.

[0072] The process of determining the complexity change is as follows: first, calculate the sum of the path lengths of each node in any random segmentation tree to obtain the complexity before inserting the new suppression efficiency; then, insert the suppression efficiency at the current moment into the random segmentation tree, calculate the sum of the path lengths of each node in the new random segmentation tree, and obtain the complexity after inserting the new suppression efficiency; finally, calculate the difference in complexity of the suppression efficiency before and after the insertion to obtain the complexity change of the random segmentation tree before and after the insertion of the new suppression efficiency, and then obtain the average value of the complexity change of all random segmentation trees before and after the insertion of the new suppression efficiency, and obtain the abnormal score of the suppression efficiency at the current moment. It should be noted that the abnormal score is a professional term in the RRCF algorithm, and this embodiment will not be described in detail here.

[0073] In an exemplary embodiment of the present invention, the normalized value of the product of the abnormality degree and the abnormality score may be used as the final abnormality score, and the final abnormality score satisfies the following relationship:

[0074] ;

[0075] In the formula, is the final abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal degree of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the normalization function.

[0076] In another embodiment, the product of the abnormal degree of the suppression efficiency at the current moment and the abnormal score of the suppression efficiency at the current moment may be normalized by other methods, such as by a deformed exponential function. This embodiment does not specifically limit the normalization method.

[0077] In an exemplary embodiment of the present invention, the abnormal conditions of the reaction parameters of the antioxidant can be monitored by the following steps:

[0078] A preset threshold is obtained, and if the final abnormality score is greater than the threshold, it is determined that the reaction parameters of the antioxidant at the current moment are abnormal.

[0079] Optionally, the threshold value can be set to 0.7 based on an empirical value. If the final abnormal score of the inhibition efficiency at the current moment is greater than 0.7, it is considered that the inhibition efficiency of the antioxidant is abnormal at the current moment, which further indicates that the reaction parameters of the antioxidant are abnormal at the current moment. Therefore, the abnormal situation of the reaction parameters of the antioxidant can be monitored in real time, and corresponding measures can be taken in time to adjust the corresponding reaction parameters to ensure the normal reaction conditions of the antioxidant.

[0080] The present invention also provides an antioxidant reaction parameter abnormality intelligent monitoring system, the system includes a memory and a processor, and a computer program is stored in the memory, the computer program integrates the function of an antioxidant reaction parameter abnormality intelligent monitoring method, when the computer program is executed, the antioxidant reaction parameter abnormality intelligent monitoring method can improve the accuracy of monitoring the abnormal situation of the antioxidant reaction parameters.

[0081] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0082] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. An intelligent monitoring method for abnormal reaction parameters of antioxidants, characterized in that: include: Obtaining the initial content of oxides in the reactor, calculating the difference between the initial content and the content of oxides at any time after the addition of the antioxidant, and taking the ratio of the difference to the initial content as the inhibition efficiency of the antioxidant at any time; Constructing a curve graph of the inhibition efficiency of the antioxidant in the historical period before the current moment, counting the proportion of data with a negative slope in the curve graph, calculating a normalized value of the product of the data proportion and the standard deviation of the inhibition efficiency in the historical period, and taking the difference between 1 and the normalized value as the accuracy of the variation trend of the inhibition efficiency in the historical period; Calculating the abnormality of the inhibition efficiency of the antioxidant at the current moment, wherein the abnormality is positively correlated with the difference between the inhibition efficiency at the current moment and the inhibition efficiency at the previous moment, and the accuracy; The RRCF algorithm is used to perform abnormal detection on the inhibition efficiency at the current moment, and the abnormal score of the inhibition efficiency at the current moment is obtained. The abnormality of the reaction parameters of the antioxidant is monitored based on the normalized value of the product of the abnormal degree and the abnormal score, and the comparison result of the preset threshold.

2. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 1, characterized in that: The accuracy of the change trend of the suppression efficiency in the historical period satisfies the following relationship: ； In the formula, The accuracy of the change trend of the inhibition efficiency of the antioxidant in the historical period before the current moment; It is the first antioxidant in history. the efficiency of suppression at each historical moment; is the average value of the inhibition efficiency of the antioxidant in the historical period; is the number of data points with negative slope in the curve graph of inhibition efficiency in the historical period; is the number of historical moments in the historical period; To preset hyperparameters; is the normalization function.

3. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 2, characterized in that: The method for obtaining the curve graph includes: A curve graph of the inhibition efficiency of the oxidant in the historical period before the current moment is constructed by taking several historical moments before the current moment as the horizontal coordinate and taking the inhibition efficiency of the oxidant at each of the historical moments as the vertical coordinate.

4. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 1, characterized in that: The abnormal degree of the inhibitory efficiency of the antioxidant at the current moment satisfies the following relationship: ； In the formula, For antioxidants at this time The abnormal degree of inhibitory efficiency; For antioxidants at this time The inhibition efficiency of For antioxidants at this time The inhibition efficiency at the previous moment; For antioxidants at this time The accuracy of the trend of suppression efficiency in previous historical periods; is the absolute value symbol.

5. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 1, characterized in that: The method of using the RRCF algorithm to perform anomaly detection on the suppression efficiency at the current moment and obtaining an anomaly score of the suppression efficiency at the current moment includes: Using the RRCF algorithm, a plurality of random segmentation trees are constructed based on the inhibition efficiency at historical moments, and the inhibition efficiency of the antioxidant at the current moment is inserted into each of the random segmentation trees; The average value of the complexity change of all random segmentation trees before and after the insertion of the new inhibition efficiency is calculated to obtain the abnormality score.

6. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 1, characterized in that: The normalized value of the product of the abnormality degree and the abnormality score is taken as the final abnormality score, and the final abnormality score satisfies the following relationship: ； In the formula, is the final abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal degree of the inhibitory efficiency of the antioxidant at the current moment; is the abnormal score of the inhibitory efficiency of the antioxidant at the current moment; is the normalization function.

7. The method for intelligently monitoring abnormal reaction parameters of an antioxidant according to claim 6, characterized in that: The abnormality of the reaction parameters of the monitoring antioxidant includes: A preset threshold is obtained, and if the final abnormality score is greater than the threshold, it is determined that the reaction parameters of the antioxidant are abnormal at the current moment.

8. An antioxidant reaction parameter abnormality intelligent monitoring system, characterized in that: The antioxidant reaction parameter abnormality intelligent monitoring system comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

Images