A chemical equipment safety monitoring method and system based on artificial intelligence

Abstract

The present invention relates to the technical field of equipment monitoring and control, and in particular to a chemical equipment safety monitoring method and system based on artificial intelligence. The method comprises the steps of: obtaining parameter items corresponding to each data point in a time series sequence of chemical equipment operation data, and determining the time series sequence of the parameter items; calculating the regional synchronization degree of the data point; determining its offset degree through the regional synchronization degree of the data point; obtaining the window segment of the data point, and determining the interval abnormality of the data point through the data characteristics of each parameter item in the window segment of the data point and the adjacent data points on the left and right sides; calculating the corrected local density of the data point based on the offset degree, local density and interval abnormality of the data point; obtaining the abnormality through the k nearest neighbor data points of the data point and the corrected local density of the data point; realizing the safety monitoring of the chemical equipment through the comparison result of the abnormality of the data point and the preset abnormal threshold, which can effectively improve the accuracy of the safety monitoring of the chemical equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring and control, and in particular to a chemical equipment safety monitoring method and system based on artificial intelligence. Background Art

[0002] Chemical equipment refers to mechanical equipment used to complete chemical reactions, material transmission, separation and storage tasks in chemical production. Its production process involves high temperature, high pressure, flammable and explosive, and toxic and hazardous substances. The safety of chemical equipment affects the normal operation of the production process and directly determines the quality of chemical products.

[0003] In the prior art, the safe operation of equipment is mostly achieved by abnormal monitoring and early warning of equipment. For example, the patent application document with publication number CN110412903A discloses an equipment monitoring device, detection system and detection method. The monitoring module of the application is used to monitor and collect internal temperature and humidity, vibration amplitude and dust concentration to generate monitoring data; the LORA transmission module is used to receive monitoring data and convert the monitoring data into wireless transmission signals, so without destroying the equipment, users can collect monitoring data of the equipment through the equipment monitoring device of the application. Receive wireless transmission signals and convert them into detection data, so as to compare the detection data with the preset value and generate detection results, thereby assisting users in the data recording, processing and analysis and early warning required for subsequent equipment tracking.

[0004] The above-mentioned prior art realizes the status monitoring of the equipment by collecting the equipment parameter data. However, the equipment will gradually age after long-term operation, and the conventional abnormal monitoring method may cause identification errors.

[0005] Based on this, how to accurately implement artificial intelligence-based safety monitoring of chemical equipment is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the technical problem of how to accurately implement chemical equipment safety monitoring based on artificial intelligence, the present invention provides a chemical equipment safety monitoring method and system based on artificial intelligence.

[0007] In the first aspect, the present invention provides a chemical equipment safety monitoring method based on artificial intelligence, which adopts the following technical solutions:

[0008] A chemical equipment safety monitoring method based on artificial intelligence, comprising the steps of:

[0009] Obtain the parameter item corresponding to each data point in the time series sequence of the chemical equipment operation data, and determine the time series sequence of the parameter item;

[0010] ;

[0011] is the regional synchronization degree of the data points, is the total number of data points in the running data time series, is the time index of the data point, is the time index of the i-th data point in the running data time series except this data point, is the number of parameter item categories, For the The parameter term is the variance of the time series, The data point corresponding to The value of the parameter item, is the number corresponding to the i-th data point in the running data time series except this data point The value of the parameter item, is the linear normalization function, is an exponential function with base e, is an absolute value; the degree of deviation is determined by the regional synchronization of the data point; the window segment of the data point is obtained, and the interval abnormality of the data point is determined by the data characteristics of each parameter item in the window segment of the data point and the adjacent data points on the left and right sides;

[0012] ; is the local density after correction of the data points, is the degree of deviation of the data point, is the local density of the data point, is the interval anomaly of the data point; the average of the ratio of the k nearest neighboring data points of the data point to the corrected local density of the data point is recorded as the anomaly of the data point; the safety monitoring of the chemical equipment is achieved by comparing the anomaly of the data point with the preset anomaly threshold.

[0013] When realizing the safety monitoring of chemical equipment, the present invention processes the multidimensional data in the time series of the operation data of the chemical equipment through the LOF algorithm, and can accurately realize the safety monitoring of the chemical equipment. In this process, the present invention takes into account that the aging of chemical equipment will affect the distribution of data points, resulting in low accuracy of the local density of data points; based on this, the present invention corrects the local density by obtaining data points with a lower degree of deviation, and can improve the accuracy of the abnormality of the data points. On this basis, the present invention also corrects the local density of data points with a higher degree of deviation by obtaining the interval abnormality of the data points, effectively improving the accuracy of the abnormality of the obtained data points, so that the safety monitoring of chemical equipment can be accurately realized.

[0014] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the method of obtaining parameter items corresponding to each data point in a time series sequence of chemical equipment operation data also includes: combining the values ​​of various parameter items generated by the operation of the equipment at the same acquisition moment into one data point, and performing preprocessing to obtain a time series sequence of chemical equipment operation data, and arranging the same type of parameter items in time series to obtain a parameter item time series sequence.

[0015] The present invention takes into account the possible existence of noise interference and data missing in the originally collected data, and therefore improves the overall quality of the data through preprocessing to facilitate subsequent data processing.

[0016] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the offset degree of the data point satisfies the relationship: ; is the degree of deviation of the data points, is a hyperparameter, is the regional synchronization degree of the data points, It is the median of regional synchronization in the time series of operating data.

[0017] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the method for obtaining the window segment of the data point includes: presetting the window segment length ; Take the data point as the center and obtain data point and obtain the window segment of the data point.

[0018] The present invention constructs a window segment by acquiring other data points on both sides of the current data point, and comprehensively considers the data changes before and after the current data point, so that the prominence of the current data point compared with the surrounding data can be accurately obtained based on this, thereby obtaining the interval anomaly of the data point.

[0019] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the interval abnormality of the data point satisfies the relationship:

[0020] ;

[0021] is the interval abnormality of the data point, is the number of parameter item categories, The data point corresponding to The value of the parameter item, is the first The parameter item mean, 2 is the number of adjacent data points of this data point, is the number of adjacent data points on one side of the data point in the window segment The mean of the parameter items, is the first Parameter term variance, is the number of adjacent data points on one side of the data point in the window segment Parameter term variance.

[0022] The present invention provides an accurate calculation formula for the interval anomaly degree of a data point. By obtaining the prominence of the current data point and its window segment data, as well as the overall difference between the window segment data of the current data point and its left and right adjacent data points, the interval anomaly degree of the current data point can be accurately obtained.

[0023] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the comparison result of the abnormality of the data point and the preset abnormality threshold includes: if the abnormality of the data point is greater than the preset abnormality threshold, the data point is an abnormal data point; otherwise, the data point is a normal data point.

[0024] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the safety monitoring of chemical equipment is achieved by comparing the abnormality of a data point with a preset abnormal threshold, including: obtaining a chemical equipment early warning result and processing it through the number of continuous abnormal data points in a time series sequence of chemical equipment operation data to achieve safety monitoring of chemical equipment.

[0025] According to a chemical equipment safety monitoring method based on artificial intelligence provided by the present invention, the chemical equipment early warning result is obtained and processed through the number of continuous abnormal data points in the time series sequence of chemical equipment operation data, including: if the proportion of the number of abnormal data points in the time series sequence of chemical equipment operation data is greater than a preset threshold, the chemical equipment early warning result is to start the early warning and issue a power-off command; otherwise, the chemical equipment early warning result is that the equipment is normal.

[0026] The present invention starts an early warning when an abnormality occurs in chemical equipment, and controls the chemical equipment to take measures, which can effectively reduce the potential safety hazards in the operation process of the chemical equipment.

[0027] In a second aspect, the present invention provides a chemical equipment safety monitoring system based on artificial intelligence, which adopts the following technical solutions:

[0028] A chemical equipment safety monitoring system based on artificial intelligence includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned chemical equipment safety monitoring method based on artificial intelligence is implemented.

[0029] By adopting the above technical solution, the above-mentioned artificial intelligence-based chemical equipment safety monitoring method is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made based on the memory and the processor for easy use.

[0030] The present invention has the following technical effects:

[0031] Based on the above technical solution, when realizing the safety monitoring of chemical equipment, the present invention processes the multidimensional data in the time series of the operation data of chemical equipment through the LOF algorithm, and can accurately realize the safety monitoring of chemical equipment. In this process, the present invention takes into account that the aging of chemical equipment will affect the distribution of data points, resulting in low accuracy of the local density of data points; based on this, the present invention corrects the local density by obtaining data points with a lower degree of deviation, and can improve the accuracy of the abnormality of the data points. On this basis, the present invention also corrects the local density of data points with a higher degree of deviation by obtaining the interval abnormality of the data points, effectively improving the accuracy of the abnormality of the obtained data points, so that the safety monitoring of chemical equipment can be accurately realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, 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-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0033] Figure 1 A schematic flow chart of a chemical equipment safety monitoring method based on artificial intelligence provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] 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.

[0035] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0036] Chemical equipment refers to mechanical equipment used to complete chemical reactions, material transmission, separation and storage in chemical production. Its production process involves high temperature, high pressure, flammable and explosive, and toxic and hazardous substances. The safety of chemical equipment affects the normal operation of the production process and directly determines the quality of chemical products. Many types of parameters will be generated during the production process of chemical equipment. When monitoring the production process of chemical equipment, an anomaly detection algorithm that can be used for medium-high dimensional data is needed.

[0037] The Local Outlier Factor (LOF) algorithm is an algorithm for anomaly detection. The algorithm obtains the degree of anomaly of a data point through the local density of the data point and other points in its k-neighborhood, and can realize anomaly detection. During the calculation process, the LOF algorithm considers both the local and global properties of the data set, and the outlier is determined relative to the density of the neighborhood points. Therefore, when there are different clusters of different densities in the data set, the LOF algorithm performs well and can be applied to anomaly monitoring of medium and high-dimensional data sets.

[0038] Based on this, an embodiment of the present invention discloses a chemical equipment safety monitoring method based on artificial intelligence. The method uses the LOF algorithm to perform abnormal monitoring on the time series of chemical equipment operation data, and can accurately obtain the chemical equipment safety monitoring results.

[0039] For details, please see Figure 1 As shown, Figure 1 A flow chart of a chemical equipment safety monitoring method based on artificial intelligence provided in an embodiment of the present invention, the method specifically includes the following steps.

[0040] S1: Obtain the parameter item corresponding to each data point in the time series sequence of chemical equipment operation data, and determine the time series sequence of the parameter item.

[0041] The parameter item category may be temperature, pressure, vibration, current, voltage, etc.; the parameter item category may be specifically set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0042] It should be noted that the abnormality of chemical equipment may be reflected in a variety of parameters. For example, equipment failure or process out of control may cause significant changes in the temperature of the equipment; bearing wear, imbalance or alignment errors may cause abnormal pressure and vibration data of the equipment; electrical failure or equipment overload may cause abnormal current and voltage of the equipment, etc.

[0043] Based on this, in order to accurately obtain the monitoring results of the chemical equipment, the embodiment of the present invention can analyze each dimension of the above parameters separately, so as to more comprehensively describe the operating status of the chemical equipment.

[0044] For example, in an embodiment of the present invention, the parameter items corresponding to each data point in the time series sequence of chemical equipment operation data are obtained, which also includes: the values ​​of various parameter items generated by the equipment operation at the same acquisition time are combined into one data point, and pre-processing is performed to obtain the time series sequence of chemical equipment operation data, and the parameter items of the same type are arranged in time series to obtain a parameter item time series sequence.

[0045] The preprocessing may be Gaussian filtering denoising, missing data interpolation, etc., which may be specifically configured according to actual needs, and the embodiment of the present invention does not impose too many restrictions thereon.

[0046] Specifically, temperature sensors and pressure sensors are arranged in the reactors, pipelines and storage tanks of the chemical equipment to collect the time series of temperature parameter items and the time series of pressure parameter items, respectively; vibration sensors, current sensors and voltage sensors are arranged in the motors, pumps and compressors of the chemical equipment to collect the time series of vibration parameter items, the time series of current parameter items and the time series of voltage parameter items.

[0047] Among them, the acquisition frequency of the sensor can be set to 10 Hz, which can be set according to actual needs.

[0048] After acquiring each parameter item in the chemical equipment data point and the corresponding parameter item time series based on the above steps, anomalies can be identified through the local density of the data points.

[0049] For example, in an embodiment of the present invention, when determining the local density of a data point, the distance between the current data point and other data points in the running data time series sequence can be obtained; the k data points closest to the current data point are obtained as the k nearest neighbor data points of the current data point; and the inverse of the average distance between the current data point and its k nearest neighbor data points is taken as the local density of the current data point.

[0050] Among them, k can be set to 10; the k value can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0051] S2: Calculate the regional synchronicity of the data points, and determine the degree of deviation through the regional synchronicity of the data points.

[0052] It should be noted that the equipment will gradually age after long-term operation, and the distribution and concentration areas of the parameters of each dimension will deviate, and the data points will shift, which may lead to uneven density distribution of the parameters in different areas. When the calculation is directly based on the LOF algorithm, the local density accuracy of the obtained data points will be low, thus affecting the accuracy of the safety monitoring results of chemical equipment.

[0053] It should be further explained that the data points are concentrated in different areas in different dimensions, but the aging degree of the equipment is consistent. The dimensional parameter items with large fluctuations usually mean that the parameter items are more sensitive to small changes in the equipment status. When the equipment ages, these sensitive parameters are more easily affected. In addition, the time distribution between data points is relatively uniform under normal conditions. The more uneven the time distribution of data points in a region, the greater the possibility of its offset.

[0054] Based on this, the embodiment of the present invention analyzes the fluctuation of each parameter item time series to obtain its weight when identifying the offset data point. The higher the fluctuation of the parameter item, the greater its weight. Combined with the difference in collection time between the current data point and its surrounding data points in the area, the uniformity of the collection time of the current data point is obtained.

[0055] For example, in the embodiment of the present invention, the regional concurrency of the data points is determined, and the specific relationship can be as follows:

[0056] ;

[0057] is the regional synchronization degree of the data points, is the total number of data points in the running data time series, is the time index of the data point, is the time index of the i-th data point in the running data time series except this data point, is the number of parameter item categories, For the The parameter term is the variance of the time series, The data point corresponding to The value of the parameter item, is the number corresponding to the i-th data point in the running data time series except this data point The value of the parameter item, is the linear normalization function, is an exponential function with base e, is an absolute value.

[0058] In the above formula, Indicates the closeness of the collection time of the current data point to other data points. The smaller it is, the closer the collection time of the current data point is to that of other data points, the more uniform the collection time of the current data point is with other data points, the smaller the possibility of offset of the current data point is, and the higher the corresponding regional synchronization is.

[0059] The smaller the variance of the parameter time series, the smaller its volatility; It means that the Manhattan distance between each parameter item of the data point is weighted by the variance of the time series of each parameter item, so that the closer the data point with smaller fluctuation is to the current data point, the higher the credibility of the regional synchronization of the current data point is.

[0060] After obtaining the regional synchronicity of each data point based on the above formula, the degree of deviation can be obtained based on the regional synchronicity of the data point.

[0061] For example, in the embodiment of the present invention, the degree of deviation of the data point is determined, and specifically, the following relationship can be referred to:

[0062] ;

[0063] is the degree of deviation of the data points, is a hyperparameter, is the regional synchronization degree of the data points, is the median of regional synchronization in the time series of running data, is an exponential function with base e.

[0064] Among them, the hyperparameter can be set to 30; it can be set specifically according to actual needs, and the embodiment of the present invention does not impose too many restrictions here.

[0065] In the above formula, the median of the regional synchronicity of all data points in the running data time series can represent the state of normal distribution of the data point collection time without deviation. The closer the regional synchronicity of the current data point is to the median of the regional synchronicity, the more uniform the collection time of the current data point is, and the lower the deviation of the data point is.

[0066] After obtaining the degree of deviation of each data point based on the above steps, continue to perform the following steps.

[0067] S3: Obtain a window segment of the data point, and determine the interval abnormality of the data point through the data characteristics of each parameter item in the window segment between the data point and the adjacent data points on the left and right sides.

[0068] It should be noted that the higher the degree of deviation of the data point, the lower its credibility in determining the local density of the data point, and the local density of the data point cannot be accurately obtained based on the degree of deviation of the data point. The aging of chemical equipment is a long-term process. Therefore, the closer the data points are to the time of collection, the smaller the deviation difference caused by aging. Based on this, the embodiment of the present invention obtains the interval abnormality of the current data point by obtaining the data change between the local window segment of the current data point and the local window segment of the adjacent data point. The lower the interval abnormality of the data point, the less likely it is to be an abnormal data point.

[0069] For example, in an embodiment of the present invention, the method for obtaining a window segment of a data point includes the following two possible implementations:

[0070] In a possible implementation, the window segment length can be preset ; Take the data point as the center and obtain data point and obtain the window segment of the data point.

[0071] in, It can be set to 49; for the convenience of analysis, the window segment length can be set to an odd number, and the window segment length can be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0072] It is understandable that the final window segment of the current data point includes the current data point itself. Thus, the embodiment of the present invention can accurately describe the data changes on both sides of the current data point by acquiring other data points on both sides of the current data point in equal amounts to construct the window segment of the current data point.

[0073] In another possible implementation, the window segment length can be preset ; Taking the data point as the end, obtain the historical data of the data point data point and obtain the window segment of the data point.

[0074] In this way, the embodiment of the present invention obtains the historical data of the current data point as the window segment data of the current data point, so that the prominence of the current data point can be obtained later according to the trend change of the historical data and the data deviation between the current data point.

[0075] After the window segment of each data point is obtained based on the above method, the interval abnormality of the current data point can be obtained based on the difference in the window segment prominence between the data point and the adjacent data points on the left and right sides.

[0076] For example, in the embodiment of the present invention, the interval abnormality of the data point is determined, and the specific relationship can be as follows:

[0077] ;

[0078] is the interval abnormality of the data point, is the number of parameter item categories, The data point corresponding to The value of the parameter item, is the first The parameter item mean, 2 is the number of adjacent data points of this data point, is the number of adjacent data points on one side of the data point in the window segment The mean of the parameter items, is the first Parameter term variance, is the number of adjacent data points on one side of the data point in the window segment Parameter term variance, is the linear normalization function, is an absolute value.

[0079] It is understandable that if is the right adjacent data point of the current data point, then When , it is the left adjacent data point of the current data point; otherwise, if is the left adjacent data point of the current data point, then When is the data point adjacent to the right of the current data point, it can be set according to actual needs.

[0080] In the above formula, Indicates the current data point The parameter item is relative to the first The prominence of the parameter item mean. The larger the value, the more prominent the current data point is relative to the data points in the window segment, and the higher the possibility of being abnormal data.

[0081] It indicates the overall difference in data characteristics between each parameter item in the current data point and each parameter item in the window segment of the adjacent data points on the left and right sides. The smaller the value, the smaller the difference between the current data point and the adjacent data points on the left and right sides, the less likely the current data point and the adjacent data points on the left and right sides are to be affected by aging or other factors, and the greater the credibility of the interval abnormality of the current data point.

[0082] After obtaining the degree of deviation and interval anomaly of each data point based on the above steps, the local density of the data point can be corrected based on the degree of deviation and interval anomaly of the data point, so as to accurately obtain the true local density of the data point, that is, continue to execute the following steps.

[0083] S4: Correct the local density of the data point based on the deviation degree and interval anomaly of the data point to obtain the corrected local density of the data point; record the average of the ratios of the k nearest neighbor data points of the data point to the corrected local density of the data point as the anomaly of the data point.

[0084] It should be noted that if the deviation of a data point is low, its contribution to the calculation of the local density of the data point is large. Therefore, the local density correction can be achieved based on the data point with a low deviation. If the deviation of a data point is high, it means that the local density of the data point is used to calculate the reliability of the abnormality of the data point. In this case, the interval abnormality of the data point can also be obtained. The data point with a low interval abnormality has a smaller degree of data deviation caused by the aging of chemical equipment, and the possibility of being real data is higher. At this time, the interval abnormality of the data point can better reflect the real abnormal state of the data point.

[0085] For example, in the embodiment of the present invention, the local density of the corrected data point is determined, and the specific details can be referred to the following relationship:

[0086] ;

[0087] is the local density after correction of the data points, is the degree of deviation of the data point, is the local density of the data point, is the interval anomaly of the data point.

[0088] The smaller the deviation degree and interval anomaly of the data point, the larger the corresponding corrected local density.

[0089] After the corrected local density of each data point is obtained based on the above formula, its abnormality can be obtained based on the corrected local density of each data point, thereby realizing the safety monitoring of chemical equipment.

[0090] S5: The safety monitoring of chemical equipment is achieved by comparing the abnormality of the data point with the preset abnormality threshold.

[0091] For example, in an embodiment of the present invention, the comparison result of the abnormality of a data point and a preset abnormality threshold includes: if the abnormality of a data point is greater than the preset abnormality threshold, the data point is an abnormal data point; otherwise, the data point is a normal data point.

[0092] The preset abnormality threshold may be set to 1.0; the preset abnormality threshold may be set specifically according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0093] It should be noted that when a chemical equipment runs abnormally, it is usually accompanied by a series of abnormal data. Based on this, the embodiment of the present invention achieves safety monitoring of the chemical equipment by acquiring the number of continuous abnormal data points in the data time series.

[0094] By way of example, in an embodiment of the present invention, safety monitoring of chemical equipment is achieved by comparing the abnormality of a data point with a preset abnormal threshold, including: obtaining early warning results of the chemical equipment and making processing based on the number of continuous abnormal data points in a time series sequence of chemical equipment operation data, so as to achieve safety monitoring of the chemical equipment.

[0095] By way of example, in an embodiment of the present invention, a warning result of chemical equipment is obtained and processed by the number of continuous abnormal data points in a time series sequence of chemical equipment operation data, including: if the proportion of the number of abnormal data points in the time series sequence of chemical equipment operation data is greater than a preset threshold, then the warning result of the chemical equipment is to start a warning and issue a power-off command; otherwise, the warning result of the chemical equipment is that the equipment is normal.

[0096] The preset threshold may be set to 30; the preset threshold may be set according to actual needs, and the embodiment of the present invention does not impose too many restrictions on this.

[0097] It can be understood that after the embodiment of the present invention monitors that the early warning result of the chemical equipment is to start the early warning, it can control the various modules of the chemical equipment to cut off the power in time by issuing a power-off instruction to each module of the chemical equipment, and notify the staff based on the early warning result, which helps the staff to perform maintenance and processing in time.

[0098] It can be seen that in the embodiment of the present invention, the parameter item corresponding to each data point in the time series sequence of the chemical equipment operation data is obtained, and the time series sequence of the parameter item is determined;

[0099] ;

[0100] is the regional synchronization degree of the data points, is the total number of data points in the running data time series, is the time index of the data point, is the time index of the i-th data point in the running data time series except this data point, is the number of parameter item categories, For the The parameter term is the variance of the time series, The data point corresponding to The value of the parameter item, is the number corresponding to the i-th data point in the running data time series except this data point The value of the parameter item, is the linear normalization function, is an exponential function with base e, is an absolute value; the degree of deviation is determined by the regional synchronization of the data point; the window segment of the data point is obtained, and the interval abnormality of the data point is determined by the data characteristics of each parameter item in the window segment of the data point and the adjacent data points on the left and right sides;

[0101] ; is the local density after correction of the data points, is the degree of deviation of the data point, is the local density of the data point, is the interval anomaly of the data point; the average of the ratio of the k nearest neighboring data points of the data point to the corrected local density of the data point is recorded as the anomaly of the data point; the safety monitoring of the chemical equipment is achieved by comparing the anomaly of the data point with the preset anomaly threshold.

[0102] In this way, the embodiment of the present invention can accurately realize the safety monitoring of chemical equipment by processing the multidimensional data in the time series of chemical equipment operation data through the LOF algorithm. In this process, the embodiment of the present invention takes into account that the aging of chemical equipment will affect the distribution of data points, resulting in low accuracy of the local density of data points; based on this, the embodiment of the present invention corrects the local density by obtaining data points with a lower degree of deviation, which can improve the accuracy of the abnormality of the data points. On this basis, the embodiment of the present invention also corrects the local density of data points with a higher degree of deviation by obtaining the interval abnormality of the data points, which effectively improves the accuracy of the abnormality of the obtained data points, thereby accurately realizing the safety monitoring of chemical equipment.

[0103] An embodiment of the present invention also discloses an artificial intelligence-based chemical equipment safety monitoring system, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, an artificial intelligence-based chemical equipment safety monitoring method provided by the present invention is implemented.

[0104] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0105] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM, a dynamic random access memory DRAM, a static random access memory SRAM, an enhanced dynamic random access memory EDRAM, a high bandwidth memory HBM, a hybrid memory cube HMC, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0106] 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.

[0107] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A chemical equipment safety monitoring method based on artificial intelligence, characterized in that: include: Obtain the parameter item corresponding to each data point in the time series sequence of the chemical equipment operation data, and determine the time series sequence of the parameter item; ； is the regional synchronization degree of the data points, is the total number of data points in the running data time series, is the time index of the data point, is the time index of the i-th data point in the running data time series except this data point, is the number of parameter item categories, For the The parameter term is the variance of the time series, The data point corresponding to The value of the parameter item, is the number corresponding to the i-th data point in the running data time series except this data point The value of the parameter item, is the linear normalization function, is an exponential function with base e, is an absolute value; The degree of deviation is determined by the regional synchronicity of the data point; the window segment of the data point is obtained, and the interval abnormality of the data point is determined by the data characteristics of each parameter item in the window segment of the data point and the adjacent data points on the left and right sides; ； is the local density after correction of the data points, is the degree of deviation of the data point, is the local density of the data point, is the interval anomaly of the data point; the average of the ratio of the k nearest neighboring data points of the data point to the corrected local density of the data point is recorded as the anomaly of the data point; the safety monitoring of the chemical equipment is achieved by comparing the anomaly of the data point with the preset anomaly threshold.

2. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 1, characterized in that: The method of obtaining the parameter items corresponding to each data point in the time series of the chemical equipment operation data also includes: The values ​​of various parameter items generated by the operation of the equipment at the same acquisition time are combined into a data point, and preprocessed to obtain the time series sequence of chemical equipment operation data. The same type of parameter items are arranged in time series to obtain the parameter item time series sequence.

3. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 1, characterized in that: The degree of deviation of the data points satisfies the relationship: ； is the degree of deviation of the data points, is a hyperparameter, is the regional synchronization degree of the data points, It is the median of regional synchronization in the time series of operating data.

4. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 1, characterized in that: The method for obtaining the window segment of the data point includes: Preset window segment length ; Take the data point as the center and obtain data point and obtain the window segment of the data point.

5. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 1, characterized in that: The interval abnormality of the data point satisfies the relationship: ； is the interval abnormality of the data point, is the number of parameter item categories, The data point corresponding to The value of the parameter item, is the first The parameter item mean, 2 is the number of adjacent data points of this data point, is the number of adjacent data points on one side of the data point in the window segment The mean of the parameter items, is the first Parameter term variance, is the number of adjacent data points on one side of the data point in the window segment Parameter term variance.

6. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 1, characterized in that: The comparison result of the abnormality of the data point and the preset abnormality threshold includes: If the abnormality of a data point is greater than the preset abnormality threshold, the data point is an abnormal data point; otherwise, the data point is a normal data point.

7. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 6 is characterized in that: The safety monitoring of chemical equipment is achieved by comparing the abnormality of the data point with the preset abnormality threshold, including: Through the number of continuous abnormal data points in the time series of chemical equipment operation data, the early warning results of chemical equipment are obtained and processed to achieve safe monitoring of chemical equipment.

8. The method for safety monitoring of chemical equipment based on artificial intelligence according to claim 7 is characterized in that: The method of obtaining the chemical equipment early warning result and processing it according to the number of continuous abnormal data points in the time series of the chemical equipment operation data includes: If the proportion of abnormal data points in the time series of chemical equipment operation data is greater than the preset threshold, the warning result of the chemical equipment is to start the warning and issue a power-off command; otherwise, the warning result of the chemical equipment is that the equipment is normal.

9. A chemical equipment safety monitoring system based on artificial intelligence, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an artificial intelligence-based chemical equipment safety monitoring method according to any one of claims 1 to 8 is implemented.

Images