Chemical storage area data processing method and system based on Internet of Things

Abstract

The present invention relates to the field of chemical storage area data detection technology, and specifically to a chemical storage area data processing method and system based on the Internet of Things. The present invention obtains a standard comparison value based on the data distribution of the chemical storage area monitoring data of the same dimension as the target monitoring data; obtains an analysis value based on the data distribution of different sampling times within the latest preset time period and the data distribution of different preset areas, and obtains abnormal data of the chemical storage area after comparison; obtains an abnormal evaluation index based on the data distribution of the abnormal data of the chemical storage area; and monitors the chemical storage area according to the abnormal evaluation index. The present invention avoids the monitoring data anomaly caused by the periodic changes in the chemical storage area environment, and improves the accuracy of abnormal detection in the chemical storage area.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical storage area data detection, and in particular to a chemical storage area data processing method and system based on the Internet of Things. Background Art

[0002] Chemical storage areas usually store a large number of volatile hazardous chemicals, and the storage conditions of chemicals are strictly required, including temperature, humidity, ventilation, etc. If the requirements are not met, it may cause chemical quality changes, degradation, or even explosions. Therefore, in order to meet the urgent needs of chemical storage area safety management, promote the transformation and upgrading of the chemical industry to intelligence and greenness, establish a management system for chemical storage areas through the application of Internet of Things technology, and comprehensively improve the management level and safety assurance capabilities of chemical storage areas.

[0003] In the prior art, the temperature, humidity and gas concentration of the chemical storage area are often monitored based on thresholds and statistical methods, which can quickly respond to data anomalies in the chemical storage area. However, this method cannot identify periodic changes in the scene environment, such as seasonal changes, weather changes, airflow disturbances and other environmental factors that cause monitoring data anomalies. Such data anomalies do not affect the current actual storage status of the chemical storage area, but will generate monitoring alarms, resulting in false alarms, thereby reducing the accuracy of abnormal monitoring in the chemical storage area. Summary of the invention

[0004] In order to solve the technical problem that the monitoring data anomalies caused by the periodic changes in the scene environment cannot be correctly distinguished, thereby reducing the accuracy of abnormal monitoring in the chemical storage area, the purpose of the present invention is to provide a chemical storage area data processing method and system based on the Internet of Things. The technical scheme adopted is as follows:

[0005] A chemical storage area data processing method based on the Internet of Things, the method comprising:

[0006] In each preset time period, the chemical storage area monitoring data of all dimensions of each preset area in the chemical storage area at each sampling time are obtained;

[0007] In each preset time period, select any chemical storage area monitoring data of any dimension in any preset area at any sampling time as the target monitoring data; obtain the standard comparison value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension as the target monitoring data; obtain the analysis value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension in the same preset area as the target monitoring data in the latest preset time period, and the data distribution characteristics of the chemical storage area monitoring data of the same dimension at the same sampling time as the target monitoring data;

[0008] According to the analysis value and the standard comparison value of the target monitoring data, the abnormal data of the chemical storage area in all the monitoring data of the chemical storage area are obtained; according to the corresponding analysis value and the standard comparison value of each abnormal data of the chemical storage area, and the data distribution of the abnormal data of the chemical storage area, the abnormal evaluation index of each abnormal data of the chemical storage area is obtained;

[0009] The chemical storage area is monitored according to the abnormal evaluation indicators.

[0010] Furthermore, the method for obtaining the standard comparison value includes:

[0011] Obtaining a normal level value of the target monitoring data according to the data quantity distribution of the chemical storage area monitoring data with different data values ​​of the same dimension as the target monitoring data within each preset time period;

[0012] The standard deviation of all chemical storage area monitoring data of the same dimension and at the same sampling time as the target monitoring data within each preset time period is used as the spatial discreteness of the target monitoring data;

[0013] In each preset time period, the ratio of the difference between the target monitoring data and the corresponding normal level value to the spatial discreteness is taken as the first ratio; the first ratios of all preset time periods are averaged to obtain the standard comparison value of the target monitoring data.

[0014] Furthermore, the method for obtaining the normal level value includes:

[0015] In each preset time period, the proportion of the chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data is calculated;

[0016] The proportion of the chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data is taken as the weight of the corresponding data value, and each data value is weighted and averaged to obtain the normal level value of the target monitoring data.

[0017] Furthermore, the method for obtaining the analysis value of the target monitoring data includes:

[0018] According to the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling time within the latest preset time period, a time series reference coefficient corresponding to the latest sampling time is obtained; the time series reference coefficient is positively correlated with the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling time;

[0019] Calculate the average value of the chemical storage area monitoring data of all preset areas in the same dimension as the target monitoring data at the latest sampling time within the latest preset time period as the first mean value; calculate the average value of the chemical storage area monitoring data of all sampling times in the same dimension as the target monitoring data in each preset area before the latest sampling time within the latest preset time period as the second mean value;

[0020] Combining the timing reference coefficient, the first mean value and the second mean value to obtain a monitoring data error;

[0021] In the latest preset time period, the ratio of the difference between the target monitoring data and the monitoring data error to the spatial discreteness is used as the analysis value of the target monitoring data.

[0022] Further, combining the timing reference coefficient, the first mean value and the second mean value to obtain a monitoring data error includes:

[0023] The product of the time series reference coefficient and the first mean is used as the time series data error; the time series reference coefficient is negatively correlated and mapped, and the product of the time series reference coefficient and the second mean is used as the spatial data error;

[0024] The sum of the temporal data error and the spatial data error is used as the monitoring data error.

[0025] Furthermore, the method for obtaining abnormal data of the chemical storage area includes:

[0026] When there is a difference in the integer part between the standard comparison value and the analysis value of the target monitoring data, the target monitoring data is used as abnormal data of the chemical storage area;

[0027] Change the target monitoring data to obtain abnormal data of all chemical storage areas.

[0028] Furthermore, the method for obtaining the abnormal evaluation index includes:

[0029] The difference between the analysis value and the standard comparison value of abnormal data in each chemical storage area is used as the data abnormality degree of the abnormal data in the corresponding chemical storage area;

[0030] According to the data distribution difference between the abnormal data of each chemical storage area and the abnormal data of the chemical storage area at the previous sampling time within a preset neighborhood range, the abnormal change degree of the abnormal data of each chemical storage area is obtained;

[0031] The product of the abnormal change degree of abnormal data in each chemical storage area and the data abnormality degree is normalized to obtain the abnormal evaluation index of the abnormal data in each chemical storage area.

[0032] Furthermore, the method for obtaining the degree of abnormal change includes:

[0033] The preset areas where abnormal data of adjacent chemical storage areas of the same dimension at the same sampling time are located form an abnormal preset area range;

[0034] Select any abnormal data of a chemical storage area as the target abnormal data; take the average of the analysis values ​​of all abnormal data of the chemical storage area in the preset first area of ​​the target abnormal data at the same sampling time as the data abnormality feature of the target abnormal data; obtain the abnormal preset area range where the target abnormal data is located;

[0035] According to the analysis value of the target abnormal data, the data abnormality feature and the abnormal preset area range, the abnormal data feature of the target abnormal data is obtained; the target abnormal data feature and the analysis value of the target abnormal data, the data abnormality feature and the abnormal preset area range are all positively correlated;

[0036] The abnormal data of the chemical storage area of ​​the same dimension at the previous sampling time of the target abnormal data in the same preset area is used as the abnormal data to be analyzed; the average value of the analysis values ​​of all the abnormal data of the chemical storage area in the preset first area of ​​the abnormal data to be analyzed at the same sampling time is used as the data abnormality feature of the abnormal data to be analyzed; and the range of the abnormal preset area where the abnormal data to be analyzed is located is obtained;

[0037] According to the analysis value of the abnormal data to be analyzed, the data abnormality feature and the abnormal preset area range, the abnormal data feature of the abnormal data to be analyzed is obtained; the abnormal data feature to be analyzed and the analysis value of the abnormal data to be analyzed, the data abnormality feature and the abnormal preset area range are all positively correlated;

[0038] The absolute value of the ratio between the target abnormal data feature and the abnormal data feature to be analyzed is used as the abnormal change degree of the target abnormal data;

[0039] Change the target abnormal data to obtain the abnormal change degree of the abnormal data in each chemical storage area.

[0040] Furthermore, monitoring the chemical storage area according to the abnormal evaluation index includes:

[0041] All abnormal data of the chemical storage area where the abnormal evaluation index is greater than a preset threshold value are regarded as real abnormal data;

[0042] The chemical storage area is monitored according to the collection time, preset area and dimension corresponding to the real abnormal data.

[0043] A chemical storage area data processing system based on the Internet of Things includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned chemical storage area data processing method based on the Internet of Things are implemented.

[0044] The present invention has the following beneficial effects:

[0045] In order to comprehensively monitor and control the chemical storage area and avoid the abnormal situation in the chemical storage area caused by the temperature, humidity and air pressure on the chemicals in the chemical storage area, the present invention obtains the chemical storage area monitoring data of all dimensions of each preset area of ​​the chemical storage area; since the chemical storage area has weather changes, in order to avoid the chemical storage area monitoring data collected on a certain day or several days may misjudge the normal chemical storage area monitoring situation, multiple preset time periods are divided, and the chemical storage area monitoring data of all sampling times in all preset time periods are sorted and analyzed; since false alarms may occur when the chemical storage area monitoring data set of a certain preset time period or several preset time periods is compared and screened as normal conditions, the data distribution of all preset time periods is analyzed to obtain the standard comparison value of the target monitoring data; since the latest preset time period may appear The temperature of the time period is different from the temperature of each preset time period in history. At this time, the chemical storage area monitoring data collected in the latest preset time period may not be able to be compared for abnormal monitoring, so the data distribution of the target monitoring data in the latest preset time period at different sampling times and different preset areas is analyzed to obtain the analysis value of the chemical storage area monitoring data at the latest sampling time; according to the analysis value and standard comparison value between the chemical storage area monitoring data, all chemical storage area abnormal data are obtained; since the chemical storage area monitoring data of different preset areas of the chemical storage area need a certain change time to change, it is impossible to judge whether the chemical storage area abnormal data is the chemical storage area monitoring data under the real abnormal situation, so it is necessary to combine the data distribution of the chemical storage area abnormal data, obtain the abnormal evaluation index of the chemical storage area abnormal data for judgment; monitor the chemical storage area through the abnormal evaluation index. The present invention avoids the abnormal monitoring data caused by the periodic changes of the chemical storage area environment and improves the accuracy of abnormal detection of the chemical storage area. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1A flow chart of a chemical storage area data processing method based on the Internet of Things provided by one embodiment of the present invention.

[0048] Figure 2 A flow chart of a method for obtaining a standard comparison value of target monitoring data provided by an embodiment of the present invention.

[0049] Figure 3 A flow chart of a method for obtaining analysis values ​​of target monitoring data provided by an embodiment of the present invention.

[0050] Figure 4 A flow chart of a method for obtaining an abnormal evaluation index provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a chemical storage area data processing method and system based on the Internet of Things proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless defined otherwise, 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 belongs.

[0053] The following is a detailed description of a specific scheme of a chemical storage area data processing method and system based on the Internet of Things provided by the present invention in conjunction with the accompanying drawings.

[0054] See also Figure 1 , which shows a flow chart of a chemical storage area data processing method based on the Internet of Things provided by an embodiment of the present invention, the method comprising:

[0055] Step S1: within each preset time period, obtaining the chemical storage area monitoring data of all dimensions of each preset area of ​​the chemical storage area at each sampling time.

[0056] The embodiment of the present invention is mainly used for monitoring abnormal conditions in chemical storage areas. In order to comprehensively monitor and control all preset areas of the chemical storage area at different times, the monitoring data of the chemical storage area at each preset area of ​​the chemical storage area at each sampling time is obtained; because the temperature, humidity, and air pressure on chemicals in the chemical storage area may cause abnormal conditions in the chemical storage area and thus cause safety accidents, the monitoring data of the chemical storage area in multiple dimensions are obtained to monitor the chemical storage area in multiple dimensions. Since there are periodic changes in the scene environment in the chemical storage area, such as weather changes, each type of weather usually lasts for a period of time, and directly using the monitoring data of the chemical storage area collected on a certain day or a few days may misjudge the normal monitoring situation of the chemical storage area, so in one embodiment of the present invention, a preset time period is set, all sampling moments in history are divided into multiple preset time periods, and the monitoring data of the chemical storage area at all sampling moments in all preset time periods are sorted and analyzed.

[0057] In one embodiment of the present invention, the preset time period is set to 1 day, and the sampling time is set to 1 second. It should be noted that in other embodiments of the present invention, the preset time period and the sampling time can be set by oneself, which is not limited here.

[0058] Among them, the smallest rectangular area that contains the area occupied by the chemical storage area is obtained, and the smallest rectangular area is divided into It should be noted that in other embodiments of the present invention, the preset area can be set voluntarily, which is not limited here.

[0059] Among them, the same number of humidity sensors, temperature sensors, pressure sensors and gas sensors are set in each preset area of ​​the chemical storage area to obtain multi-dimensional monitoring data of the chemical storage area. It should be noted that in other embodiments of the present invention, sensors of other dimensions can also be added to monitor the chemical storage area, which is not limited here.

[0060] The embodiment of the present invention obtains chemical storage area monitoring data, analyzes and obtains the analysis value of the chemical storage area monitoring data and compares it with the standard comparison value, screens out abnormal data of the chemical storage area in the chemical storage area monitoring data, and analyzes the abnormal evaluation index of the abnormal data of the chemical storage area, and then monitors the chemical storage area. The specific analysis process is shown in the following embodiment.

[0061] Step S2: In each preset time period, select any chemical storage area monitoring data of any dimension in any preset area at any sampling time as the target monitoring data; obtain the standard comparison value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension as the target monitoring data; obtain the analysis value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension in the same preset area of ​​the target monitoring data in the latest preset time period, and the data distribution characteristics of the chemical storage area monitoring data of the same dimension at the same sampling time.

[0062] The purpose of the embodiment of the present invention is to monitor whether there is an abnormal situation in the chemical storage area, that is, whether the monitoring data of the chemical storage area is abnormal, so it is necessary to analyze the standards of the chemical storage area under normal conditions and perform comparative screening. In actual situations, the chemical storage area monitoring data collected in each preset time period is very large and relatively chaotic. In addition, there will be differences in the scene environment in each preset time period. For example, there are different weather conditions, and the temperature in the morning and at noon is quite different. It is possible to obtain chemical storage area monitoring data with large differences in the same preset area of ​​the chemical storage area. That is, when using a certain preset time period or a certain number of preset time periods of chemical storage area monitoring data sets as normal conditions for comparative screening, it may cause false alarms of abnormal conditions in the chemical storage area. Therefore, all chemical storage area monitoring data in all preset time periods are analyzed. Therefore, in the embodiment of the present invention, the standard comparison value of the target monitoring data is obtained according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension as the target monitoring data.

[0063] See also Figure 2 , which shows a flow chart of a method for obtaining a standard comparison value of target monitoring data, the method comprising:

[0064] Step S201: obtaining a normal level value of the target monitoring data according to the data quantity distribution of the chemical storage area monitoring data with different data values ​​of the same dimension as the target monitoring data within each preset time period.

[0065] Among all the chemical storage area monitoring data in each preset time period, the number of chemical storage area monitoring data corresponding to abnormal situations is relatively small. For the target monitoring data, among all the chemical storage area monitoring data of the same dimension, the more the number of chemical storage area monitoring data with the same data value, the more likely the chemical storage area monitoring data corresponding to the data value is to be normal, that is, the higher the confidence of the chemical storage area monitoring data corresponding to the data value, the more likely it is to reflect the normal level of the target monitoring data. Therefore, in the embodiment of the present invention, the normal level value of the target monitoring data is obtained according to the data distribution characteristics of all the chemical storage area monitoring data with the same dimension as the target monitoring data.

[0066] Preferably, in one embodiment of the present invention, the method for obtaining the normal level value includes:

[0067] In each preset time period, obtain the percentage of chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data;

[0068] The proportion of the number of chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data is used as the weight of the corresponding data value to perform weighted averaging on each data value to obtain the normal level value of the target monitoring data. In one embodiment of the present invention, the normal level value calculation formula is as follows:

[0069]

[0070] In the formula, Indicates the serial number of the preset time period, Indicates the dimension to which the target monitoring data belongs; Indicates Normal level value of target monitoring data in a preset time period; Indicates The types of monitoring data of the same chemical storage area among all monitoring data of the chemical storage area with the same dimension of the target monitoring data in a preset time period; Indicates the type serial number of the monitoring data of the same chemical storage area; Indicates The number of monitoring data of the same chemical storage area; Indicates The number of all chemical storage area monitoring data with the same dimension as the target monitoring data in a preset time period; Indicates The data values ​​of the same chemical storage area monitoring data.

[0071] In the normal level calculation formula, The more monitoring data of the same chemical storage area, the more Confidence of monitoring data of the same chemical storage area The higher it is, the more likely it is the chemical storage area monitoring data under normal circumstances; the corresponding data values ​​of each type of the same chemical storage area monitoring data are weighted and averaged, and the normal degree and data value of all chemical storage area monitoring data of the same dimension as the target monitoring data are analyzed to reflect the possible normal level of the target monitoring data and obtain the normal level value of the target monitoring data.

[0072] Step S202: taking the standard deviation of all chemical storage area monitoring data of the same dimension at the same sampling time as the target monitoring data within each preset time period as the spatial discreteness of the target monitoring data.

[0073] The discreteness of all chemical storage area monitoring data of the same dimension at the same sampling time as the target monitoring data is analyzed. If the discreteness is smaller, it means that the chemical storage area monitoring data of all preset areas are closer to the same data value. At this time, the data value is more likely to be a normal data value, that is, the target monitoring data is more likely to be normal.

[0074] In an embodiment of the present invention, the standard deviation of all chemical storage area monitoring data of the same dimension at the same sampling time of the target monitoring data is calculated, and the standard deviation is used as the spatial discreteness. The spatial discreteness further characterizes the normality of the target detection data itself.

[0075] Step S203: In each preset time period, the ratio of the difference between the target monitoring data and the corresponding normal level value to the spatial discreteness is taken as the first ratio; the first ratios of all preset time periods are averaged to obtain the standard comparison value of the target monitoring data.

[0076] In order to eliminate the data changes that may be caused by temperature changes, the embodiment of the present invention quantifies the standard value of the target monitoring data by the overall normality of all chemical storage area monitoring data of the same dimension as the target monitoring data in each preset time period. In one embodiment of the present invention, the idea of ​​standard score is used to analyze the variation measurement of the target monitoring data, and then the standard comparison value of the target monitoring data is obtained. The calculation formula of the standard comparison value is as follows:

[0077]

[0078] In the formula, A serial number indicating a preset time period; Indicates the dimension to which the target monitoring data belongs; Indicates the corresponding preset area of ​​target monitoring data; Indicates the corresponding sampling time of the target monitoring data; Indicates the standard comparison value of all target monitoring data of the preset time period; Indicates the number of preset time periods; Indicates The data value of the target monitoring data in a preset time period; Indicates Normal level value of target monitoring data in a preset time period; Indicates The standard deviation of the chemical storage area monitoring data of all preset areas in the same dimension and sampling time as the target monitoring data in a preset time period; Represents the first ratio.

[0079] In the standard comparison value calculation formula, the idea of ​​standard score is used to convert the data value of the target monitoring data within each preset time period into With normal level The deviation is quantized by the first ratio after quantization To indicate the The comparison value of the target monitoring data in a preset time period; in order to reduce the impact of possible special weather on the chemical storage area monitoring data in a preset time period or several preset time periods, the deviation values ​​of all preset time periods are averaged to obtain the standard comparison value of the target monitoring data, that is, the standard comparison value of the chemical storage area monitoring data in each dimension of each preset area at each sampling time in different preset time periods is obtained.

[0080] It should be noted that the standard score is a technical means well known to those skilled in the art and will not be elaborated here.

[0081] The above process obtains the standard comparison value of the target monitoring data through the overall analysis of the chemical storage area monitoring data in all historical preset time periods, thereby determining whether the chemical storage area monitoring data at the current sampling time in the latest preset time period is normal. Since the scene environment in the latest preset time period is different from the scene environment in all historical preset time periods, for example, the temperature in the latest preset time period is different from the temperature in each historical preset time period, the chemical storage area monitoring data collected in the latest preset time period may not be able to be compared for abnormal monitoring. Therefore, in the embodiment of the present invention, the chemical storage area monitoring data at the latest sampling time is converted into a deviation value relative to the chemical storage area monitoring data in the past sampling time and other areas, thereby obtaining the analysis value of the chemical storage area monitoring data at the latest sampling time.

[0082] See also Figure 3 , which shows a flow chart of a method for obtaining an analysis value of target monitoring data, the method comprising:

[0083] Step S211: Obtain a timing reference coefficient corresponding to the latest sampling moment according to the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling moment within the latest preset time period; the timing reference coefficient is positively correlated with the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling moment.

[0084] For the chemical storage area monitoring data within the latest preset time period, the more chemical storage area monitoring data have been collected at the current sampling moment, the greater the reference value of the current sampling moment, and the greater the importance of all chemical storage area monitoring data in the same area at this time, so as the sampling moment increases, the timing reference coefficient continues to increase.

[0085] In one embodiment of the present invention, the timing reference coefficient of the first sampling moment in the latest preset time period is set to 0.1, and as the sampling moment increases, the timing reference coefficient increases in steps of 0.001. It should be noted that in other embodiments of the present invention, the timing reference coefficient can be set voluntarily and is not limited here.

[0086] Among them, a positive correlation means that the dependent variable will increase as the independent variable increases, and the dependent variable will decrease as the independent variable decreases. The specific relationship can be a multiplication relationship, an addition relationship, or the power of an exponential function, which is determined by actual application; a negative correlation means that the dependent variable will decrease as the independent variable increases, and the dependent variable will increase as the independent variable decreases. It can be a subtraction relationship, a division relationship, etc., which is determined by actual application.

[0087] Step S212: Calculate the average value of the chemical storage area monitoring data of the same dimension as the target monitoring data in all preset areas at the latest sampling time within the latest preset time period as the first mean value; calculate the average value of the chemical storage area monitoring data of the same dimension as the target monitoring data in each preset area at each sampling time before the latest sampling time within the latest preset time period as the second mean value.

[0088] In this step, the first mean is used to represent the average value of the chemical storage area monitoring data of all preset areas at the latest sampling time within the latest preset time period. The larger the first mean is, the larger the overall data value of the chemical storage area monitoring data of all preset areas at the latest sampling time is, and the overall data distribution in space is taken into account; the second mean is the average of all chemical storage area monitoring data before the latest sampling time of the same preset area. The second mean is similar to the first mean analysis process, and the second mean takes into account the overall data distribution in time series.

[0089] Step S213: Combine the time series reference coefficient, the first mean and the second mean to obtain the monitoring data error. In the latest preset time period, the difference between the target monitoring data and the monitoring data error and the ratio of the spatial discreteness are used as the analysis value of the target monitoring data.

[0090] Preferably, in one embodiment of the present invention, combining the timing reference coefficient, the first mean value and the second mean value to obtain the monitoring data error includes:

[0091] The product of the time series reference coefficient and the first mean is taken as the time series data error; the time series reference coefficient is negatively correlated and mapped, and the product with the second mean is taken as the spatial data error; the sum of the time series data error and the spatial data error is taken as the monitoring data error.

[0092] In one embodiment of the present invention, the calculation formula of the analysis value of the target monitoring data is as follows:

[0093]

[0094] In the formula, Indicates the latest preset time period; Indicates the analysis value of the target monitoring data in the latest preset time period; Indicates the data value of the target monitoring data in the latest preset time period; Indicates the timing reference coefficient corresponding to the latest sampling moment; represents the second mean; represents the first mean; It represents the standard deviation of the chemical storage area monitoring data of all preset areas with the same dimension and sampling time as the target monitoring data in the latest preset time period; Indicates the monitoring data error of the target monitoring data in the latest preset time period.

[0095] In the calculation formula of the analysis value, the idea of ​​standard fraction is used to analyze The deviation value of the chemical storage area monitoring data relative to the past sampling time and other areas. The more the number of chemical storage area monitoring data collected before the latest sampling time, the higher the second mean of the chemical storage area monitoring data at the past sampling time. The stronger the reference, the more The weight of The weight of Deviation value of chemical storage area monitoring data relative to past sampling time and other areas To reflect The variation measure of The analysis value of .

[0096] Step S3: According to the analysis value and standard comparison value of the target monitoring data, the abnormal data of the chemical storage area in all the monitoring data of the chemical storage area are obtained; according to the corresponding analysis value and standard comparison value of each abnormal data of the chemical storage area, and the data distribution of the abnormal data of the chemical storage area, the abnormal evaluation index of each abnormal data of the chemical storage area is obtained.

[0097] According to the aforementioned process, the analysis value and the standard comparison value of the target monitoring data are obtained, that is, the analysis value and the standard comparison value of the monitoring data of each chemical storage area are obtained. The difference between the analysis value and the standard comparison value can be used to obtain the difference between the data value of each chemical storage area monitoring data relative to the normal situation. Therefore, in an embodiment of the present invention, the abnormal data of the chemical storage area in all the chemical storage area monitoring data are obtained according to the analysis value and the standard comparison value of the target monitoring data.

[0098] Preferably, in one embodiment of the present invention, the method for obtaining abnormal data of a chemical storage area includes:

[0099] When there is a difference in the integer part between the standard comparison value and the analysis value of the target monitoring data, the target monitoring data is used as abnormal data of the chemical storage area; the target monitoring data is changed to obtain abnormal data of all chemical storage areas.

[0100] Due to seasonal changes, weather changes, and temperature changes that may occur in different preset time periods, the multi-dimensional chemical storage area monitoring data such as humidity, temperature, and air pressure in the chemical storage area will change significantly. Since there are phenomena of heat conduction and heat exchange in the chemical storage area, and gas diffusion also gradually spreads from the center to the surroundings, it takes a certain amount of time for the chemical storage area monitoring data in different preset areas of the chemical storage area to change. At this time, it is impossible to determine whether the abnormal data in the chemical storage area is the chemical storage area monitoring data under real abnormal conditions. Therefore, in the embodiment of the present invention, according to the corresponding analysis value and standard comparison value of each chemical storage area abnormal data, and the data distribution of the abnormal data in the chemical storage area, the abnormal evaluation index of each chemical storage area abnormal data is obtained.

[0101] See also Figure 4 , which shows a flow chart of a method for obtaining an abnormal evaluation index, the method comprising:

[0102] Step S301: taking the difference between the analysis value of abnormal data of each chemical storage area and the standard comparison value as the data abnormality degree of the abnormal data of the corresponding chemical storage area.

[0103] The greater the difference between the analysis value of the abnormal data in the chemical storage area and the standard comparison value, the higher the possibility that the abnormal data in the chemical storage area is an abnormal situation.

[0104] Step S302: Obtain the abnormal change degree of the abnormal data of each chemical storage area according to the data distribution difference between the abnormal data of each chemical storage area and the abnormal data of the chemical storage area at the previous sampling time within a preset neighborhood range.

[0105] Preferably, in one embodiment of the present invention, the method for obtaining the degree of abnormal change includes:

[0106] The preset areas where abnormal data of adjacent chemical storage areas of the same dimension at the same sampling time are located constitute the abnormal preset area range.

[0107] Select any abnormal data of a chemical storage area as the target abnormal data; take the average of the analysis values ​​of all abnormal data of chemical storage areas in the adjacent preset first area of ​​the target abnormal data at the same sampling time as the data abnormality feature of the target abnormal data; obtain the abnormal preset area range where the target abnormal data is located;

[0108] In one embodiment of the present invention, around each abnormal data of the chemical storage area, the monitoring data of the chemical storage area in eight adjacent preset areas with the same dimension and the same sampling time as each abnormal data of the chemical storage area are used as the preset first area. It should be noted that the preset first area can be set voluntarily and is not limited here.

[0109] According to the analysis value of the target abnormal data, the data abnormality characteristics and the abnormal preset area range, the abnormal data characteristics of the target abnormal data are obtained; the target abnormal data characteristics and the analysis value of the target abnormal data, the data abnormality characteristics and the abnormal preset area range are all positively correlated;

[0110] The abnormal data of the chemical storage area of ​​the same dimension at the previous sampling time of the target abnormal data in the same preset area is used as the abnormal data to be analyzed; the average value of the analysis values ​​of all the abnormal data of the chemical storage area in the preset adjacent preset first area of ​​the abnormal data to be analyzed at the same sampling time is used as the data abnormality feature of the abnormal data to be analyzed; and the range of the abnormal preset area where the abnormal data to be analyzed is located is obtained;

[0111] According to the analysis value of the abnormal data to be analyzed, the data abnormality characteristics and the abnormal preset area range, the abnormal data characteristics of the abnormal data to be analyzed are obtained; the abnormal data characteristics to be analyzed and the analysis value of the abnormal data to be analyzed, the data abnormality characteristics and the abnormal preset area range are all positively correlated.

[0112] In one embodiment of the present invention, the abnormal data feature calculation formula is as follows:

[0113]

[0114] In the formula, Indicates target abnormal data; Abnormal data features representing target abnormal data; A preset first region representing target abnormal data; Indicates the abnormal preset area range of the target abnormal data; Represents the analysis value of the target abnormal data; Data anomaly features representing target abnormal data; Indicates the radius of the preset abnormal area where the target abnormal data is located; represents the absolute value function.

[0115] In one embodiment of the present invention, the minimum circumscribed circle of the abnormal preset area where the target abnormal data is located is obtained, and the average distance between the edge of the abnormal preset area where the target abnormal data is located and the center of the circle is calculated as the radius of the abnormal preset area where the target abnormal data is located.

[0116] In the abnormal data feature calculation formula, the analysis value of the target abnormal data The larger the value, the more likely the target abnormal data is an abnormal situation; the data abnormality characteristics of the target abnormal data The larger the value is, the more abnormal data there are in the chemical storage area within the first neighborhood of the target abnormal data preset, and the more likely the target abnormal data is an abnormal situation; the radius of the abnormal preset area where the target abnormal data is located The larger it is, the larger the preset abnormal area where the target abnormal data is located. At this time, the larger the abnormal diffusion range is, and the more likely the target abnormal data is an abnormal situation.

[0117] The absolute value of the ratio between the abnormal data characteristics of the target abnormal data and the abnormal data characteristics of the abnormal data to be analyzed is taken as the abnormal change degree of the target abnormal data; wherein, the larger the absolute value of the abnormal data characteristics of the target abnormal data is relative to the abnormal data characteristics of the abnormal data to be analyzed, the greater the fluctuation of the target abnormal data relative to the abnormal data to be analyzed at the previous sampling moment, the farther the abnormal range spreads, and the more likely the target abnormal data is an abnormal situation.

[0118] Step S303: normalize the product of the abnormal change degree of the abnormal data of each chemical storage area and the data abnormality degree to obtain the abnormal evaluation index of the abnormal data of each chemical storage area.

[0119] Preferably, in one embodiment of the present invention, the abnormal evaluation index calculation formula is as follows:

[0120]

[0121] In the formula, Indicates target abnormal data; Anomaly evaluation index representing target abnormal data; Represents the analysis value of the target abnormal data; Indicates the standard comparison value of the target abnormal data; Abnormal data features representing target abnormal data; Abnormal data characteristics representing abnormal data to be analyzed; Indicates the data anomaly degree of the target abnormal data; represents the normalization function; represents the absolute value function.

[0122] In the calculation formula of abnormal evaluation index, the degree of abnormality of target abnormal data is The larger the value, the greater the difference between the analysis value of the target abnormal data and the standard comparison value, and the more likely the target abnormal data is an abnormal situation. The absolute value of the abnormal data feature of the target abnormal data relative to the abnormal data feature of the abnormal data to be analyzed The larger the value is, the greater the fluctuation of the target abnormal data relative to the abnormal data to be analyzed at the previous sampling moment, the farther the abnormal range spreads, and the more likely the target abnormal data is an abnormal situation.

[0123] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of numerical values, which will not be described in detail.

[0124] Change the target abnormal data and obtain the abnormal evaluation index of the abnormal data in each chemical storage area.

[0125] Step S4: Monitor the chemical storage area according to the abnormal evaluation index.

[0126] Preferably, in one embodiment of the present invention, monitoring the chemical storage area according to the abnormal evaluation index includes:

[0127] All abnormal data of chemical storage areas with abnormal evaluation indicators greater than the preset threshold are regarded as real abnormal data;

[0128] When the data processing system in the chemical storage area detects real abnormal data, it will issue an alarm to notify relevant management personnel and trigger relevant control systems to adjust the internal environment of the chemical storage area. It will then record the collection time, preset area, and dimension of the real abnormal data for subsequent analysis and investigation.

[0129] The anomaly detection model and the chemical storage area monitoring data at each sampling moment are regularly updated and optimized to adapt to changes in data distribution and system operation.

[0130] The performance of the chemical storage area is evaluated and monitored at regular intervals, including indicators such as alarm accuracy, false alarm rate, and monitoring failure rate. Based on the evaluation results and combined with the user experience during use, the system is adjusted and improved.

[0131] In one embodiment of the present invention, the preset threshold is set to 0.625. It should be noted that the preset threshold can be set voluntarily and is not limited here.

[0132] At this point, the monitoring of the chemical storage area has been completed.

[0133] In summary, within each preset time period, the chemical storage area monitoring data of all dimensions of each preset area of ​​the chemical storage area at each sampling time is obtained; within each preset time period, the chemical storage area monitoring data of any dimension of any preset area at any sampling time is selected as the target monitoring data; according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension as the target monitoring data, the standard comparison value of the target monitoring data is obtained; according to the data distribution characteristics of the chemical storage area monitoring data of different sampling times of the same dimension of the same preset area as the target monitoring data within the latest preset time period, and the data distribution characteristics of the chemical storage area monitoring data of different preset areas of the same dimension at the same sampling time as the target monitoring data, the analysis value of the target monitoring data is obtained; according to the analysis value and the standard comparison value of the target monitoring data, the abnormal data of the chemical storage area in all the chemical storage area monitoring data is obtained; according to the corresponding analysis value and the standard comparison value of each chemical storage area abnormal data, and the data distribution of the abnormal data of the chemical storage area, the abnormal evaluation index of each chemical storage area abnormal data is obtained; the chemical storage area is monitored according to the abnormal evaluation index.

[0134] An embodiment of the present invention also provides a chemical storage area data processing system based on the Internet of Things, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a chemical storage area data processing method based on the Internet of Things can be implemented.

[0135] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0136] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A chemical storage area data processing method based on the Internet of Things, characterized in that: The method comprises: In each preset time period, the chemical storage area monitoring data of all dimensions of each preset area in the chemical storage area at each sampling time are obtained; In each preset time period, select any chemical storage area monitoring data of any dimension in any preset area at any sampling time as the target monitoring data; obtain the standard comparison value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension as the target monitoring data; obtain the analysis value of the target monitoring data according to the data distribution characteristics of the chemical storage area monitoring data of the same dimension in the same preset area as the target monitoring data in the latest preset time period, and the data distribution characteristics of the chemical storage area monitoring data of the same dimension at the same sampling time as the target monitoring data; According to the analysis value and the standard comparison value of the target monitoring data, the abnormal data of the chemical storage area in all the monitoring data of the chemical storage area are obtained; according to the corresponding analysis value and the standard comparison value of each abnormal data of the chemical storage area, and the data distribution of the abnormal data of the chemical storage area, the abnormal evaluation index of each abnormal data of the chemical storage area is obtained; Monitoring the chemical storage area according to the abnormal evaluation index; The method for obtaining the standard comparison value includes: Obtaining a normal level value of the target monitoring data according to the data quantity distribution of the chemical storage area monitoring data with different data values ​​of the same dimension as the target monitoring data within each preset time period; The standard deviation of all chemical storage area monitoring data of the same dimension and at the same sampling time as the target monitoring data within each preset time period is used as the spatial discreteness of the target monitoring data; In each preset time period, the ratio of the difference between the target monitoring data and the corresponding normal level value to the spatial discreteness is taken as a first ratio; the first ratios of all preset time periods are averaged to obtain a standard comparison value of the target monitoring data; The method for obtaining the analysis value of the target monitoring data includes: According to the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling time within the latest preset time period, a time series reference coefficient corresponding to the latest sampling time is obtained; the time series reference coefficient is positively correlated with the number of chemical storage area monitoring data of the same dimension as the target monitoring data collected before the latest sampling time; Calculate the average value of the chemical storage area monitoring data of all preset areas in the same dimension as the target monitoring data at the latest sampling time within the latest preset time period as the first mean value; calculate the average value of the chemical storage area monitoring data of all sampling times in the same dimension as the target monitoring data in each preset area before the latest sampling time within the latest preset time period as the second mean value; Combining the timing reference coefficient, the first mean value and the second mean value to obtain a monitoring data error; In the latest preset time period, the ratio of the difference between the target monitoring data and the monitoring data error to the spatial discreteness is used as the analysis value of the target monitoring data; The method for obtaining the abnormal evaluation index includes: The difference between the analysis value and the standard comparison value of abnormal data in each chemical storage area is used as the data abnormality degree of the abnormal data in the corresponding chemical storage area; According to the data distribution difference between the abnormal data of each chemical storage area and the abnormal data of the chemical storage area at the previous sampling time within a preset neighborhood range, the abnormal change degree of the abnormal data of each chemical storage area is obtained; The product of the abnormal change degree of abnormal data in each chemical storage area and the data abnormality degree is normalized to obtain the abnormal evaluation index of the abnormal data in each chemical storage area.

2. According to a method for processing chemical storage area data based on the Internet of Things in claim 1, it is characterized in that: The method for obtaining the normal level value includes: In each preset time period, the proportion of the chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data is calculated; The proportion of the chemical storage area monitoring data with different data values ​​under the same dimension of the target monitoring data is taken as the weight of the corresponding data value, and each data value is weighted and averaged to obtain the normal level value of the target monitoring data.

3. The method for processing chemical storage area data based on the Internet of Things according to claim 1 is characterized in that: Combining the timing reference coefficient, the first mean value, and the second mean value to obtain a monitoring data error includes: The product of the time series reference coefficient and the first mean is used as the time series data error; the time series reference coefficient is negatively correlated and mapped, and the product of the time series reference coefficient and the second mean is used as the spatial data error; The sum of the temporal data error and the spatial data error is used as the monitoring data error.

4. The method for processing chemical storage area data based on the Internet of Things according to claim 1 is characterized in that: The method for obtaining abnormal data of the chemical storage area includes: When there is a difference in the integer part between the standard comparison value and the analysis value of the target monitoring data, the target monitoring data is used as abnormal data of the chemical storage area; Change the target monitoring data to obtain abnormal data of all chemical storage areas.

5. The method for processing chemical storage area data based on the Internet of Things according to claim 1 is characterized in that: The method for obtaining the abnormal change degree includes: The preset areas where abnormal data of adjacent chemical storage areas of the same dimension at the same sampling time are located form an abnormal preset area range; Select any abnormal data of a chemical storage area as the target abnormal data; take the average of the analysis values ​​of all abnormal data of the chemical storage area in the preset first area of ​​the target abnormal data at the same sampling time as the data abnormality feature of the target abnormal data; obtain the abnormal preset area range where the target abnormal data is located; According to the analysis value of the target abnormal data, the data abnormality feature and the abnormal preset area range, the abnormal data feature of the target abnormal data is obtained; the target abnormal data feature and the analysis value of the target abnormal data, the data abnormality feature and the abnormal preset area range are all positively correlated; The abnormal data of the chemical storage area of ​​the same dimension at the previous sampling time of the target abnormal data in the same preset area is used as the abnormal data to be analyzed; the average value of the analysis values ​​of all the abnormal data of the chemical storage area in the preset first area of ​​the abnormal data to be analyzed at the same sampling time is used as the data abnormality feature of the abnormal data to be analyzed; and the range of the abnormal preset area where the abnormal data to be analyzed is located is obtained; According to the analysis value of the abnormal data to be analyzed, the data abnormality feature and the abnormal preset area range, the abnormal data feature of the abnormal data to be analyzed is obtained; the abnormal data feature to be analyzed and the analysis value of the abnormal data to be analyzed, the data abnormality feature and the abnormal preset area range are all positively correlated; The absolute value of the ratio between the target abnormal data feature and the abnormal data feature to be analyzed is used as the abnormal change degree of the target abnormal data; Change the target abnormal data to obtain the abnormal change degree of the abnormal data in each chemical storage area.

6. The method for processing chemical storage area data based on the Internet of Things according to claim 1 is characterized in that: The chemical storage area is monitored according to the abnormal evaluation indicators, including: All abnormal data of the chemical storage area where the abnormal evaluation index is greater than a preset threshold value are regarded as real abnormal data; The chemical storage area is monitored according to the collection time, preset area and dimension corresponding to the real abnormal data.

7. A chemical storage area data processing system based on the Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a chemical storage area data processing method based on the Internet of Things as described in any one of claims 1 to 6 are implemented.

Images