Hazardous waste treatment process monitoring system, method and equipment based on Internet of Things

Abstract

The invention discloses a hazardous waste treatment process monitoring system, method and equipment based on the Internet of Things, relates to the technical field of data management, and realizes whole-process monitoring and real-time early warning of a hazardous waste treatment process by analyzing and processing sensor data in each monitoring time period. Firstly, data values of all sensors are collected, the fluctuation degree of sensor data is evaluated by calculating indexes such as fluctuation indexes and slope changes, the abnormal monitoring time period is found in time, and abnormal indexes are generated to further judge whether abnormity occurs or not. And then, data correction is performed through the correction reading and the liquid level change model, and finally a composite abnormal alarm is given out to ensure that an operator responds in time. The abnormal detection capability of the system is effectively improved, the accuracy and the real-time performance of the hazardous waste treatment process data are enhanced, and potential problems in the hazardous waste treatment process can be quickly identified and treated, so that the risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and specifically to an Internet of Things-based monitoring system, method, and device for hazardous waste treatment processes. Background Art

[0002] The hazardous waste treatment process monitoring system collects key data in the treatment process in real time by deploying various sensors (such as liquid level sensors, gas concentration sensors, temperature sensors, etc.), and uses wireless communication technology to transmit the data to a central processing platform for real-time analysis and judgment. It can achieve comprehensive monitoring of all links in the hazardous waste treatment process and early warning of potential abnormalities, such as leakage, gas concentration exceeding the standard, and other problems.

[0003] For example, the invention patent with publication number CN115600935A discloses a system for the full-process management of hazardous waste, which relates to the technical field of hazardous waste management. The system includes a hazardous waste registration and declaration module, a hazardous waste storage management module, and a hazardous waste transfer management module; the hazardous waste registration and declaration module is used for the application and approval of hazardous waste disposal, and a hazardous waste electronic declaration form is set in the hazardous waste registration and declaration module to record basic information of hazardous waste, information of hazardous waste transportation units, and information of hazardous waste disposal units. The hazardous waste transfer management module is used to record hazardous waste transportation information and manage the hazardous waste transfer process, and the hazardous waste storage management module is used to manage the hazardous waste storage library and supervise the hazardous waste transfer process. The system of the present invention shares the information of the entire process of hazardous waste disposal, avoids the large accumulation of hazardous waste and the unreasonable use of the hazardous waste storage library due to insufficient information, and improves the disposal efficiency of hazardous waste.

[0004] For example, the invention patent with bulletin number CN117726257B announces a method for treating hazardous waste in industrial parks based on an artificial intelligence big data model. Taking the management party of the industrial park as the operation entity, for the storage process of hazardous waste where humans may make mistakes, an abnormal analysis model can be used to judge the production situation of the source manufacturer and whether mis-packaging may have occurred.

[0005] Based on the above findings, in the existing technical solutions, there may be data instability caused by fluctuations of each sensor in the hazardous waste treatment process due to environmental factors or equipment failures, which may lead to data errors. Summary of the Invention

[0006] Aiming at the deficiencies of the existing technology, the present invention provides an Internet of Things-based monitoring system, method, and device for hazardous waste treatment processes, and solves the problems described in the above background art.

[0007] To achieve the above object, the present invention is realized through the following technical solutions: A hazardous waste treatment process monitoring system based on the Internet of Things, including a hazardous waste monitoring module, which is used to monitor and obtain hazardous waste treatment process data, analyze the hazardous waste treatment process data, and obtain the first fluctuation index of the data of each sensor in each monitoring time period.

[0008] A hazardous waste monitoring anomaly judgment module, which is used to perform anomaly judgment according to the first fluctuation index of the data of each sensor in each monitoring time period, and obtain the judgment result of the hazardous waste treatment process data in each monitoring time period.

[0009] A hazardous waste treatment process data processing module, which is used to process the hazardous waste treatment process data in each monitoring time period according to the judgment result of the hazardous waste treatment process data in each monitoring time period.

[0010] An alarm verification module, which is used to analyze the data after processing the hazardous waste treatment process data in each monitoring time period, and perform a composite anomaly alarm prompt.

[0011] Further, the process of analyzing the hazardous waste treatment process data is as follows: Set several monitoring time periods, obtain the hazardous waste treatment process data in each monitoring time period, including the average data value of the data of each sensor in each monitoring time period, the maximum reading of the sensor data, the minimum reading of the sensor data, and the transient shock amplitude offset of the sensor data, and extract the transient shock amplitude offset definition value of the hazardous waste sensor data stored in the database. Take the average data value of the data of each sensor and the expected data value of the sensor data in each monitoring time period as the data deviation amplitude value of the data of each sensor in each monitoring time period.

[0012] Take the difference between the maximum reading and the minimum reading of the sensor data of each sensor in each monitoring time period as the change amplitude of the data of each sensor in each monitoring time period.

[0013] Further, the process of obtaining the first fluctuation index of the data of each sensor in each monitoring time period is as follows: Extract the data deviation amplitude value of the data of each sensor and the data deviation amplitude definition value of the sensor in each monitoring time period, the change amplitude of the data of each sensor and the data change definition amplitude of the sensor in each monitoring time period, and the transient shock amplitude offset of the sensor data and the transient shock amplitude offset definition value of the sensor data in each monitoring time period for comparison, and introduce an influence correction factor to obtain the first fluctuation index of the data of each sensor in each monitoring time period. The first fluctuation index of the data of each sensor in each monitoring time period is used to represent the fluctuation degree of the data value of each sensor in each monitoring time period.

[0014] Further, the abnormal judgment is performed according to the first fluctuation index of the data of each sensor in each monitoring time period. The specific process is as follows: Extract the first fluctuation index of the data of each sensor in each monitoring time period and compare it with the fluctuation index threshold of the sensor stored in the database. Count the number of sensors whose first fluctuation index of the data of each sensor in each monitoring time period is higher than or equal to the fluctuation index threshold of the sensor, which is recorded as the fluctuation number of the sensors in each monitoring time period.

[0015] Extract the fluctuation number of the sensors in each monitoring time period and compare it with the fluctuation number threshold of the sensors in each monitoring time period stored in the database for judgment. If the fluctuation number of the sensors in a certain monitoring time period is higher than or equal to the fluctuation number threshold of the sensors, mark this monitoring time period as an abnormal monitoring time period, and thus obtain each abnormal monitoring time period.

[0016] Further, according to each abnormal monitoring time period, extract the slope of the data of each sensor in each abnormal monitoring time period and the slope of the data of each sensor in the adjacent time period of each abnormal monitoring time period. Take the deviation value between the slope of the data of each sensor in each abnormal monitoring time period and the slope of the data of each sensor in the adjacent time period of each abnormal monitoring time period as the slope change value of the data of each sensor in each abnormal monitoring time period. Compare the slope change value of the data of each sensor in the abnormal monitoring time period with the slope change definition value of the sensor data, and extract the change amplitude of the data value of each sensor in each abnormal monitoring time period and compare it with the data change definition amplitude of the sensor, and introduce a weight factor to obtain the abnormal index of the data of each sensor in each abnormal monitoring time period.

[0017] Compare the abnormal index of the data of each sensor in each abnormal monitoring time period with the set abnormal index threshold of the sensor data. If the abnormal index of the reading of a certain sensor in a certain abnormal monitoring time period exceeds the set abnormal index threshold, mark the judgment result of the hazardous waste treatment process data in this abnormal monitoring time period as abnormal, and thus obtain the judgment result of the hazardous waste treatment process data in each monitoring time period.

[0018] Further, the processing of the hazardous waste treatment process data in each monitoring time period is as follows: Extract the data value of each sensor in each abnormal monitoring time period, and obtain the weight distribution value of each sensor stored in the database. Multiply the data value of each sensor in each abnormal monitoring time period by the weight distribution value corresponding to this sensor to obtain the corrected value of the data value of each sensor in each abnormal monitoring time period, and replace the data value of each sensor in each abnormal monitoring time period with the corrected value of the data value of this sensor to obtain the final value of the data of each sensor in each abnormal monitoring time period.

[0019] Further, the analysis of the data obtained by processing the hazardous waste treatment process data in each monitoring time period is as follows: Obtain the liquid levels of the leaky tanks at the initial time points and the termination time points of each abnormal monitoring time period. Based on the liquid levels of the leaky tanks at the initial time points and the termination time points of each abnormal monitoring time period, obtain the actual liquid level change values for each abnormal monitoring time.

[0020] According to the final values of the data of each sensor in each abnormal monitoring time period and the data change values of the associated sensor in each abnormal monitoring time, establish a linear physical relationship model of the associated sensor. Substitute the final values of the data of each sensor in each abnormal monitoring time period into the linear physical relationship model of the associated sensor to obtain the theoretical liquid level change values for each abnormal monitoring time period.

[0021] Further, perform a composite abnormal alarm prompt. The specific process is as follows: Obtain the Pearson correlation coefficient between the actual liquid level change value of each abnormal monitoring time and the theoretical liquid level change value in each abnormal monitoring time period. Perform a composite abnormal alarm prompt based on the Pearson correlation coefficient between the actual liquid level change value of each abnormal monitoring time and the theoretical liquid level change value in each abnormal monitoring time period, and push the data to the operation and maintenance terminal.

[0022] Compare the actual liquid level deviation value of each abnormal monitoring time period with the set actual liquid level deviation threshold. If the actual liquid level deviation value of a certain abnormal monitoring time period is higher than or equal to the actual liquid level deviation threshold, perform a composite abnormal alarm prompt for the certain abnormal monitoring time period and push the data to the operation and maintenance terminal.

[0023] The second aspect of the present invention also provides a method for a hazardous waste treatment process monitoring system based on the Internet of Things, including: Monitoring and obtaining hazardous waste treatment process data, analyzing the hazardous waste treatment process data to obtain the first fluctuation index of the data of each sensor in each monitoring time period.

[0024] Perform abnormal judgment based on the first fluctuation index of the data of each sensor in each monitoring time period to obtain the judgment result of the hazardous waste treatment process data in each monitoring time period.

[0025] Process the hazardous waste treatment process data in each monitoring time period according to the judgment result of the hazardous waste treatment process data in each monitoring time period.

[0026] Analyze the data obtained by processing the hazardous waste treatment process data in each monitoring time period and perform a composite abnormal alarm prompt.

[0027] The third aspect of the present invention also provides an electronic device, including: a processor and a memory for storing instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements the hazardous waste treatment process monitoring system based on the Internet of Things.

[0028] The present invention has the following beneficial effects:

[0029] (1) The present invention provides an Internet of Things-based monitoring system, method and device for the hazardous waste treatment process. Through the analysis and processing of sensor data in each monitoring time period, the whole process monitoring and real-time warning of the hazardous waste treatment process are realized. First, the data values of each sensor are collected, and the fluctuation degree of the sensor data is evaluated by calculating indexes such as the fluctuation index and the slope change, and the abnormal monitoring time period is found in time, and an abnormal index is generated to further judge whether an abnormality occurs. Then, data correction is carried out through the corrected reading and the liquid level change model to ensure the data accuracy of the system. Finally, the system issues a composite abnormal alarm according to the set threshold value to ensure that the operator responds in time. The abnormal detection ability of the system is effectively improved, the accuracy and real-time performance of the data are enhanced, and the potential problems in the hazardous waste treatment process can be quickly identified and processed, so as to ensure environmental safety, improve operation and maintenance efficiency and reduce operation risks.

[0030] (2) The present invention calculates the deviation amplitude and change amplitude of the readings of each sensor, and combines the vibration frequency to obtain the first fluctuation index of the data to reflect the fluctuation degree of the data. By comparing the difference between the sensor reading and the expected value, the potential abnormal fluctuations and equipment failures can be effectively revealed, the equipment state can be monitored in real time, the abnormal situation can be found in time, and accurate data support can be provided for subsequent fault judgment and treatment, so as to improve the reliability and safety of the system, optimize operation and maintenance management and reduce environmental risks.

[0031] (3) The present invention calculates the abnormal index of the sensor data in each abnormal monitoring time period, and finally judges whether there is an abnormal situation, provides an accurate abnormal warning function. Through the comprehensive analysis of data fluctuations, it helps to identify potential problems in the treatment process in time, improve the accuracy of fault detection, helps to ensure the safety and stability of the hazardous waste treatment process, and at the same time enhances the intelligence and automation level of the system.

[0032] (4) By correcting and analyzing the data during the abnormal monitoring period, the present invention further improves the accuracy and real-time performance of the monitoring system. First, the system adjusts the sensor readings during the abnormal monitoring period according to the weights of each sensor, performs a difference process with the liquid level change in the liquid leakage tank, and calculates the actual liquid level change value. Then, by establishing a physical relationship model between the liquid level change and the associated sensors, the theoretical liquid level change is compared with the actual liquid level change to obtain the liquid level deviation value, which is compared with the set threshold value, thereby triggering a composite abnormal alarm. This series of steps ensures the accurate detection and timely response to abnormal situations, can effectively eliminate errors and external interferences, enhance the adaptability of the system in complex environments, and improve the safety, stability, and intelligent level of the hazardous waste treatment process. In addition, the composite abnormal alarm system can quickly feedback the abnormal data to the operation and maintenance terminal, providing effective support for timely intervention and fault troubleshooting.

[0033] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic diagram of the modules of the present invention;

[0035] Figure 2 is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0037] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inner", "perimeter", etc. indicating the orientation or positional relationship are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0038] Please refer to Figure 1 , the embodiments of the present invention provide a technical solution: a hazardous waste treatment process monitoring system based on the Internet of Things, including a hazardous waste monitoring module for monitoring and obtaining hazardous waste treatment process data, analyzing the hazardous waste treatment process data, and obtaining the first fluctuation index of the data of each sensor in each monitoring period.

[0039] A hazardous waste monitoring anomaly judgment module, which is used to perform anomaly judgment based on the first fluctuation index of the data of each sensor in each monitoring time period, and obtain the judgment results of the hazardous waste treatment process data in each monitoring time period.

[0040] A hazardous waste treatment process data processing module, which is used to process the hazardous waste treatment process data in each monitoring time period according to the judgment results of the hazardous waste treatment process data in each monitoring time period.

[0041] An alarm verification module, which is used to analyze the data after processing the hazardous waste treatment process data in each monitoring time period, and perform a composite anomaly alarm prompt.

[0042] Specifically, the analysis of the hazardous waste treatment process data is as follows: Set several monitoring time periods, obtain the hazardous waste treatment process data in each monitoring time period, including the average data value of the data of each sensor in each monitoring time period, the maximum reading of the sensor data, the minimum reading of the sensor data, and the transient shock amplitude offset of the sensor data, and extract the transient shock amplitude offset boundary value of the hazardous waste sensor data stored in the database. The average data value of the data of each sensor in each monitoring time period and the expected data value of the sensor data are used as the data deviation amplitude values of the data of each sensor in each monitoring time period.

[0043] It should be noted that the transient shock amplitude offset is a parameter in which the data amplitude of the sensor suddenly jumps due to external instantaneous shocks (such as collisions, vibrations, etc.). In this embodiment, the transient shock amplitude offset of the sensor data refers to the maximum instantaneous amplitude difference value at which the sensor reading deviates from the set normal baseline due to external shocks within the monitoring time period. For example, when the transport vehicle brakes suddenly or the road surface is bumpy, the data of the vibration sensor may suddenly increase from a stable value of 50 mV to 120 mV. At this time, the transient shock amplitude offset is 70 mV. When the transient shock amplitude offset increases significantly, the maximum reading of the sensor data represents the instantaneous shock intensity. Therefore, the maximum reading of the sensor data may also change accordingly, and the average reading may increase due to the superposition of multiple instantaneous peaks. Such abnormal fluctuations may indicate that the equipment or container during transportation has withstood an unexpected physical shock, and may also be caused by the internal components of the sensor being temporarily displaced or the signal being distorted due to severe vibration, resulting in a decrease in the reliability of the reading.

[0044] The difference between the maximum reading and the minimum reading of the sensor data of each sensor in each monitoring time period is used as the change amplitude of the sensor data of each sensor in each monitoring time period.

[0045] Specifically, the first fluctuation index of the data of each sensor in each monitoring time period is obtained, and the specific process is as follows: Extract the data deviation amplitude value of each sensor in each monitoring time period and the data deviation amplitude definition value of the sensor, the change amplitude of the data of each sensor in each monitoring time period and the data change definition amplitude of the sensor, and the transient impact amplitude offset of the sensor data and the transient impact amplitude offset definition value of the sensor data for comparison, and introduce an influence correction factor to obtain the first fluctuation index of the data of each sensor in each monitoring time period. The first fluctuation index of the data of each sensor in each monitoring time period is used to represent the fluctuation degree of the data value of each sensor in each monitoring time period.

[0046] It should be noted that the specific analysis conditions for the first fluctuation index of the data of each sensor in each monitoring time period are as follows:

[0047]

[0048] In the formula, R ij represents the first fluctuation index of the data of the j-th sensor in the i-th monitoring time period, PC ij represents the data deviation amplitude value of the j-th sensor in the i-th monitoring time period, U ij represents the data change amplitude of the j-th sensor in the i-th monitoring time period, ZD ij represents the transient impact amplitude offset of the data of the j-th sensor in the i-th monitoring time period, PC 1 represents the set data deviation amplitude definition value of the sensor, U 1 represents the set data change definition amplitude of the sensor, ZD 1 represents the set transient impact amplitude offset definition value of the sensor data, represents the correction factor corresponding to the set data deviation amplitude value, represents the correction factor corresponding to the set data change amplitude, represents the correction factor corresponding to the set transient impact amplitude offset. i represents the number of each monitoring time period, i = 1, 2, 3,..., n, n represents the total number of monitoring time periods, j represents the number of each sensor, j = 1, 2, 3,..., m, and m represents the total number of sensors.

[0049] It should be noted that during the monitoring process, external disturbances (such as mechanical vibration, temperature change, bumps during transportation, etc.) will cause fluctuations in the sensor values, and the data change amplitude of the sensor refers to the degree of fluctuation of the data within the monitoring time. When the value read by the sensor changes significantly, it indicates that significant vibration or other influencing factors have occurred. Therefore, the difference between the maximum reading and the minimum reading of the sensor data can reflect the fluctuation amplitude of the sensor during this time period.

[0050] It should be noted that each sensor has a specific natural frequency or inherent frequency. When the vibration frequency of the external environment approaches the natural frequency of the sensor, resonance may occur. Resonance will cause the vibration amplitude of the sensor to increase sharply, thereby amplifying the fluctuation of the data. By comparing the actual vibration frequency of the sensor with the reference vibration frequency, it is possible to evaluate whether resonance has occurred, and based on this, it can help to judge the severity of the data fluctuation.

[0051] In a specific embodiment, there is a close relationship among several parameters such as the data deviation amplitude value, the data change amplitude value, and the transient shock amplitude offset of the sensor data. First, as a measure of the systematic deviation between the sensor output and the true value, the data deviation amplitude value directly affects the dynamic range of the data change amplitude value. When the deviation amplitude value of the sensor increases, the fluctuation range of the data output by the sensor expands accordingly, manifested as a significant increase in the data change amplitude value. Secondly, the transient shock amplitude offset directly exacerbates the data deviation amplitude value. Under the stable operating state of the system, the three usually remain at a low level. At this time, the dynamic characteristics of the sensor data meet the expectations; however, when the device encounters abnormal vibration or external interference, the sudden increase in the transient shock amplitude offset will break the original balance, resulting in the cumulative increase of the data deviation amplitude value and the non-steady expansion of the data change amplitude value, forming a negative feedback loop in which the three reinforce each other.

[0052] In a specific embodiment, the value ranges of the correction factor corresponding to the data deviation amplitude value, the correction factor corresponding to the data change amplitude value, and the correction factor corresponding to the transient shock amplitude offset are usually between 0 and 1. The correction factor corresponding to the data deviation amplitude value is determined by a pre-set mapping table. For example, by constructing a mapping relationship between the data deviation amplitude value and the correction factor. The data deviation amplitude value of the sensor detected in real time can be input into this mapping table to quickly obtain the corresponding correction factor. This helps to adjust the sensor readings, reduce data deviation caused by errors or environmental factors, and optimize the accuracy and reliability of the monitoring results. Similarly, the correction factor corresponding to the data change amplitude value can also be determined in a similar way. By analyzing the change amplitude of the readings of each sensor under different working conditions, a mapping table between the data change amplitude value and the correction factor is established. The data change amplitude value of the sensor detected in real time is input into the mapping table. After finding the corresponding correction factor, the real-time data can be corrected. This correction factor can effectively reduce the error caused by the drastic fluctuation of the data and improve the stability and reliability of the data. In addition, the correction factor corresponding to the transient shock amplitude offset can also be determined by establishing a mapping relationship. According to the influence of different transient shock amplitude offsets on the system operation, a mapping table between the transient shock amplitude offset and the correction factor is constructed. When the transient shock amplitude offset is input into this mapping table, the corresponding correction factor can be quickly obtained. This helps to ensure the accuracy of the data.

[0053] Specifically, anomaly judgment is performed based on the first fluctuation index of the data of each sensor in each monitoring time period. The specific process is as follows: Extract the first fluctuation index of the data of each sensor in each monitoring time period and compare it with the fluctuation index threshold of the sensor stored in the database. Count the number of sensors whose first fluctuation index of the data of each sensor in each monitoring time period is higher than or equal to the fluctuation index threshold of the sensor, which is recorded as the fluctuation number of the sensor in each monitoring time period.

[0054] Extract the fluctuation number of the sensor in each monitoring time period and compare and judge it with the fluctuation number threshold of the sensor in each monitoring time period stored in the database. If the fluctuation number of the sensor in a certain monitoring time period is higher than or equal to the fluctuation number threshold of the sensor, mark this monitoring time period as an abnormal monitoring time period, and thus obtain each abnormal monitoring time period.

[0055] It should be noted that if the fluctuation number of the sensor in a certain monitoring time period is lower than the fluctuation number threshold of the sensor, mark this monitoring time period as a normal monitoring time period.

[0056] Specifically, obtain the judgment results of the hazardous waste treatment process data in each monitoring time period. The specific process is as follows: According to each abnormal monitoring time period, extract the slope of the data of each sensor in each abnormal monitoring time period and the slope of the data of each sensor in the adjacent time period of each abnormal monitoring time period. Use the deviation value between the slope of the data of each sensor in each abnormal monitoring time period and the slope of the data of each sensor in the adjacent time period of each abnormal monitoring time period as the slope change value of the data of each sensor in each abnormal monitoring time period. Compare the slope change value of the data of each sensor in each abnormal monitoring time period with the slope change definition value of the sensor data, and extract the change amplitude of the data value of each sensor in each abnormal monitoring time period and compare it with the data change definition amplitude of the sensor, and introduce a weight factor to obtain the anomaly index of the data of each sensor in each abnormal monitoring time period.

[0057] It should be noted that the specific analysis conditions for the anomaly index of the data of each sensor in each abnormal monitoring time period are as follows:

[0058]

[0059] In the formula, E bj represents the anomaly index of the data of the j-th sensor in the b-th abnormal monitoring time period, K bj represents the slope change value of the data of the j-th sensor in the b-th abnormal monitoring time period, V bj represents the data change amplitude of the j-th sensor in the b-th abnormal monitoring time period, V 1$\Delta K_0$ represents the defined amplitude of data change of the set sensor, $\Delta K$ represents the defined value of slope change of the set sensor data, $\beta_2$ represents the weight factor corresponding to the defined data deviation amplitude value, $\beta_1$ represents the weight factor corresponding to the defined slope change value, $b$ represents the number of each abnormal monitoring time period, $b = 1, 2, 3, \cdots, k$, $k$ represents the total number of abnormal monitoring time periods, $j$ represents the number of each sensor, $j = 1, 2, 3, \cdots, m$, and $m$ represents the total number of sensors.

[0060] In a specific embodiment, the change amplitude and slope change value of each sensor data in each monitoring time period do not exist in isolation. There is also a close correlation between the slope change value and the data change amplitude. Under normal circumstances, the readings of the sensor will change gradually over time, and the slope reflects the rate of this change. If the data change amplitude of the sensor is large, then the slope during its change process is usually also large. For example, a sudden change in the reading may mean that a certain parameter in the system has fluctuated violently, resulting in a rapid change in the sensor reading within a short period of time, thus increasing the slope change value. Therefore, a large change amplitude is usually accompanied by a large slope change, which may indicate a drastic change in the state of the monitored object.

[0061] It should be noted that the weight factor corresponding to the data deviation amplitude value of the sensor is determined through a preset mapping relationship. For example, the data deviation amplitude value of the sensor and the preset deviation amplitude values stored in the database form a mapping set. By inputting the data deviation amplitude values of the sensors in each monitoring time period obtained in real time into this mapping set, through the mapping relationship with the preset deviation amplitude values, the corresponding weight factor of the data deviation amplitude value of the sensor can be obtained. Similarly, the weight factor corresponding to the slope change value of the sensor is also determined through a preset mapping relationship. For example, the slope change value of the sensor and the preset slope change values stored in the database form a mapping set. By inputting the slope change values of the sensors in each monitoring time period obtained in real time into this mapping set, through the mapping relationship with the preset slope change values, the corresponding weight factor of the slope change value of the sensor can be obtained, thus providing a more accurate basis for subsequent anomaly detection and alarm mechanisms.

[0062] Compare the anomaly index of each sensor data in each abnormal monitoring time period with the anomaly index threshold of the set sensor data. If the anomaly index of a certain sensor reading in a certain abnormal monitoring time period exceeds the set anomaly index threshold, then mark the judgment result of the hazardous waste treatment process data in this abnormal monitoring time period as abnormal, thereby obtaining the judgment results of the hazardous waste treatment process data in each monitoring time period.

[0063] It should be noted that the adjacent time periods of each abnormal monitoring time period in this embodiment are the next monitoring time periods of each abnormal monitoring time period. If the next monitoring time period of each abnormal monitoring time period is also an abnormal monitoring time period, the continuous abnormal time periods are combined into one abnormal monitoring time period.

[0064] It should be noted that the first fluctuation index of the data is a comprehensive index used to preliminarily measure the degree of fluctuation of the sensor data within a single monitoring time period. It is calculated by comparing the reading deviation amplitude, change amplitude, vibration frequency with the preset boundary values in real time and combining the influence correction factor of the environmental or equipment state; while the abnormal index of each sensor data in each abnormal monitoring time period is a deep index further generated by analyzing the sudden change of the sensor data slope (the trend difference between adjacent time periods) and the data change amplitude for the time periods that have been marked as abnormal. The key of the abnormal index of each sensor data in each abnormal monitoring time period lies in identifying the persistence and trend of the data change within the abnormal time period.

[0065] Specifically, the processing of the hazardous waste treatment process data in each monitoring time period is as follows: extract the data values of each sensor in each abnormal monitoring time period, obtain the weight distribution values of each sensor stored in the database, multiply the data values of each sensor in each abnormal monitoring time period by the corresponding weight distribution value of the sensor to obtain the corrected value of the data value of each sensor in each abnormal monitoring time period, and replace the corrected value of the data value of each sensor in each abnormal monitoring time period with the data value of the sensor to obtain the final value of the data of each sensor in each abnormal monitoring time period.

[0066] It should be noted that the specific analysis conditions for the corrected value of the data value of each sensor in each abnormal monitoring time period are:

[0067]

[0068] In the formula, w bj represents the corrected value of the data value of the j-th sensor in the b-th abnormal monitoring time period, c bj represents the data value of the j-th sensor in the b-th abnormal monitoring time period, y j1 represents the weight distribution value of the j-th sensor, j represents the number of each sensor, j = 1, 2, 3,..., m, and m represents the total number of sensors.

[0069] It should be noted that the sensors used in each abnormal monitoring time period are usually of different types. For example, when monitoring parameters in different dimensions, since the measurement objects, dimensions, and importance of different types of sensors are different, when performing weighted average processing on them, the premise of data standardization and weight rationality needs to be met. Assigning weights according to the sensor type and calculating the correction value are mainly to comprehensively consider the error characteristics and reliability differences of different types of sensors, so as to improve the overall accuracy of the data. The weight assignment value is set based on the contribution of each sensor in abnormal judgment (such as the priority of gas leakage is higher than the temperature and humidity fluctuation). Weighted processing can comprehensively reflect the overall abnormal degree, but it is necessary to avoid deviations caused by inconsistent dimensions or improper weight assignment, which helps to ensure the effective integration of multi-source heterogeneous data.

[0070] Specifically, analyze the data processed from the data of the hazardous waste treatment process in each monitoring time period. The specific process is as follows: Obtain the liquid level of the liquid leakage tank at the initial time point and the end time point of each abnormal monitoring time period, and obtain the actual liquid level change value of each abnormal monitoring time according to the liquid level of the liquid leakage tank at the initial time point and the end time point of each abnormal monitoring time period.

[0071] It should be noted that obtaining the actual liquid level change value of each abnormal monitoring time is specifically to perform a difference process on the liquid level of the liquid leakage tank at the initial time point and the end time point of each abnormal monitoring time period to obtain the actual liquid level change value of each abnormal monitoring time.

[0072] According to the final value of the data of each sensor in each abnormal monitoring time period and the data change value of the associated sensor in each abnormal monitoring time, establish a linear physical relationship model of the associated sensor, and substitute the final value of the data of each sensor in each abnormal monitoring time period into the linear physical relationship model of the associated sensor to obtain the theoretical liquid level change value of each abnormal monitoring time period.

[0073] It should be noted that in this embodiment, the sampling frequency and time granularity of the sensor data and the liquid level data of the liquid leakage tank are the same.

[0074] It should be noted that the time granularity refers to the minimum time interval of data acquisition, that is, the time span between two adjacent sampling points. For example, if the sampling frequency is to collect 1000 sample points per second, the time granularity is 1 millisecond.

[0075] It should be noted that for the linear model of the physical relationship of the associated sensors, the relationship between the liquid level change and the data of the associated sensors should satisfy a linear relationship. For example, the liquid level drop and the reading of the evaporation rate sensor should satisfy that the change in the liquid level is equal to the conversion coefficient between the liquid level change and the evaporation amount multiplied by the evaporation amount multiplied by the time interval. Here, the change in the liquid level, that is, the drop in the liquid level of the leaky tank (the unit is usually meters or other volume units), and the evaporation amount refers to the amount of liquid evaporated, and the unit is "the volume of liquid per unit time". In this embodiment, the value of the conversion coefficient between the liquid level change and the evaporation amount depends on the preset area of the leaky tank, and the value of the time interval in this embodiment is the length of each abnormal monitoring time period.

[0076] It should be noted that the establishment of the linear model of the physical relationship of the associated sensors is based on the linear coupling mechanism between the liquid level change amount and the sensor monitoring parameters. By introducing the conversion coefficient related to the preset leaky tank area and the time interval parameter, the sensor reading is mapped to the theoretical liquid level change value. Specifically, the model can be expressed as the liquid level change amount Δh = k·S·Δt, where k is the conversion coefficient related to the cross-sectional area A of the leaky tank (k = 1 / A, unit m-1), S is the evaporation amount monitored by the sensor (unit m 3 / s), and Δt is the length of the abnormal monitoring time period. This model linearly correlates the current signal (representing the liquid level height) output by the liquid level transmitter with parameters such as the evaporation amount collected by the sensor, and uses the equivalent time sampling method, that is, the length of each abnormal monitoring time period, to ensure the accurate measurement of the time difference Δt. Thus, on the premise of knowing the geometric parameters of the tank, the quantitative conversion between the liquid level change and the sensor data is realized.

[0077] Specifically, for the composite abnormal alarm prompt, the specific process is as follows: Obtain the Pearson correlation coefficient between the actual liquid level change value at each abnormal monitoring time and the theoretical liquid level change value in each abnormal monitoring time period, and perform a composite abnormal alarm prompt according to the Pearson correlation coefficient between the actual liquid level change value at each abnormal monitoring time and the theoretical liquid level change value in each abnormal monitoring time period, and push the data to the operation and maintenance terminal.

[0078] It should be noted that the Pearson correlation coefficient between the actual liquid level change values at each abnormal monitoring time and the theoretical liquid level change values in each abnormal monitoring time period reflects the degree of linear correlation between the actual change of sensor data and the theoretical expectation in the time dimension. The value range is [-1, 1]. The larger the absolute value, the stronger the correlation. The reason for using this coefficient for composite abnormal alarms is that under normal operating conditions, the actual liquid level change should be highly correlated with the theoretical prediction (close to 1). If there are abnormalities such as sensor errors or external interferences, the actual value will deviate from the theoretical trend, resulting in a significant decrease in the correlation coefficient. For example, when the detected correlation coefficient is lower than a preset threshold (such as 0.4), an alarm can be triggered to achieve more accurate abnormal positioning and improve the reliability of abnormal detection.

[0079] It should be noted that the Internet of Things-based monitoring system for hazardous waste treatment processes also includes a database for storing the transient shock amplitude offset boundary values of hazardous waste sensor data, the fluctuation index thresholds of sensors, the fluctuation quantity thresholds of sensors in each monitoring time period, the weights of each sensor, the transient shock amplitude offset boundary values of hazardous waste sensor data, etc.

[0080] The second aspect of the present invention also provides an Internet of Things-based method for monitoring hazardous waste treatment processes, including: monitoring and obtaining hazardous waste treatment process data, analyzing the hazardous waste treatment process data to obtain the first fluctuation index of the data of each sensor in each monitoring time period.

[0081] Perform abnormal judgment based on the first fluctuation index of the data of each sensor in each monitoring time period to obtain the judgment results of the hazardous waste treatment process data in each monitoring time period.

[0082] Process the hazardous waste treatment process data in each monitoring time period according to the judgment results of the hazardous waste treatment process data in each monitoring time period.

[0083] Analyze the processed data of the hazardous waste treatment process data in each monitoring time period and give a composite abnormal alarm prompt.

[0084] The third aspect of the present invention also provides an electronic device, including: a processor and a memory for storing instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements the Internet of Things-based monitoring system for hazardous waste treatment processes.

[0085] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0086] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. An Internet-of-Things-based monitoring system for hazardous waste treatment processes, characterized in that, It includes the following steps: A hazardous waste monitoring module, which is used to monitor and obtain hazardous waste treatment process data, analyze the hazardous waste treatment process data, and obtain the first fluctuation index of the data of each sensor in each monitoring time period; A hazardous waste monitoring anomaly judgment module, which is used to perform anomaly judgment according to the first fluctuation index of the data of each sensor in each monitoring time period, and obtain the judgment result of the hazardous waste treatment process data in each monitoring time period; A hazardous waste treatment process data processing module, which is used to process the hazardous waste treatment process data in each monitoring time period according to the judgment result of the hazardous waste treatment process data in each monitoring time period; An alarm verification module, which is used to analyze the data after processing the hazardous waste treatment process data in each monitoring time period, and perform a composite anomaly alarm prompt.

2. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 1, wherein: The analysis of the hazardous waste treatment process data is specifically as follows: Set several monitoring time periods, obtain the hazardous waste treatment process data in each monitoring time period, including the average data value of the data of each sensor in each monitoring time period, the maximum reading of the sensor data, the minimum reading of the sensor data, and the transient impact amplitude offset of the sensor data, and extract the transient impact amplitude offset definition value of the hazardous waste sensor data stored in the database. The average data value of the data of each sensor in each monitoring time period and the expected data value of the sensor data are used as the data deviation amplitude value of the data of each sensor in each monitoring time period; The difference between the maximum reading and the minimum reading of the sensor data of each sensor in each monitoring time period is used as the change amplitude of the data of each sensor in each monitoring time period.

3. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 2, wherein: The specific process of obtaining the first fluctuation index of the data of each sensor in each monitoring time period is as follows: Extract the comparison between the data deviation amplitude value of the data of each sensor in each monitoring time period and the data deviation amplitude definition value of the sensor, the change amplitude of the data of each sensor in each monitoring time period and the data change definition amplitude of the sensor, and the transient impact amplitude offset of the sensor data and the transient impact amplitude offset definition value of the sensor data, and introduce an influence correction factor to obtain the first fluctuation index of the data of each sensor in each monitoring time period. The first fluctuation index of the data of each sensor in each monitoring time period is used to represent the fluctuation degree of the data value of each sensor in each monitoring time period.

4. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 1, wherein: The specific process of performing anomaly judgment according to the first fluctuation index of the data of each sensor in each monitoring time period is as follows: Extract the first fluctuation index of the data of each sensor in each monitoring time period and compare it with the fluctuation index threshold of the sensor stored in the database. Count the number of sensors whose first fluctuation index of the data of each sensor in each monitoring time period is higher than or equal to the fluctuation index threshold of the sensor, which is recorded as the fluctuation number of the sensors in each monitoring time period; Extract the fluctuation number of the sensors in each monitoring time period and compare it with the fluctuation number threshold of the sensors in each monitoring time period stored in the database for judgment. If the fluctuation number of the sensors in a certain monitoring time period is higher than or equal to the fluctuation number threshold of the sensors, mark this monitoring time period as an abnormal monitoring time period, and thus obtain each abnormal monitoring time period.

5. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 4, wherein: The process of obtaining the judgment results of the hazardous waste treatment process data for each monitoring time period is as follows: According to each abnormal monitoring time period, extract the slopes of the data of each sensor in each abnormal monitoring time period and the slopes of the data of each sensor in the adjacent time periods of each abnormal monitoring time period. Take the deviation values between the slopes of the data of each sensor in each abnormal monitoring time period and the slopes of the data of each sensor in the adjacent time periods of each abnormal monitoring time period as the slope change values of the data of each sensor in each abnormal monitoring time period. Compare the slope change values of the data of each sensor in each abnormal monitoring time period with the slope change threshold value of the sensor data, and extract the amplitude of the data value change of each sensor in each abnormal monitoring time period and compare it with the data change threshold amplitude of the sensor. Introduce a weight factor to obtain the abnormality index of the data of each sensor in each abnormal monitoring time period; Compare the abnormality index of the data of each sensor in each abnormal monitoring time period with the set abnormality index threshold value of the sensor data. If the abnormality index of the reading of a certain sensor in a certain abnormal monitoring time period exceeds the set abnormality index threshold value, mark the judgment result of the hazardous waste treatment process data in that abnormal monitoring time period as abnormal. Thus, the judgment results of the hazardous waste treatment process data for each monitoring time period are obtained.

6. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 1, characterized in that: The process of processing the hazardous waste treatment process data for each monitoring time period is as follows: Extract the data values of each sensor in each abnormal monitoring time period, and obtain the weight distribution values of each sensor stored in the database. Multiply the data values of each sensor in each abnormal monitoring time period by the corresponding weight distribution value of the sensor to obtain the corrected values of the data values of each sensor in each abnormal monitoring time period, and replace the data values of each sensor in each abnormal monitoring time period with the corrected values of the data values of each sensor to obtain the final values of the data of each sensor in each abnormal monitoring time period.

7. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 6, characterized in that: The process of analyzing the processed data of the hazardous waste treatment process data for each monitoring time period is as follows: According to the liquid levels of the leaky tanks at the initial time points and the termination time points of each abnormal monitoring time period, obtain the actual liquid level change values of each abnormal monitoring time; According to the final values of the data of each sensor in each abnormal monitoring time period and the data change values of the associated sensors in each abnormal monitoring time, establish a linear physical relationship model of the associated sensors. Substitute the final values of the data of each sensor in each abnormal monitoring time period into the linear physical relationship model of the associated sensors to obtain the theoretical liquid level change values of each abnormal monitoring time period.

8. The monitoring system for the hazardous waste treatment process based on the Internet of Things according to claim 7, characterized in that: The process of performing a composite abnormal alarm prompt is as follows: Obtain the Pearson correlation coefficient between the actual liquid level change value of each abnormal monitoring time and the theoretical liquid level change value of each abnormal monitoring time period. Perform a composite abnormal alarm prompt according to the Pearson correlation coefficient between the actual liquid level change value of each abnormal monitoring time and the theoretical liquid level change value of each abnormal monitoring time period, and push the data to the operation and maintenance terminal.

9. A method applied to the Internet of Things-based hazardous waste treatment process monitoring system according to any one of claims 1-8, characterized in that: Monitor and obtain the hazardous waste treatment process data, analyze the hazardous waste treatment process data, and obtain the first fluctuation index of the data of each sensor in each monitoring time period; Anomaly judgment is performed based on the first fluctuation index of the data of each sensor in each monitoring time period, and the judgment results of the hazardous waste treatment process data in each monitoring time period are obtained; The hazardous waste treatment process data in each monitoring time period is processed according to the judgment results of the hazardous waste treatment process data in each monitoring time period; The data obtained by processing the hazardous waste treatment process data in each monitoring time period is analyzed, and a composite anomaly alarm prompt is given.

10. An electronic device, characterized in that, It includes: A processor and a memory for storing instructions executable by the processor; When the processor is configured to execute the instructions, the electronic device implements the Internet of Things-based hazardous waste treatment process monitoring system according to any one of claims 1-8.

Images