Chemical production safety risk dynamic monitoring system based on Internet of Things

Abstract

The invention relates to the technical field of dynamic monitoring, in particular to a chemical production safety risk dynamic monitoring system based on the Internet of Things, and the system comprises a data collection module which receives pressure monitoring data, temperature fluctuation data and pipeline flow velocity data measured by a chemical plant sensor in real time, records the time of each data point, and transmits the data to a server; and sorting and organizing the data according to a time sequence. According to the method, by receiving and sorting the chemical plant sensor data and implementing comprehensive analysis of the multi-scale time window, the capture and analysis precision of key parameter fluctuation is remarkably improved, the scientificity of rapid identification and risk prediction of variable fluctuation in the chemical process is promoted, potential risks can be found in advance, and the risk prediction efficiency is improved. Therefore, production strategies are adjusted in real time, production safety is ensured, pre-judgment of future risks is optimized by analyzing long-term trends and periodic changes, the accuracy of risk assessment and the safety of operation are improved, the probability of accidents is reduced, and the safety of personnel and equipment is protected.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic monitoring, and in particular to a dynamic monitoring system for chemical production safety risks based on the Internet of Things. Background Art

[0002] The technical field of dynamic monitoring mainly involves technologies for monitoring and managing the status of processes, facilities, or environments by using real-time data collection and analysis. This field employs various sensors, computational methods, and communication technologies to collect key operating parameters and environmental data in real time. These data are then processed through advanced algorithms to evaluate the performance and safety status of the system in real time. Dynamic monitoring technology plays a key role in improving operational efficiency, preventing accidents, and maintaining system health. It is widely applied in multiple industries such as chemical, energy, transportation, and construction to enhance safety and reliability, while optimizing resource utilization and operation and maintenance strategies.

[0003] Among them, the dynamic monitoring system for chemical production safety risks refers to a real-time monitoring system specifically designed for chemical production, which is used to dynamically evaluate possible safety risks by continuously tracking the operating conditions and environmental changes of chemical facilities. This system integrates a variety of sensors to monitor key safety indicators such as temperature, pressure, and chemical substance concentration, and uses data analysis and machine learning technologies to predict and prevent potential accidents. The purpose of this system is to detect abnormal conditions in a timely manner, reduce the likelihood of accidents, protect the safety of factory assets and employees, and thus improve the overall safety and efficiency of chemical production.

[0004] The commonly used standardized data processing methods in the prior art have poor adaptability in dynamic environments, especially in the monitoring of rapidly changing parameters, where the response is not agile enough. This is particularly evident in chemical production. For example, rapid changes in temperature and pressure may not be captured in a timely manner, resulting in a delay in the response to potential risks. In addition, the lack of the ability to analyze long-term trends and periodic changes in data makes it difficult to conduct effective risk prediction, affecting the timeliness of decision-making and the effectiveness of safety management. These technical limitations may lead to a lag in safety measures in actual operations, increasing the risk of production interruptions and accidents. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose a dynamic monitoring system for chemical production safety risks based on the Internet of Things.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: A dynamic monitoring system for chemical production safety risks based on the Internet of Things includes:

[0007] The data acquisition module receives in real time the pressure monitoring data, temperature fluctuation data, and pipeline flow rate data measured by the sensors in the chemical plant, records the time of each data point, sorts and organizes the data in a time series, integrates the data points including pressure values, temperature readings, and flow rate information, and generates an initial data set;

[0008] Based on the initial data set, the multi-scale time window analysis module sets windows with different time scales, conducts statistical analysis on the data within each time window, calculates the average value, maximum value, and minimum value of the data points, analyzes temperature fluctuations and pressure changes, and generates a multi-scale analysis result;

[0009] The state recognition and transition module uses the multi-scale analysis result to monitor the data fluctuations within multiple time windows, identifies them through the state indicators of whether the data points are abnormal or normal, conducts frequency analysis and classification on the data state changes, and generates a state marking result;

[0010] The trend prediction and anomaly detection module uses the state marking result to analyze the long-term trends and periodic changes in the data. For the change trends of reactor performance and catalyst efficiency, it predicts the abnormal development within a future time period with reference to the data changes, and generates a trend and anomaly prediction result;

[0011] The risk assessment module conducts a risk assessment on the pressure monitoring and temperature fluctuation parameters in the chemical process according to the trend and anomaly prediction result, compares the data deviation under standard operating conditions, conducts quantitative and classification processing on the risks, and generates the current risk state of chemical production.

[0012] The initial data set specifically includes pressure values, temperature readings, and flow rate information. The multi-scale analysis result includes the average value, maximum value, and minimum value of the data points. The state marking result specifically refers to the state indicators of whether the data points are abnormal or normal, the frequency analysis and classification of data state changes. The trend and anomaly prediction result includes long-term trends, periodic changes, change trends of reactor performance, and change trends of catalyst efficiency. The current risk state of chemical production includes quantitative processing of risks and classification processing of risks.

[0013] As a further solution of the present invention, the specific steps for obtaining the initial data set are as follows:

[0014] Receive the pressure monitoring data, temperature fluctuation data, and pipeline flow rate data of the chemical plant, record the time stamps of each data point, and obtain the sensor raw data marked with time;

[0015] According to the sensor raw data marked with time, sort the data in the order of time stamps to generate a time-serialized sensor data set;

[0016] From the time - serialized sensor dataset, calculate the average values of pressure, temperature, and flow rate respectively, using the formulas:

[0017]

[0018] and

[0019]

[0020] and

[0021]

[0022] to obtain the initial dataset;

[0023] where p i represents the pressure value measured at the i - th time point, t i represents the temperature value measured at the i - th time point, f i represents the flow rate value measured at the i - th time point, n represents the total number of data points, P represents the calculated average pressure value, T represents the calculated average temperature value, and F represents the calculated average flow rate value.

[0024] As a further aspect of the present invention, the steps for obtaining the multi - scale analysis result are specifically as follows:

[0025] Based on the initial dataset, set multiple time - scale windows, calculate the average value, maximum value, and minimum value for the data within each window, and generate the statistical parameter definitions for each window;

[0026] Apply the statistical parameter definitions of each window to the corresponding time - window data for statistical processing of the data to obtain the result of the statistical analysis;

[0027] Analyze the result of the statistical analysis, including temperature fluctuations and pressure changes, using the formula:

[0028]

[0029] Calculate the normalized change index for each window to obtain the multi - scale analysis result;

[0030] where x max represents the maximum value of each time window, which is the maximum value selected from all data points within each window, x min represents the minimum value of each time window, which is the minimum value selected from all data points within each window, w represents the weight factor of the time window for adjusting the influence degree of the differential window data, n represents the total number of windows included in the statistical analysis, and S represents the normalized change index.

[0031] As a further solution of the present invention, the step of obtaining the status marking result is specifically as follows:

[0032] Based on the multi-scale analysis result, monitor the data fluctuations within multiple time windows, pay attention to the abnormal fluctuations of data points, detect the abnormal status of each time window, and generate a data status recognition result;

[0033] Analyze the data status recognition result, count the occurrence frequency of each status in multiple time windows, and determine the common degree of the status by comparing the frequency with a preset threshold to obtain a frequency analysis result;

[0034] [[ID=⑨]]Based on the frequency analysis result, classify the data status and use the formula:

[0035]

[0036] Calculate and generate a status marking result;

[0037] Among them, f i represents the frequency of status i, the number of occurrences of multiple identified statuses, s i is the severity weight of status i, used to evaluate the potential impact of each status on the system, C represents the status marking result after classification processing, f j represents the frequency value of the jth data point or status, and n represents the total number of frequencies.

[0038] As a further solution of the present invention, the step of obtaining the trend and anomaly prediction result is specifically as follows:

[0039] Extract key data from the status marking result, analyze the data in each time window to identify long-term trends and periodic fluctuations, and generate a preliminary trend analysis result through statistical methods and trend line analysis;

[0040] Analyze the preliminary trend analysis result, calculate the consistency and deviation of each trend, and use a normalization method to quantify the change trends of reactor performance and catalyst efficiency to generate a change trend analysis result;

[0041] Based on the change trend analysis result, apply a mathematical model to predict the abnormal development in a future time period and use the formula:

[0042]

[0043] Generate a trend and anomaly prediction result;

[0044] Among them, a i represents the trend change coefficient, reflecting the intensity of the trend in each time period, t i is the time variable, and its square is used to emphasize the non-linear growth of the influence of time on the trend, bi is the baseline adjustment coefficient, used to fine-tune the model output to match the initial conditions, and n represents the total number of data points.

[0045] As a further solution of the present invention, the step of obtaining the current risk state of the chemical production is specifically as follows:

[0046] Analyze the trend and anomaly prediction results, extract key data points of pressure monitoring and temperature fluctuations, evaluate the deviation of parameters from the standard operating conditions, and generate a preliminary risk assessment result;

[0047] Analyze the risk assessment result, use statistical methods to quantify the potential impact of each risk factor, identify the highest-risk and lowest-risk areas, and obtain a quantified risk result;

[0048] Based on the quantified risk result, classify the risks, and use the formula:

[0049]

[0050] Calculate the overall risk state and generate the current risk state of the chemical production;

[0051] where w i represents the weight coefficient of each risk parameter, adjusts the degree of influence on the safety of the production process, r i is the quantified value of multiple risk parameters, measuring the risk level of each parameter, d i is the ratio of the deviation from the standard operating conditions, R represents the overall risk state, and n represents the total number of risk factors.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0053] In the present invention, by receiving and sorting the chemical plant sensor data and implementing the comprehensive analysis of multi-scale time windows, the accuracy of capturing and analyzing the fluctuations of key parameters is significantly improved, which promotes the rapid identification of variable fluctuations in the chemical process and the scientific nature of risk prediction, helps to discover potential risks in advance, so as to adjust the production strategy in real time and ensure production safety. By analyzing the long-term trend and periodic changes, the prediction of future risks is optimized, the accuracy of risk assessment and the safety of operation are improved, thereby reducing the probability of accidents and protecting the safety of personnel and equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is the system flow chart of the present invention;

[0055] Figure 2 is the flow chart of the step of obtaining the initial data set of the present invention;

[0056] Figure 3Flowchart of the steps for obtaining the multi-scale analysis results of the present invention;

[0057] Figure 4 Flowchart of the steps for obtaining the status marking results of the present invention;

[0058] Figure 5 Flowchart of the steps for obtaining the trend and anomaly prediction results of the present invention;

[0059] Figure 6 Flowchart of the steps for obtaining the current risk status of chemical production of the present invention. Detailed implementation manners

[0060] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0061] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0062] Embodiment 1

[0063] Please refer to Figure 1 , a dynamic monitoring system for chemical production safety risks based on the Internet of Things includes:

[0064] The data acquisition module receives in real time the pressure monitoring data, temperature fluctuation data and pipeline flow rate data measured by the sensors in the chemical plant, records the time of each data point, sorts and organizes the data in time series, integrates the data points including pressure values, temperature readings and flow rate information, and generates an initial data set;

[0065] The multi-scale time window analysis module sets windows with different time scales based on the initial data set, performs statistical analysis on the data within each time window, calculates the average value, maximum value and minimum value of the data points, analyzes the temperature fluctuation and pressure change, and generates multi-scale analysis results;

[0066] The state recognition and transfer module uses the multi-scale analysis results to monitor the data fluctuations within multiple time windows, identifies them through the status indicators of data points being abnormal or normal, conducts frequency analysis and classification on the changes in data states, and generates state marking results;

[0067] The trend prediction and anomaly detection module uses the state marking results to analyze the long-term trends and periodic changes in the data. For the change trends of reactor performance and catalyst efficiency, it predicts the abnormal development within future time periods with reference to the data changes, and generates trend and anomaly prediction results;

[0068] The risk assessment module conducts risk assessment on the pressure monitoring and temperature fluctuation parameters in the chemical process according to the trend and anomaly prediction results, compares the data deviations under standard operating conditions, quantifies and classifies the risks, and generates the current risk status of chemical production.

[0069] The initial data set specifically includes pressure values, temperature readings, and flow rate information. The multi-scale analysis results include the average value, maximum value, and minimum value of data points. The state marking results specifically refer to the status indicators of data points being abnormal or normal, the frequency analysis and classification of data state changes. The trend and anomaly prediction results include long-term trends, periodic changes, reactor performance change trends, and catalyst efficiency change trends. The current risk status of chemical production includes the quantification process of risks and the classification process of risks.

[0070] Please refer to Figure 2 , and the steps for obtaining the initial data set are specifically as follows:

[0071] Receive the pressure monitoring data, temperature fluctuation data, and pipeline flow rate data of the chemical plant, record the time stamps of each data point, and obtain the time-stamped original sensor data;

[0072] In this process, first initialize the data acquisition system to ensure that all sensors are operating normally and synchronize the time stamp function. Whenever the sensor detects changes in pressure, temperature, or flow rate, the system will automatically record the data and the corresponding time stamp. These data are then sent to the central database through a secure data transmission protocol. The database sorts and marks the received data using the time stamp, which not only ensures the accuracy of the data but also improves the efficiency of subsequent processing.

[0073] Sort the data according to the time stamp order based on the time-stamped original sensor data to generate a time-serialized sensor data set;

[0074] The key here involves the data integration and processing flow. The system will extract all the unprocessed raw data from the database, and then use a sorting algorithm to sort this data according to the timestamp. The sorted data is organized in a time series, which not only makes the data easier to analyze, but also facilitates subsequent data processing and use. Mark and filter the abnormal data to ensure the high quality and reliability of the generated time-series sensor dataset.

[0075] From the time-series sensor dataset, calculate the average values of pressure, temperature, and flow rate respectively, using the formulas:

[0076]

[0077] and

[0078]

[0079] and

[0080]

[0081] to obtain the initial dataset;

[0082] where p i represents the pressure value measured at the i-th time point, t i represents the temperature value measured at the i-th time point, f i represents the flow rate value measured at the i-th time point, n represents the total number of data points, P represents the calculated average pressure value, T represents the calculated average temperature value, and F represents the calculated average flow rate value.

[0083] Formulas:

[0084]

[0085] and

[0086]

[0087] and

[0088]

[0089] The advantage of the formulas is that through the averaging process, the random fluctuations of individual measurement values can be suppressed, providing more stable and reliable data for iterative analysis.

[0090] Detailed explanation of the formulas and the derivation process of formula calculation:

[0091] Set p i = 101, 102, 100, t i = 35, 36, 37, f i= 3.5, 3.6, 3.7, n = 3, and the calculation process is as follows:

[0092]

[0093] The results show that the average pressure is 101 kPa, the average temperature is 36 °C, and the average flow rate is 3.6 m / s. These average values will be used as basic data for system monitoring and performance evaluation.

[0094] Please refer to Figure 3 , and the steps for obtaining the multi-scale analysis results are specifically as follows:

[0095] Based on the initial data set, set multiple time-scale windows, calculate the average value, maximum value, and minimum value for the data within each window, and generate the statistical parameter definitions for each window;

[0096] Based on the initial data set, multiple time-scale windows are set, and the average value, maximum value, and minimum value are calculated for the data within each window. These parameters help capture the fluctuations of the data on different time scales. In this way, the behavior patterns of the data can be analyzed. Analyzing these data can assist in predicting trends or identifying potential anomalies in future time periods. The average value provides the central tendency of the data, while the maximum and minimum values reveal the extreme changes of the data. These statistical measures are basic tools in data analysis for capturing the key features of time series data. By calculating these parameters for each window, the statistical parameter definitions for each window are generated.

[0097] Apply the statistical parameter definitions of each window to the corresponding time-window data for statistical processing of the data to obtain the results of the statistical analysis;

[0098] Apply the statistical parameter definitions to the corresponding time-window data for statistical processing of the data. This process involves analyzing the data points within each time window and calculating their average value, maximum value, and minimum value. These calculations help plot the distribution and volatility of the data within the time window. The results of the statistical analysis provide the basis for subsequent data analysis. These results show the behavior and changes of the data within different time windows and provide a basis for data analysis and decision-making.

[0099] Analyze the results of the statistical analysis, including temperature fluctuations and pressure changes, using the formula:

[0100]

[0101] Calculate the standardized change index for each window to obtain the multi-scale analysis results;

[0102] where, x maxRepresents the maximum value for each time window, which is the maximum value selected from all data points within each window, x min Represents the minimum value for each time window, which is the minimum value selected from all data points within each window. w represents the weight factor of the time window, used to adjust the influence degree of the differentiated window data. n represents the total number of windows included in the statistical analysis. S represents the standardized change index.

[0103] Formula:

[0104]

[0105] The benefit of the formula is that by combining the absolute value of the extreme difference and the weight factor of the time window, it provides a method to measure the volatility of data on a differentiated scale, which helps to identify the key and significant changes in data on a differentiated time scale.

[0106] Detailed explanation of the formula and the derivation process of formula calculation:

[0107] Set in the example, there is the following data: x max = 90, x min = 30, w = 0.5, n = 10. Then the calculation process of the formula is as follows:

[0108] 1. Calculate the absolute value of the extreme difference:

[0109] |x max -x min | = |90 - 30| = 60

[0110] 2. Apply the weight factor:

[0111] 60·0.5 = 30

[0112] 3. Calculate the sum of the weighted differences of all windows and divide by the number of windows:

[0113]

[0114] 4. Calculate the standardized volatility index:

[0115]

[0116] The result shows that within the selected time window of the time series, the standardized volatility index is 1.732, which means that the data within the window shows a medium degree of change after being adjusted by the weight. This value can be used to evaluate the stability or risk level of the data.

[0117] Please refer to Figure 4 , and the specific steps for obtaining the status marking result are as follows:

[0118] Based on the multi-scale analysis results, monitor the data fluctuations within multiple time windows, focus on the abnormal fluctuations of data points, detect the abnormal state of each time window, and generate the data state recognition results;

[0119] After performing multi-scale analysis, the data fluctuations within multiple time windows are monitored. In terms of detecting abnormal states, the statistical threshold method is adopted to identify the abnormal or normal state of data points. This method establishes a threshold model based on past data fluctuations, and any data point exceeding this threshold is regarded as abnormal. This method effectively differentiates between normal and abnormal data states. By marking abnormal data, the occurrence patterns and frequencies of this data can be iteratively analyzed, laying a foundation for subsequent state classification. This method not only enhances the transparency of data processing but also improves the response speed and accuracy of the monitoring system.

[0120] Analyze the data state recognition results, count the occurrence frequencies of each state within multiple time windows, and determine the commonness of the state by comparing the frequencies with the preset thresholds to obtain the frequency analysis results;

[0121] Once multiple states are recognized, the next step is to quantify the commonness of these states through frequency analysis. This process involves counting and analyzing the number of occurrences of each state in different time windows. The statistical results will be used to compare and evaluate the distribution and frequencies of multiple states. Through this information, identify which states are common and which are relatively rare, which helps to understand the universality and abnormality of the states. The frequency analysis results are obtained by collecting and sorting the number of occurrences of each state, which not only assists in analyzing the data fluctuation patterns but also promotes the prediction of future trends.

[0122] Based on the frequency analysis results, classify the data states and adopt the formula:

[0123]

[0124] Calculate and generate the state marking results;

[0125] where, f i represents the frequency of state i, the number of occurrences of the recognized multiple states, s i is the severity weight of state i, used to evaluate the potential impact of each state on the system, C represents the state marking result after classification processing, f j represents the frequency value of the jth data point or state, and n represents the total number of frequencies.

[0126] Formula:

[0127]

[0128] The benefit of the formula is that by introducing a weight coefficient and a standardization process, it reflects the criticality and urgency of the differentiated state, making the state classification more scientific and targeted.

[0129] Detailed explanation of the formula and the derivation process of formula calculation:

[0130] Suppose within the monitoring period of a target, three states are monitored, with frequencies f1 = 30, f2 = 15, f3 = 5 respectively, and the corresponding severity weights are s1 = 0.5, s2 = 1.5, s3 = 2. First, calculate the square root of the sum of squares of the frequencies, that is

[0131]

[0132] Then calculate the weighted sum:

[0133]

[0134] The result shows that this value reflects the comprehensive state index after referring to the frequency and state severity, assisting in quantifying the impact of each state on the whole, so as to conduct effective resource allocation and early warning.

[0135] Please refer to Figure 5 , and the specific steps for obtaining the trend and anomaly prediction results are as follows:

[0136] Extract key data from the state marking results, analyze the data in each time window to identify long-term trends and periodic fluctuations, and generate preliminary trend analysis results through statistical methods and trend line analysis;

[0137] Extract key data from the state marking results, by analyzing the data changes within each time window, identify long-term trends and periodic fluctuations. This process involves continuous monitoring of the data to ensure that the data at each time point can be recorded and analyzed. Use statistical methods to determine the standard deviation and average value of the data. These statistical indicators assist in determining the fluctuation range and central tendency of the data. Apply moving average or exponential smoothing to analyze the potential trends of the data, and generate preliminary trend analysis results through detailed operations, laying a foundation for the next step of analysis.

[0138] Analyze the preliminary trend analysis results, calculate the consistency and deviation of each trend, and use a standardization method to quantify the change trends of reactor performance and catalyst efficiency, generating change trend analysis results;

[0139] Analyze the results of the preliminary trend analysis, including identifying periodicity and long-term trends in the data using time series analysis, and evaluating their potential impact on system performance, including trends in reactor performance and catalyst efficiency, by calculating the persistence and directionality of each trend. This requires the use of an autoregressive moving average model to quantify the stability of the trends and predict their behavior over future time periods, generating trend analysis results that will directly affect the prediction of the system's future behavior.

[0140] Based on the trend analysis results, apply a mathematical model to predict abnormal developments over future time periods using the formula:

[0141]

[0142] Generate trend and anomaly prediction results;

[0143] where a i represents the trend change coefficient, reflecting the intensity of the trend in each time period, t i is the time variable, and its square is used to emphasize the non-linear growth of the impact of time on the trend, b i is the baseline adjustment coefficient, used to fine-tune the model output to match the initial conditions, and n represents the total number of data points.

[0144] Formula:

[0145]

[0146] The advantage of the formula is that by combining the exponential and quadratic terms, it improves the sensitivity of the prediction model to the response of the time variable, simulating the behavior of the data points when approaching the critical point and making the prediction more accurate.

[0147] Detailed explanation of the formula and the derivation process of the formula calculation:

[0148] Set the data as a i = 0.5, t i = 3, b i = 0.1, calculate

[0149]

[0150] Calculate the exponential part

[0151]

[0152] Substitute into the formula for calculation

[0153]

[0154] The results show that at the given t iAt time = 3, the predicted value D is approximately 4.381, indicating that at this time point, the system shows abnormal behavior and requires monitoring and analysis.

[0155] Please refer to Figure 6 , and the steps for obtaining the current risk status of chemical production are specifically as follows:

[0156] Analyze the trend and abnormal prediction results, extract key data points for pressure monitoring and temperature fluctuations, evaluate the deviation of parameters from the standard operating conditions, and generate a preliminary risk assessment result;

[0157] The process of analyzing the trend and abnormal prediction results involves extracting and comparing key data points for pressure monitoring and temperature fluctuations. 720 data points are obtained from the monitoring data of 24 hours per day in the past month. These data points are used to detect the deviation from the standard operating conditions. By calculating the deviation value of each data point, an array M of deviation values is obtained, where n1 is the number of data points, and the average value m1 of the array of deviation values is calculated. A deviation threshold T = m1 is set, and the data points exceeding this threshold are marked. In this way, the system can automatically identify potential risk areas and generate a preliminary risk assessment result. The execution of this step helps to locate the problem area, thus providing data support for subsequent risk management.

[0158] Analyze the risk assessment result, use statistical methods to quantify the potential impact of each risk factor, identify the highest-risk and lowest-risk areas, and obtain a quantified risk result;

[0159] Analyze the risk assessment result. The statistical methods applied in this process include calculating the potential impact of each risk factor and quantifying the risk value of each factor. This involves calculation and model adjustment. A weight coefficient is set for each risk factor. These weight coefficients are obtained based on the degree of influence of past data and expert evaluation. The adjustment of each weight factor is to more accurately reflect its contribution to the overall risk status. Through this refined risk assessment process, the highest-risk and lowest-risk areas can be more accurately distinguished, and a quantified risk result can be obtained accordingly. This process not only improves the efficiency of risk management but also makes the risk response measures more targeted.

[0160] Based on the quantified risk result, classify the risks, using the formula:

[0161]

[0162] Calculate the overall risk status and generate the current risk status of chemical production;

[0163] where, w i represents the weight coefficient of each risk parameter, adjusts the degree of influence on the safety of the production process, ri is the quantified value of multiple risk parameters, measuring the risk level of each parameter, d i is the ratio of the deviation from the standard operating conditions. R represents the overall risk status, and n represents the total number of risk factors.

[0164] Formula:

[0165]

[0166] The benefit of the formula is that by introducing the weight coefficient and adjustment coefficient, the multiple risk parameters are weighted, enabling the influence of each parameter to be adjusted according to the actual situation, thereby achieving the precision of risk assessment. The definitions of the multiple parameters in the formula are as follows, where w i represents the weight of multiple risk factors, r i represents the risk value of multiple factors, d i is the deviation ratio.

[0167] Detailed explanation of the formula and the derivation process of formula calculation:

[0168] Suppose there are three risk factors, and their weights are set as w1 = 0.3, w2 = 0.5, w3 = 0.2 respectively. The risk values of multiple factors are r1 = 20, r2 = 50, r3 = 30 respectively, and the deviation ratios are d1 = 0.1, d2 = 0.3, d3 = 0.2 respectively. Then the formula calculation process is as follows:

[0169]

[0170] The result shows that according to the current risk parameter settings and actual monitoring data, the overall risk rating is 37, which means that risk management and countermeasure formulation can be carried out based on this risk value to ensure the safety of chemical production.

[0171] The above is only the preferred embodiment of the present invention, and it does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A dynamic monitoring system for chemical production safety risks based on the Internet of Things, characterized in that, The system includes: The data acquisition module receives in real time the pressure monitoring data, temperature fluctuation data, and pipeline flow rate data measured by the sensors in the chemical plant, records the time of each data point, sorts and organizes the data in time series, integrates the data points including pressure values, temperature readings, and flow rate information, and generates an initial data set. The multi-scale time window analysis module, based on the initial data set, sets windows with different time scales, performs statistical analysis on the data within each time window, calculates the average value, maximum value, and minimum value of the data points, analyzes the temperature fluctuation and pressure change, and generates a multi-scale analysis result. The state recognition and transition module uses the multi-scale analysis result to monitor the data fluctuation within multiple time windows, identifies through the state indicators of abnormal or normal data points, conducts frequency analysis and classification on the data state change, and generates a state marking result. The trend prediction and anomaly detection module uses the state marking result to analyze the long-term trend and periodic change in the data, and for the change trends of reactor performance and catalyst efficiency, predicts the abnormal development within a future time period with reference to the data change, and generates a trend and anomaly prediction result. The risk assessment module, based on the trend and anomaly prediction result, conducts a risk assessment on the pressure monitoring and temperature fluctuation parameters in the chemical process, compares the data deviation under standard operating conditions, and performs quantification and classification processing on the risk to generate the current risk state of the chemical production.

2. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 1, wherein The initial data set specifically includes pressure values, temperature readings, and flow rate information. The multi-scale analysis result includes the average value, maximum value, and minimum value of the data points. The state marking result specifically refers to the state indicators of abnormal or normal data points, the frequency analysis and classification of the data state change. The trend and anomaly prediction result includes long-term trend, periodic change, reactor performance change trend, and catalyst efficiency change trend. The current risk state of the chemical production includes the quantification processing of the risk and the classification processing of the risk.

3. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 2, wherein, The specific steps for obtaining the initial data set are as follows: Receive the pressure monitoring data, temperature fluctuation data, and pipeline flow rate data of the chemical plant, record the time stamp of each data point, and obtain the sensor raw data marked with time. According to the sensor raw data marked with time, sort the data in the order of time stamp to generate a time-serialized sensor data set. From the time-serialized sensor data set, calculate the average values of pressure, temperature, and flow rate respectively, using the formulas: and and to obtain the initial data set. where p i represents the pressure value measured at the i-th time point, t i represents the temperature value measured at the i-th time point, f i represents the flow rate value measured at the i-th time point, n represents the total number of data points, P represents the calculated average pressure value, T represents the calculated average temperature value, and F represents the calculated average flow rate value.

4. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 3, characterized in that The specific steps for obtaining the multi-scale analysis result are as follows: Based on the initial data set, set multiple time scale windows, calculate the average value, maximum value, and minimum value for the data within each window, and generate the statistical parameter definition for each window. Apply the statistical parameter definition of each window to the corresponding time window data for statistical processing of the data to obtain the result of statistical analysis. Analyze the result of the statistical analysis, including temperature fluctuation and pressure change, using the formula: Calculate the standardized change index for each window to obtain the multi-scale analysis result. Among them, x max represents the maximum value of each time window, which is the maximum value selected from all data points within each window, x min represents the minimum value of each time window, which is the minimum value selected from all data points within each window, w represents the weight factor of the time window, used to adjust the influence degree of the differentiated window data, n represents the total number of windows included in the statistical analysis, and S represents the standardized change index.

5. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 4, wherein, The specific steps for obtaining the state marking result are as follows: Based on the multi-scale analysis results, monitor the data fluctuations within multiple time windows, pay attention to the abnormal fluctuations of data points, detect the abnormal states of each time window, and generate data state recognition results; Analyze the data state recognition results, count the occurrence frequencies of each state in multiple time windows, and determine the commonness of the states by comparing the frequencies with the preset thresholds to obtain frequency analysis results; Based on the frequency analysis results, perform classification processing on the data states, using the formula: Calculate and generate state marking results; Among them, f i represents the frequency of state i, the number of occurrences of multiple recognized states, s i is the severity weight of state i, used to evaluate the potential impact of each state on the system, C represents the state marking result after classification processing, f j represents the frequency value of the j-th data point or state, and n represents the total number of frequencies.

6. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 5, characterized in that The specific steps for obtaining the trend and anomaly prediction results are as follows: Extract key data from the state marking results, analyze the data in each time window to identify long-term trends and periodic fluctuations, and generate preliminary trend analysis results through statistical methods and trend line analysis; Analyze the preliminary trend analysis results, calculate the consistency and deviation of each trend, and use standardization methods to quantify the change trends of reactor performance and catalyst efficiency to generate change trend analysis results; Based on the change trend analysis results, apply a mathematical model to predict the abnormal development in the future time period, using the formula: Generate trend and anomaly prediction results; Among them, a i represents the trend change coefficient, reflecting the intensity of the trend in each time period, and t i is the time variable, and its square is used to emphasize the non-linear growth of the influence of time on the trend. b i is the baseline adjustment coefficient, used to fine-tune the model output to match the initial conditions, and n represents the total number of data points.

7. The dynamic monitoring system for chemical production safety risks based on the Internet of Things according to claim 6, characterized in that The specific steps for obtaining the current risk state of the chemical production are as follows: Analyze the trend and anomaly prediction results, extract the key data points of pressure monitoring and temperature fluctuations, evaluate the deviation of the parameters from the standard operating conditions, and generate preliminary risk assessment results; Analyze the risk assessment results, use statistical methods to quantify the potential impact of each risk factor, and identify the highest-risk and lowest-risk areas to obtain quantified risk results; Based on the quantified risk results, perform classification processing on the risks, using the formula: Calculate the overall risk state and generate the current risk state of the chemical production; Among them, w i represents the weight coefficient of each risk parameter, and adjusts the influence degree on the safety of the production process. r i is the quantified value of multiple risk parameters, measuring the risk level of each parameter. d i is the ratio of the deviation from the standard operating conditions. R represents the overall risk status, and n represents the total number of risk factors.

Images