Intelligent management method and system for chemical industry park

Abstract

The invention relates to the technical field of chemical industry park management, and provides a chemical industry park intelligent management method and system, and the method comprises the steps: constructing a risk assessment model; acquiring real-time data in the chemical industry park from a plurality of data sources in real time; performing multi-source fusion processing on the data set to obtain a fused data set; based on the fused data set, real-time risk assessment is carried out through a risk assessment model, and a management decision is made; instructions are sent to related personnel and corresponding devices through the information transmission device, and corresponding tasks are completed. A risk assessment model is constructed by integrating a plurality of data sources, and multi-source fusion processing is performed based on real-time and historical data, so that comprehensive and real-time risk assessment and management decision of a chemical industry park are realized, and the problem that an existing intelligent chemical industry park management system is lack of a comprehensive intelligent management method is solved. And the management requirements of the chemical industry park are difficult to comprehensively meet.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical park management, and in particular to a chemical park intelligent management method and system. Background Art

[0002] With the rapid development of the chemical industry, chemical parks, as important gathering places for chemical enterprises, have become increasingly complex and risky to manage. Traditional chemical park management relies heavily on manual inspections and paper records. This management model is not only inefficient, but also difficult to achieve comprehensive, real-time monitoring and timely response to all aspects of the park. At the same time, the high risk and environmental pollution pressure of the chemical industry require management to have higher accuracy and foresight to effectively prevent and respond to potential safety and environmental issues.

[0003] In order to improve the management level of chemical parks and achieve safe, environmentally friendly, efficient and sustainable development, smart chemical park management came into being. Smart chemical park management integrates modern information technology and Internet of Things technology to achieve real-time collection, processing and analysis of various data in the park, providing strong data support for park management. However, most of the smart chemical park management systems on the market currently focus on single-aspect management, such as safety management or environmental management, lack comprehensive intelligent management methods, and are difficult to fully meet the management needs of chemical parks. Summary of the invention

[0004] The purpose of the present invention is to provide a chemical park intelligent management system and method, system, equipment and storage medium, aiming to solve the problem that most of the existing smart chemical park management systems focus on single-aspect management, lack comprehensive intelligent management methods, and are difficult to fully meet the management needs of chemical parks.

[0005] The embodiments of the present invention are implemented by the following technical solutions:

[0006] A chemical park intelligent management method comprises the following steps:

[0007] Collect historical data in the park through several data sources to obtain an initial historical data set, pre-process the initial historical data set to obtain an integrated historical data set, and build a risk assessment model based on the integrated historical data set;

[0008] Collecting real-time data in the chemical park from several data sources in real time to obtain an initial real-time data set, and preprocessing the initial real-time data set to obtain an integrated real-time data set;

[0009] Perform multi-source fusion processing on the integrated real-time data set and the integrated historical data set to obtain a fused data set;

[0010] Based on the fused data set, real-time risk assessment is performed through the risk assessment model, and management decisions are made based on the assessment results;

[0011] Execute management decisions, send instructions to relevant personnel and corresponding equipment through information transmission equipment, and the relevant personnel and corresponding equipment complete the corresponding tasks according to the instructions.

[0012] Optionally, the plurality of data sources include:

[0013] Environmental monitoring equipment, used to monitor environmental parameters within the chemical park;

[0014] Process equipment status monitoring equipment, used to monitor the operating status of each process equipment in the chemical park;

[0015] Personnel activity monitoring equipment, used to monitor the activities of personnel in the chemical park;

[0016] Inventory management systems to provide inventory data on chemical raw materials, products and waste;

[0017] Energy management systems to monitor energy consumption;

[0018] Safety management system, used to record safety accidents, hidden danger inspections, safety training and related data.

[0019] Optionally, the specific process of constructing a risk assessment model based on integrating historical data sets is:

[0020] Extract risk features from the integrated historical data set to obtain a risk feature data set;

[0021] Set risk level classification standards and set corresponding labels for each risk level. According to the risk level classification standards and the labels corresponding to each risk level, perform sample annotation on the risk feature data set and verify the accuracy of the labels to obtain a sample annotation set.

[0022] The sample annotation set is divided into training set, validation set and test set. Based on the training set, validation set and test set, a risk assessment model is constructed through machine learning algorithm.

[0023] Optionally, the specific process of constructing the risk assessment model through the machine learning algorithm is:

[0024] Let X be a sample in the training set, X contains features, as shown in the following formula (1):

[0025] X={x1,x2,…,x K} (1)

[0026] The output of the risk assessment model is the predicted risk level corresponding to the sample The expression of the risk assessment model is shown in the following formula (2):

[0027]

[0028] Where M is the total number of decision trees, T m (X) is the prediction result of the mth decision tree for sample X;

[0029] The objective function is constructed based on the fused data set, and the risk assessment model is optimized through the objective function.

[0030] Optionally, the specific process of constructing the objective function based on the fused data set and optimizing the risk assessment model through the objective function is:

[0031] The objective function is constructed based on the fused data set. The expression of the objective function is shown in the following formula (3):

[0032]

[0033] Among them, n is the number of training samples, y i is the true risk level of the i-th sample, is the predicted risk level of the i-th sample, is the loss function, Ω(T m ) is the regularization term of the mth decision tree.

[0034] Optionally, the specific expression of the loss function is shown in the following formula (4):

[0035]

[0036] Where D is the total number of categories, y id is an indicator, which is 1 when the i-th sample belongs to the d-th class, otherwise it is 0. is the prediction score of the ith sample in the dth category, and j is a category index.

[0037] Optionally, the specific expression of the regularization term is as shown in the following formula (5):

[0038]

[0039] Among them, G is the number of leaf nodes of the decision tree, ω k is the weight of the kth leaf node, and γ and λ are both regularization coefficients.

[0040] Optionally, the integrated real-time data set and the integrated historical data set are subjected to multi-source fusion processing to obtain a fused data set. The specific process is:

[0041] Assume that the feature vector of the integrated real-time data set is as shown in formula (6):

[0042]

[0043] Assume that the feature vector of the integrated historical data set is as shown in the following formula (7):

[0044]

[0045] Then the feature vector of the fused data set is as shown in formula (8):

[0046]

[0047] Among them, select represents the feature selection function, and transform represents the feature transformation function;

[0048] [X real-time ,X historical ] indicates direct concatenation, which concatenates the feature vectors in the real-time data set and the historical data set into a new vector;

[0049] [select(X real-time ),select(X historical )] indicates that the features are selected and then stitched together. Before stitching, some features are selected for stitching according to their importance;

[0050] transform([X real-time ,X historical ]) indicates concatenation after transformation. For features with high correlation or redundancy, concatenation is performed after dimensionality reduction.

[0051] Optionally, a feedback mechanism is established to collect information and data from the relevant personnel and corresponding equipment during the task execution process, and the risk assessment model is continuously optimized and updated based on the feedback information and data from the task execution process, and the management decision-making process and strategy are optimized and improved according to the execution effect of the management decision.

[0052] The present invention also provides a chemical park intelligent management system for implementing the method, comprising:

[0053] Data collection and integration module, used to collect data from several data sources and perform preprocessing and integration, provide a data visualization interface, and display real-time and historical data;

[0054] A model building module is used to build a risk assessment model and train the risk assessment model to obtain a risk assessment model that meets system requirements;

[0055] The risk assessment module is used for risk assessment and provides real-time display and query functions of risk assessment results;

[0056] Management decision making module, used to make management decisions based on risk assessment results and prediction rules;

[0057] The instruction execution and feedback module is used to send instructions to relevant personnel and equipment, monitor the execution status, collect feedback information during the execution process, and conduct evaluation and optimization.

[0058] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0059] The present invention provides a chemical park intelligent management method and system, the method comprising the steps of: collecting historical data in the park through several data sources to obtain an initial historical data set, preprocessing the initial historical data set to obtain an integrated historical data set, and constructing a risk assessment model based on the integrated historical data set; collecting real-time data in the chemical park from several data sources in real time to obtain an initial real-time data set, preprocessing the initial real-time data set to obtain an integrated real-time data set; multi-source fusion processing of the integrated real-time data set and the integrated historical data set to obtain a fused data set; based on the fused data set, performing real-time risk assessment through a risk assessment model, and making management decisions according to the assessment results; executing management decisions, sending instructions to relevant personnel and corresponding equipment through information transmission equipment, and the relevant personnel and corresponding equipment complete corresponding tasks according to the instructions. By integrating multiple data sources, constructing a risk assessment model, and performing multi-source fusion processing based on real-time and historical data, comprehensive and real-time risk assessment and management decision-making for chemical parks are achieved. This method not only improves the intelligence level of chemical park management, but also enhances the accuracy and predictability of management, providing strong guarantees for the safety, environmental protection and efficient operation of chemical parks, thus solving the problem that most existing smart chemical park management systems focus on single-aspect management, lack comprehensive intelligent management methods, and are difficult to fully meet the management needs of chemical parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a schematic diagram of the process of the intelligent management method for a chemical park in Example 1 of the present invention;

[0061] Figure 2 This is a schematic diagram of the structure of the chemical park intelligent management system in Example 2 of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0063] Example 1

[0064] A chemical park intelligent management method comprises the following steps:

[0065] Step 1: Collect historical data in the park through several data sources to obtain an initial historical data set, pre-process the initial historical data set to obtain an integrated historical data set, and build a risk assessment model based on the integrated historical data set;

[0066] In this embodiment, data sources include but are not limited to:

[0067] Environmental monitoring equipment is used to monitor environmental parameters within the chemical park, including but not limited to air quality (such as PM2.5, VOCs concentration, toxic gas concentration), water quality (such as pH value, heavy metal content, chemical oxygen demand), noise level, soil pollution, etc. These data are helpful to assess the environmental pollution status of the park, timely discover and warn of potential environmental risks, and provide a scientific basis for environmental management.

[0068] Process equipment status monitoring equipment is used to monitor the operating status of various process equipment in the chemical park, including equipment parameters such as temperature, pressure, flow, vibration, speed, energy consumption, as well as equipment power on and off status, fault alarm information, etc.; through real-time monitoring of equipment status, equipment failures and potential safety hazards can be discovered in time, production accidents can be prevented, and production efficiency and equipment reliability can be improved.

[0069] Personnel activity monitoring equipment is used to monitor the activities of personnel in the chemical park, such as personnel location, entry and exit records, working hours, helmet wearing, etc.; these data help to strengthen the safety management of the park, ensure the safety of personnel, prevent illegal operations and unsafe behaviors, and also help with personnel dispatch and rapid evacuation in emergency situations.

[0070] The inventory management system is used to provide inventory data on chemical raw materials, products and waste, including information such as material type, quantity, storage location, shelf life, safety labeling, etc. Inventory management data is of great significance for ensuring production continuity, preventing material shortages or surpluses, and optimizing resource allocation. At the same time, inventory management of hazardous chemicals is directly related to the safety and environmental protection of the park.

[0071] The energy management system is used to monitor energy consumption, including the consumption of water, electricity, gas and other energy, consumption time periods, energy efficiency and other indicators; through energy management data, the energy utilization efficiency of the park can be evaluated, energy-saving potential can be discovered, energy-saving measures can be formulated, and production costs can be reduced. At the same time, it also helps to reduce carbon emissions and promote green production.

[0072] The safety management system is used to record safety accidents, hidden danger inspections, safety training and related data, including the time, location, cause, loss, handling results, etc. of the accident, as well as the frequency of hidden danger inspections, problems found, rectification, etc., as well as safety training plans, participants, training content and other information; safety management data is an important part of the park's safety management, it helps to summarize experience and lessons, improve the safety management system, improve employees' safety awareness, and reduce the occurrence of safety accidents. At the same time, it also provides an important reference for making management decisions.

[0073] Extract risk features from the integrated historical data set to obtain a risk feature data set;

[0074] Set risk level classification standards and set corresponding labels for each risk level. According to the risk level classification standards and the labels corresponding to each risk level, perform sample annotation on the risk feature data set and verify the accuracy of the labels to obtain a sample annotation set.

[0075] The sample annotation set is divided into a training set, a validation set, and a test set. Based on the training set, the validation set, and the test set, a risk assessment model is constructed through a machine learning algorithm.

[0076] Let X be a sample in the training set, X contains features, as shown in the following formula (1):

[0077] X={x1,x2,…,x K} (1)

[0078] The output of the risk assessment model is the predicted risk level corresponding to the sample The expression of the risk assessment model is shown in the following formula (2):

[0079]

[0080] Among them, M is the total number of decision trees, T m (X) is the prediction result of the mth decision tree for sample X;

[0081] The objective function is constructed based on the fused data set, and the risk assessment model is optimized through the objective function.

[0082] The expression of the objective function is shown in formula (3):

[0083]

[0084] Among them, n is the number of training samples, y i is the true risk level of the i-th sample, is the predicted risk level of the i-th sample, is the loss function, Ω(T m) is the regularization term of the mth decision tree.

[0085] The specific expression of the loss function is shown in the following formula (4):

[0086]

[0087] Where D is the total number of categories, y id is an indicator, which is 1 when the i-th sample belongs to the d-th class, otherwise it is 0. is the prediction score of the ith sample in the dth category, and j is a category index.

[0088] The specific expression of the regularization term is shown in the following formula (5):

[0089]

[0090] Among them, G is the number of leaf nodes of the decision tree, ω k is the weight of the kth leaf node, and γ and λ are both regularization coefficients.

[0091] Step 2: real-time data in the chemical park is collected from several data sources in real time to obtain an initial real-time data set, and the initial real-time data set is preprocessed to obtain an integrated real-time data set;

[0092] Real-time data collection

[0093] First, it is necessary to clarify which data sources will be used for real-time data collection. Use appropriate data collection techniques to extract data from various data sources, including pulling data through API interfaces, subscribing to data using IoT protocols such as MQTT, querying data directly from databases, or obtaining data through file transfer. The frequency of data collection should be determined based on the importance and real-time requirements of the data. High-frequency collection may be required for key parameters. The collected real-time data is first received and stored in a temporary or real-time data warehouse to form an initial real-time data set. This data usually includes timestamps, data source identifiers, data values, and possible metadata. The initial real-time data set is cleaned to eliminate erroneous, duplicated, or inconsistent data. This may include processing missing values, correcting outliers, and removing invalid or irrelevant data items. Convert the data into a unified format and standard units for subsequent processing and analysis. This may include data type conversion, timestamp unification, data encoding standardization, etc. If the data volume is huge, it may be necessary to aggregate the data to reduce the number of data points, or compress the data to save storage space and network bandwidth. Aggregation operations may be performed based on time windows. At this stage, although it is mainly pre-processing within a single data source, it can also prepare for subsequent multi-source data fusion. For example, the same type of data from different monitoring points can be associated so that the situation of the entire park can be comprehensively considered in subsequent analysis. Implement data quality control measures to ensure that the quality of pre-processed data meets the analysis requirements. Including data quality inspection, data quality scoring and data quality report generation. After the above pre-processing steps, the processed data is integrated into a new data set, namely the integrated real-time data set. This data set will serve as the basis for subsequent risk assessment and management decisions. Through the above process, the data collected from several data sources in the chemical park in real time is converted into a high-quality, unified format integrated real-time data set, which provides reliable data support for subsequent real-time risk assessment and management decisions.

[0094] Step 3: Perform multi-source fusion processing on the integrated real-time data set and the integrated historical data set to obtain a fused data set;

[0095] Assume that the feature vector of the integrated real-time data set is as shown in formula (6):

[0096]

[0097] Assume that the feature vector of the integrated historical data set is as shown in the following formula (7):

[0098]

[0099] Then the feature vector of the fused data set is as shown in formula (8):

[0100]

[0101] Among them, select represents the feature selection function, and transform represents the feature transformation function;

[0102] [X real-time ,X historical ] indicates direct concatenation, which concatenates the feature vectors in the real-time data set and the historical data set into a new vector;

[0103] [select(X real-time ),select(X historical )] indicates that the features are selected and then stitched together. Before stitching, some features are selected for stitching according to their importance;

[0104] transform([X real-time ,X historical ]) indicates concatenation after transformation. For features with high correlation or redundancy, concatenation is performed after dimensionality reduction.

[0105] Step 4: Based on the fused data set, conduct real-time risk assessment through the risk assessment model, and make management decisions based on the assessment results;

[0106] After the real-time data set and the historical data set are integrated through multi-source fusion processing to obtain a fused data set, the data set is input into the risk assessment model. The model will calculate and output real-time risk assessment results based on the characteristics in the data, such as environmental monitoring parameters, process equipment status, personnel activities, inventory levels, energy consumption, etc., and the relationship between them.

[0107] Risk assessment results may include, but are not limited to:

[0108] Risk level: such as low, medium, high or a specific numerical score.

[0109] Risk types: such as environmental risks, production safety risks, personnel safety risks, etc.

[0110] Risk Source: Identify the specific factors or areas that cause risk.

[0111] Risk trends: predict the future development trend of risks.

[0112] Analyze the assessment results. After obtaining the risk assessment results, you need to conduct an in-depth analysis of these results to understand the nature, severity and possible impact of the risks. This may require a comprehensive judgment based on the actual situation of the park, historical experience and professional knowledge.

[0113] Make management decisions. Based on the analysis of risk assessment results, make specific management decisions. Management decisions may include but are not limited to:

[0114] Preventive measures: Develop preventive measures for potential risk sources to reduce the possibility of risk occurrence.

[0115] Emergency response plan: When risks occur, develop a detailed emergency response plan to quickly control the situation and reduce losses.

[0116] Resource allocation: Based on the risk assessment results, rationally allocate human, material and financial resources to ensure the effectiveness of risk management.

[0117] Supervision and inspection: Strengthen supervision and inspection of high-risk areas or activities to promptly identify and eliminate safety hazards.

[0118] Step 5. Execute management decisions, send instructions to relevant personnel and corresponding equipment through information transmission equipment, and the relevant personnel and corresponding equipment complete the corresponding tasks according to the instructions; establish a feedback mechanism to collect information and data from the relevant personnel and corresponding equipment during the task execution process, and continuously optimize and update the risk assessment model based on the information and data during the task execution process that are fed back, and optimize and improve the management decision-making process and strategy according to the execution effect of the management decision.

[0119] Example 2

[0120] A chemical park intelligent management system for realizing a chemical park intelligent management method, comprising:

[0121] The data collection and integration module 101 is used to collect data from several data sources and perform preprocessing and integration, provide a data visualization interface, and display real-time and historical data. This module is responsible for real-time data collection from multiple data sources, such as environmental monitoring equipment, process equipment sensors, inventory management systems, safety management systems, etc. Data collection methods may include API interface calls, IoT protocol subscriptions, database queries, etc. The collected raw data is preprocessed by cleaning, verification, format conversion, etc. to ensure data quality and consistency; the preprocessed data is integrated into a unified data set for subsequent analysis and processing. At the same time, this module is also responsible for the storage and management of historical data; it provides an intuitive data visualization interface to display real-time and historical data, helping users quickly understand the operation status of the park.

[0122] The model building module 102 is used to build a risk assessment model and train the risk assessment model to obtain a risk assessment model that meets the system requirements; design a suitable risk assessment model framework according to the characteristics and business needs of the chemical park; use historical data sets to train the risk assessment model, and through algorithm optimization and parameter adjustment, enable the model to accurately assess the risk situation of the park; use a verification data set to verify the trained model to ensure its accuracy and reliability; with the continuous accumulation of new data, regularly update and optimize the model to adapt to changes in the park environment and business needs.

[0123] The risk assessment module 103 is used for risk assessment and provides real-time display and query functions of risk assessment results; displays risk assessment results in the form of graphics, reports, etc., including information such as risk level, risk type, and risk source; provides a query function for risk assessment results, and users can query historical risk assessment records based on time, region, risk type, and other conditions.

[0124] The management decision-making module 104 is used to make management decisions based on risk assessment results and prediction rules; automatically or semi-automatically make management decisions based on risk assessment results and decision-making rules; and review the made management decisions to ensure the rationality and feasibility of the decisions.

[0125] The instruction execution and feedback module 105 is used to send instructions to relevant personnel and equipment, monitor the execution status, collect feedback information during the execution process, and evaluate and optimize; monitor the execution status of instructions in real time to ensure that the instructions are executed in a timely and accurate manner; collect feedback information during the execution process, including execution results, problems encountered, improvement suggestions, etc.; evaluate the execution effect, and optimize management decisions and systems based on the feedback information to improve management efficiency and effectiveness.

[0126] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A chemical park intelligent management method, characterized in that: The following steps are involved: Collect historical data in the park through several data sources to obtain an initial historical data set, pre-process the initial historical data set to obtain an integrated historical data set, and build a risk assessment model based on the integrated historical data set; Collecting real-time data in the chemical park from several data sources in real time to obtain an initial real-time data set, and preprocessing the initial real-time data set to obtain an integrated real-time data set; Perform multi-source fusion processing on the integrated real-time data set and the integrated historical data set to obtain a fused data set; Based on the fused data set, real-time risk assessment is performed through the risk assessment model, and management decisions are made based on the assessment results; Execute management decisions, send instructions to relevant personnel and corresponding equipment through information transmission equipment, and the relevant personnel and corresponding equipment complete the corresponding tasks according to the instructions.

2. The intelligent management method for a chemical park according to claim 1, characterized in that: Several of the data sources include: Environmental monitoring equipment, used to monitor environmental parameters within the chemical park; Process equipment status monitoring equipment, used to monitor the operating status of each process equipment in the chemical park; Personnel activity monitoring equipment, used to monitor the activities of personnel in the chemical park; Inventory management systems to provide inventory data on chemical raw materials, products and waste; Energy management systems to monitor energy consumption; Safety management system, used to record safety accidents, hidden danger inspections, safety training and related data.

3. The intelligent management method for a chemical park according to claim 1, characterized in that: The specific process of building a risk assessment model based on integrating historical data sets is as follows: Extract risk features from the integrated historical data set to obtain a risk feature data set; Set risk level classification standards and set corresponding labels for each risk level. According to the risk level classification standards and the labels corresponding to each risk level, perform sample annotation on the risk feature data set and verify the accuracy of the labels to obtain a sample annotation set. The sample annotation set is divided into training set, validation set and test set. Based on the training set, validation set and test set, a risk assessment model is constructed through machine learning algorithm.

4. The intelligent management method for a chemical park according to claim 3, characterized in that: The specific process of constructing the risk assessment model through machine learning algorithm is as follows: Let X be a sample in the training set, X contains features, as shown in the following formula (1): X={x1,x2,…,x K }(1) The output of the risk assessment model is the predicted risk level corresponding to the sample The expression of the risk assessment model is shown in the following formula (2): Where M is the total number of decision trees, T m (X) is the prediction result of the mth decision tree for sample X; The objective function is constructed based on the fused data set, and the risk assessment model is optimized through the objective function.

5. The intelligent management method for a chemical park according to claim 4, characterized in that: The specific process of constructing the objective function based on the fusion data set and optimizing the risk assessment model through the objective function is as follows: The objective function is constructed based on the fused data set. The expression of the objective function is shown in the following formula (3): Among them, n is the number of training samples, y i is the true risk level of the i-th sample, is the predicted risk level of the i-th sample, is the loss function, Ω(T m ) is the regularization term of the mth decision tree.

6. The intelligent management method for a chemical park according to claim 5, characterized in that: The specific expression of the loss function is shown in the following formula (4): Where D is the total number of categories, y id is an indicator, which is 1 when the i-th sample belongs to the d-th class, otherwise it is 0. is the prediction score of the ith sample in the dth category, and j is a category index.

7. The intelligent management method for a chemical park according to claim 5, characterized in that: The specific expression of the regularization term is shown in the following formula (5): Among them, G is the number of leaf nodes of the decision tree, ω k is the weight of the kth leaf node, and γ and λ are both regularization coefficients.

8. The intelligent management method for a chemical park according to claim 1, characterized in that: The specific process of performing multi-source fusion processing on the integrated real-time data set and the integrated historical data set to obtain a fused data set is as follows: Assume that the feature vector of the integrated real-time data set is as shown in formula (6): Assume that the feature vector of the integrated historical data set is as shown in the following formula (7): Then the feature vector of the fused data set is as shown in formula (8): Among them, select represents the feature selection function, and transform represents the feature transformation function; [X real-time ,X historical ] indicates direct concatenation, which concatenates the feature vectors in the real-time data set and the historical data set into a new vector; [select(X real-time ),select(X historical )] indicates that the features are selected and then stitched together. Before stitching, some features are selected for stitching according to their importance; transform([X real-time ,X historical ]) indicates concatenation after transformation. For features with high correlation or redundancy, concatenation is performed after dimensionality reduction.

9. The intelligent management method for a chemical park according to claim 1, characterized in that: Establish a feedback mechanism to collect information and data from the relevant personnel and corresponding equipment during the task execution process, continuously optimize and update the risk assessment model based on the feedback information and data from the task execution process, and optimize and improve the management decision-making process and strategy according to the execution effect of the management decision.

10. A chemical park intelligent management system implementing the method described in any one of claims 1 to 9, characterized in that: include: Data collection and integration module, used to collect data from several data sources and perform preprocessing and integration, provide a data visualization interface, and display real-time and historical data; A model building module is used to build a risk assessment model and train the risk assessment model to obtain a risk assessment model that meets system requirements; The risk assessment module is used for risk assessment and provides real-time display and query functions of risk assessment results; Management decision making module, used to make management decisions based on risk assessment results and prediction rules; The instruction execution and feedback module is used to send instructions to relevant personnel and equipment, monitor the execution status, collect feedback information during the execution process, and conduct evaluation and optimization.

Images