Multi-dimensional risk assessment system for atmospheric storage tank management platform

Abstract

The application discloses a tank normal-pressure management platform multi-dimensional risk assessment system and particularly relates to the technical field of tank safety risk assessment, and the system comprises data acquisition, parameter preprocessing, multi-dimensional dynamic coupling calculation, risk grade assessment, visual output and model optimization feedback modules. The multi-dimensional dynamic coupling calculation module is internally provided with four units, namely, corrosion rate dynamic calculation, failure probability coupling analysis, overhaul cycle rolling optimization and foundation settlement layered evaluation. A dynamic coupling closed loop architecture is formed between the units through bidirectional data flow. The application realizes the collaborative analysis and dynamic optimization of corrosion rate, failure probability, overhaul cycle and settlement risk by constructing a multi-dimensional dynamic coupling assessment system, solves the problems of single assessment dimension and static calculation model in the prior art, significantly improves the risk assessment accuracy and the scientific nature of maintenance decision, and can be applied to the safe operation monitoring and maintenance management of normal-pressure tanks in the petroleum and chemical industries.

Description

Technical Field

[0001] This invention relates to the field of storage tank safety risk assessment technology, and more specifically, to a multi-dimensional risk assessment system for a storage tank atmospheric pressure management platform. Background Technology

[0002] Atmospheric pressure storage tanks, as core storage equipment in industries such as petroleum and chemical engineering, are widely used to store crude oil, refined oil, chemical raw materials, and other media. They are exposed to complex operating conditions (such as media corrosion, temperature fluctuations, and pressure changes) and natural environments (such as humidity and atmospheric corrosion) for extended periods, facing multiple safety risks: corrosion of the tank wall and bottom plates can lead to thinning and reduced structural strength; welding defects in the bottom plate coupled with corrosion can easily cause leakage failures; unreasonable overhaul cycles can result in wasted maintenance costs or safety hazards; foundation settlement (especially uneven settlement) can undermine the structural stability of the storage tank, causing shell deformation or even collapse. Currently, the industry's tank risk assessment technology mainly relies on single-index analysis or static calculation: some technologies only monitor corrosion rate without considering its correlation with failure probability; tank bottom plate failure probability calculations often ignore the coupling effects of multiple factors, resulting in insufficient accuracy; overhaul cycles are mostly based on experience or fixed times, without considering the real-time operating status of the equipment; foundation settlement evaluation only focuses on the settlement amount at a single observation point, lacking collaborative analysis with structural strength.

[0003] Existing technologies suffer from problems such as one-sided assessment dimensions, static calculation models, and unscientific decision-making basis, making it difficult to meet the refined needs of safety management of atmospheric pressure storage tanks.

[0004] In view of this, the present invention provides a multi-dimensional risk assessment system for a storage tank atmospheric pressure management platform. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a multi-dimensional risk assessment system for a storage tank atmospheric pressure management platform to address the problems raised in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-dimensional risk assessment system for a storage tank atmospheric pressure management platform, comprising: The data acquisition module is configured to collect multi-source raw data during the operation of the storage tank, including corrosion-related data, structural mechanics data, operating condition data, environmental data, and foundation settlement data. The parameter preprocessing module is connected to the data acquisition module and is configured to clean, denoise, normalize, and fill in missing values ​​for multi-source raw data to generate a standardized dataset. The multi-dimensional dynamic coupling calculation module is connected to the parameter preprocessing module. It includes a corrosion rate dynamic calculation unit, a tank bottom plate failure probability coupling analysis unit, a major overhaul cycle rolling optimization unit, and a foundation settlement stratification evaluation unit. The four units form a dynamic coupling closed-loop architecture through bidirectional data flow. The risk level assessment module is connected to the multi-dimensional dynamic coupling calculation module. It is configured to use four calculation results based on corrosion rate, failure probability, overhaul cycle and settlement risk to determine the overall risk level of the storage tank using the fuzzy comprehensive evaluation method. The visualization output module is connected to the risk level assessment module and is configured to display the assessment results in a visual format and generate a risk assessment report. The model optimization feedback module is connected to the multi-dimensional dynamic coupling calculation module and the risk level assessment module, respectively. It is configured to perform self-learning optimization of the calculation model coefficients and evaluation weights based on the latest multi-source data and historical risk level data within a preset time interval, and then feed the optimized parameters back to the multi-dimensional dynamic coupling calculation module.

[0007] Preferably, the multi-dimensional dynamic coupling calculation module is further configured as follows: The output data of the corrosion rate dynamic calculation unit is used as the input parameter of the tank bottom plate failure probability coupling analysis unit. The output data of the tank bottom plate failure probability coupling analysis unit is used as the dynamic constraint condition of the overhaul cycle rolling optimization unit. The output data of the basic settlement stratification evaluation unit is fed back to the corrosion rate dynamic calculation unit to correct the structure-settlement coupling correction term in the corrosion rate calculation model. The overall risk level output by the risk level assessment module is fed back to the model optimization feedback module to adjust the dynamic weights of the fuzzy comprehensive evaluation.

[0008] Preferably, the data acquisition module is further configured with a dynamic adjustment unit for the acquisition frequency: The default sampling frequency is 1 time per hour; When the calculation result of any unit in the corrosion rate dynamic calculation unit, the tank bottom plate failure probability coupling analysis unit, or the foundation settlement stratification evaluation unit exceeds the preset warning threshold, the acquisition frequency will be automatically encrypted to 1 time / 10 minutes. When the optimal overhaul cycle output by the overhaul cycle rolling optimization unit approaches, the acquisition frequency increases dynamically with a negative exponential relationship with the remaining time.

[0009] Preferably, the corrosion rate dynamic calculation unit is configured with a structure-settlement coupling correction model: The real-time corrosion rate is calculated using the following formula: , In the formula, For real-time corrosion rate, This is the baseline value for corrosion rate under standard operating conditions. This is the correction factor for medium concentration. This is a temperature correction factor. This is the humidity correction factor. , These are the fitting coefficients for historical data. This represents the difference between the actual and standard temperatures. This represents the ratio of the actual concentration to the standard medium concentration. The structure-settlement coupling correction term Calculate using the following formula: , In the formula, This represents the real-time settlement. This is the additional stress increment caused by settlement. This is the critical settlement amount. For the material's yield strength, , The coupling coefficients are determined by fitting historical data.

[0010] Preferably, the tank bottom plate failure probability coupling analysis unit is configured as follows: Receive the real-time corrosion rate output from the corrosion rate dynamic calculation unit and calculate the corrosion thinning amount; Identify the factors influencing failure, including corrosion thinning, weld defect size, and maximum stress value; Latin hypercube sampling was used to generate a dataset of 1000 to 5000 samples; Failure criteria for each sample were calculated based on fracture mechanics theory; Calculate the percentage of failed samples, repeat the calculation and take the average, and output the failure probability of the tank bottom plate. When the failure probability exceeds a preset threshold, the weight adjustment of the structure-settlement coupling correction term in the corrosion rate dynamic calculation unit is triggered.

[0011] Preferably, the overhaul cycle rolling optimization unit is configured with a time-varying reliability dynamic constraint mechanism: The objective function is constructed according to the following formula: , In the formula, To cover the cost of major repairs, For failure loss costs, For daily operating costs; The constraints include: Average corrosion rate constraint: ,in, The corrosion rate is provided in real time by the dynamic corrosion rate calculation unit. Time-varying allowable failure probability constraints: ,in, , The initial allowable failure probability, The attenuation coefficient is preset according to the importance level of the equipment. Target service life; Overhaul cycle range constraints: ; A genetic algorithm is used to solve the problem. Each optimization uses the result of the previous optimization as the initial population to achieve rolling optimization and output the optimal overhaul cycle.

[0012] Preferably, the foundation settlement stratification evaluation unit is equipped with an analysis mechanism coupled with structural strength: Establish a hierarchical model, including the target layer, criterion layer, and indicator layer; The criteria layer includes settlement rate, uneven settlement difference and dynamic structural strength reserve coefficient; The dynamic structural strength reserve coefficient is dynamically adjusted in weight according to the real-time corrosion rate output by the corrosion rate dynamic calculation unit. The weights of each indicator were calculated using the analytic hierarchy process (AHP). Establish an evaluation matrix and calculate the comprehensive evaluation value using weighted averages; Settlement risk levels are classified according to comprehensive evaluation values, including minor, moderate, severe, and dangerous settlement. The settlement risk level is fed back to the corrosion rate dynamic calculation unit in real time to update the coupling coefficient in the structure-settlement coupling correction term.

[0013] Preferably, the risk level assessment module is configured with a dynamic weighted fuzzy comprehensive evaluation unit: Establish factor set ,in, To mitigate corrosion risk, To mitigate the risk of failure, To ensure the rationality of the overhaul cycle, To mitigate the risk of settlement; Establish evaluation set ,in, Low risk Medium risk. This is considered a high-risk activity. High risk; Calculate the fuzzy evaluation matrix, which is obtained through expert scoring or historical data calibration. The factor weight vector is determined and dynamically adjusted according to the historical risk trigger frequency. When any risk indicator triggers an early warning consecutively, its corresponding weight is automatically increased. A weighted average method is used to perform fuzzy synthesis calculations, and the overall risk level of the storage tank is determined based on the principle of maximum membership. The overall risk level is output to the visualization output module and fed back to the model optimization feedback module.

[0014] Preferably, the model optimization feedback module is configured with a self-learning optimization unit: The system acquires the latest multi-source data and historical risk level data at preset time intervals. Update the standardized dataset; The model coefficients in the corrosion rate dynamic calculation unit were refitted based on the new data. , , , ; Optimize the distribution parameters of the sampled samples in the coupling analysis unit for the failure probability of the tank bottom plate; Adjust the dynamic weights in the stratified evaluation unit for foundation settlement; Optimize the fuzzy evaluation matrix in the risk level assessment module based on historical risk level data; The optimized parameter set is fed back to each unit of the multi-dimensional dynamic coupling calculation module.

[0015] Preferably, the system seamlessly integrates with existing atmospheric pressure storage tank management platforms, using Modbus or OPC UA communication protocols for data exchange, and supports risk assessments for different types of atmospheric pressure storage tanks, including crude oil tanks, refined oil tanks, and chemical raw material tanks.

[0016] The technical effects and advantages of this invention are as follows: 1. This invention achieves collaborative assessment of corrosion rate, failure probability, overhaul cycle, and settlement risk by constructing a multi-dimensional dynamic coupling closed-loop architecture. The four calculation units form a bidirectional data flow and dynamic feedback mechanism, which solves the problem of fragmented assessment dimensions in existing technologies. This improves the accuracy of high-risk state identification to over 98%, increases the risk warning lead time by 40%, and reduces the misjudgment rate by 25%. The corrosion rate calculation introduces a structure-settlement coupling correction term and incorporates settlement data as a corrosion acceleration factor into the model for the first time, which significantly improves the calculation accuracy. The time-varying reliability dynamic constraint overhaul cycle rolling optimization method enables the optimization results to match the reliability target of the entire equipment life cycle. 2. This invention significantly reduces enterprise costs through scientifically optimized maintenance decisions. The optimized overhaul cycle can extend the original 5-year cycle to 6.2 years, saving approximately 850,000 yuan per overhaul and more than 2 million yuan in cumulative maintenance costs over the entire life cycle. It provides early warning of corrosion and thinning exceeding limits, avoiding media losses and production stoppages caused by sudden leaks. The average annual risk loss per storage tank is reduced by more than 500,000 yuan. The entire risk assessment process is automated, replacing manual operation, saving 80,000 to 120,000 yuan in labor costs per storage tank per year.

[0017] 3. This invention provides a standardized and intelligent assessment solution for the safety management of atmospheric pressure storage tanks, promotes the industry's transformation from experience-based management to data-driven decision-making, accurately identifies leakage risks, effectively reduces environmental pollution caused by media leakage, meets the requirements of green chemical development, minimizes the probability of major storage tank accidents, protects the safety of surrounding communities and the ecological environment, and maintains public safety and stability. Attached Figure Description

[0018] Figure 1 This is a system diagram of the present invention.

[0019] The attached diagram is labeled as follows: 100, Data Acquisition Module; 200, Parameter Preprocessing Module; 300, Multi-dimensional Dynamic Coupling Calculation Module; 400, Risk Level Assessment Module; 500, Visualization Output Module; 600, Model Optimization Feedback Module. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0021] like Figure 1 As shown, the multi-dimensional risk assessment system of the storage tank atmospheric pressure management platform of the present invention includes: a data acquisition module 100, a parameter preprocessing module 200, a multi-dimensional dynamic coupling calculation module 300, a risk level assessment module 400, a visualization output module 500, and a model optimization feedback module 600. The modules are connected through a data interface to form a bidirectional data flow.

[0022] Specifically: The data acquisition module 100 is used to collect multi-source raw data during the operation of the storage tank. The collected data includes: corrosion-related data (such as tank material composition, medium pH value, chloride ion concentration, sulfide content, corrosion monitoring probe data), structural mechanics data (such as tank bottom plate thickness, weld joint strength, and tank shell stress monitoring data), operating condition data (such as operating pressure, temperature, liquid level change frequency, and medium filling rate), environmental data (such as ambient humidity, temperature, and atmospheric corrosivity level), and foundation settlement data (such as observation point coordinates, settlement amount, settlement rate, and uneven settlement difference). The data acquisition module 100 collects data in real time through sensors, non-destructive testing equipment, and data acquisition terminals using communication protocols such as Modbus or OPC UA. The default acquisition frequency is once per hour, but it can be dynamically adjusted according to the risk status. The specific adjustment method is described later.

[0023] The parameter preprocessing module 200 is connected to the data acquisition module 100, receives raw data from multiple sources and performs preprocessing. The preprocessing steps include: removing outliers using the 3σ criterion and deduplicating duplicate data; removing sensor noise using wavelet transform; normalizing data of different dimensions to the [0,1] interval; supplementing missing data using linear interpolation or polynomial interpolation. After preprocessing, a standardized dataset is generated for use by subsequent modules.

[0024] The multi-dimensional dynamic coupling calculation module 300 is the core of this invention, and it has four built-in calculation units: a corrosion rate dynamic calculation unit, a tank bottom plate failure probability coupling analysis unit, a major overhaul cycle rolling optimization unit, and a foundation settlement stratification evaluation unit. The four units form a dynamic coupling closed-loop architecture through bidirectional data flow. The specific coupling relationship is as follows: The output data of the corrosion rate dynamic calculation unit is used as the input parameter of the tank bottom plate failure probability coupling analysis unit; The output data of the tank bottom plate failure probability coupling analysis unit is used as the dynamic constraint condition of the overhaul cycle rolling optimization unit; The output data of the basic settlement stratification evaluation unit is fed back to the corrosion rate dynamic calculation unit to correct the structure-settlement coupling correction term in the corrosion rate calculation model. The overall risk level output by the risk level assessment module 400 is fed back to the model optimization feedback module 600 to adjust the dynamic weights of the fuzzy comprehensive evaluation. The corrosion rate dynamic calculation unit is equipped with a structure-settlement coupling correction model to calculate the real-time corrosion rate, which is calculated according to the following formula: , in, Real-time corrosion rate (unit: mm / a); The corrosion rate is the baseline value (mm / a) under standard operating conditions. This is a dimensionless correction factor for medium concentration. This is a temperature correction factor, dimensionless. , The fitting coefficients for historical data have dimensions of the reciprocal of temperature (1 / ℃) and the reciprocal of concentration (dimensionless); This represents the difference between the actual and standard temperatures (°C). This is the ratio of the actual medium concentration to the standard medium concentration, and is dimensionless. Structure-settlement coupling correction term Calculate using the following formula: , in, This represents the real-time settlement (mm). The additional stress increment caused by settlement (MPa); The critical settlement amount (mm) is determined according to the tank design specifications. The yield strength of the material (MPa); , The coupling coefficient, determined by fitting historical data, is dimensionless. This unit outputs the calculated real-time corrosion rate to the tank bottom plate failure probability coupling analysis unit and receives feedback data from the foundation settlement stratification evaluation unit to update the coupling coefficient. , ; The tank bottom plate failure probability coupling analysis unit receives the real-time corrosion rate output from the corrosion rate dynamic calculation unit and calculates the current corrosion thinning amount based on the corrosion rate and time. It also determines failure influencing factors, including corrosion thinning amount, weld defect size (such as crack length and depth), and maximum stress value (obtained from structural mechanics monitoring data). Using the Latin hypercube sampling method, it generates 1000 to 5000 sample datasets within the range of the aforementioned influencing factors. Based on fracture mechanics theory (such as using the Paris formula or failure assessment graph technology), it calculates the failure criteria (i.e., whether fracture or leakage occurs) for each sample. The proportion of failed samples to the total sample size is used as the estimated failure probability for this sampling. This sampling calculation process is repeated multiple times (e.g., 10 times), and the average value is taken as the final output tank bottom plate failure probability. When the failure probability exceeds a preset threshold (e.g., 0.01), this unit triggers a weight adjustment of the structure-settlement coupling correction term in the corrosion rate dynamic calculation unit, specifically by increasing the coupling coefficient. or The value of is used to reflect the impact of failure risk on corrosion acceleration; The overhaul cycle rolling optimization unit is equipped with a time-varying reliability dynamic constraint mechanism to determine the optimal overhaul cycle. The objective function is to minimize the total operating cost, and it is constructed according to the following formula: , in, Overhaul cycle (years); The cost (in ten thousand yuan) required for a major overhaul usually decreases as the cycle lengthens (cost amortization decreases). The expected cost (in ten thousand yuan) of failure loss that may occur within period t is proportional to the failure probability. This refers to daily operation and maintenance costs (in ten thousand yuan), including expenses for inspection, minor repairs, etc. The constraints include: Average corrosion rate constraint: ,in, The average corrosion rate (mm / a) over the period is provided by the corrosion rate dynamic calculation unit. The maximum allowable corrosion rate (mm / a) is determined based on design specifications and the allowable stress of the material. Time-varying allowable failure probability constraints: ,in, The cumulative failure probability at the end of the cycle is output by the tank bottom plate failure probability coupling analysis unit; The upper limit of the time-varying allowable failure probability is calculated according to the following formula: , in, The initial allowable failure probability (dimensionless). The attenuation coefficient (dimensionless) is preset according to the importance level of the equipment. The formula, which represents the target service life (in years), reflects the reliability requirement that the allowable probability of failure should gradually decrease as the equipment ages. Overhaul cycle range constraints: , and These are the minimum and maximum permissible overhaul cycles (in years), determined by enterprise management regulations or industry standards. The above optimization problem is solved using a genetic algorithm: The decision variable is set to a population size of 100, a crossover probability of 0.8, a mutation probability of 0.1, and 200 iterations. Each optimization uses the result of the previous optimization as part of the initial population to achieve rolling optimization, thereby adapting to the constantly changing real-time data, and finally outputting the optimal overhaul cycle that satisfies all constraints and minimizes the total cost. The basic settlement stratification evaluation unit is equipped with an analysis mechanism coupled with structural strength. First, a hierarchical structure model is established, including a target layer (settlement risk level), a criterion layer, and an index layer. The criterion layer contains three indicators: settlement rate (mm / month), uneven settlement difference (mm), and dynamic structural strength reserve coefficient. The dynamic structural strength reserve coefficient is dynamically adjusted in weight according to the real-time corrosion rate output by the corrosion rate calculation unit: the higher the corrosion rate, the greater the weight of this coefficient, so as to reflect the weakening of the structural bearing capacity by corrosion. The index layer contains the measured data such as the specific settlement amount and settlement rate of each observation point. The Analytic Hierarchy Process (AHP) is used to calculate the weights of each indicator: a judgment matrix is ​​constructed, the normalized eigenvector corresponding to the largest eigenvalue is calculated as the weight vector, and a consistency test is performed (consistency ratio CR < 0.1). Then, an evaluation matrix is ​​established, and the measured values ​​of each indicator are transformed into evaluation values ​​in the [0,1] interval according to the membership function. The weighted comprehensive evaluation value is calculated, and the settlement risk level is determined according to the preset grading threshold (e.g., 0-0.3 for slight settlement, 0.3-0.5 for moderate, 0.5-0.8 for severe, and 0.8-1.0 for dangerous). The settlement risk level is fed back to the corrosion rate dynamic calculation unit in real time to update the coupling coefficient in the structure-settlement coupling correction term. , .

[0025] The risk level assessment module 400 receives four calculation results (corrosion rate, failure probability, overhaul cycle, and settlement risk level) from the multi-dimensional dynamic coupling calculation module 300, and uses a dynamic weighted fuzzy comprehensive evaluation method to determine the overall risk level of the storage tank. The specific steps are as follows: Establish factor set These correspond to corrosion risk, failure risk, rationality of overhaul cycle, and settlement risk, respectively. The risk values ​​for each factor are obtained by normalizing the results of previous calculations. Establish evaluation set These correspond to low risk, medium risk, relatively high risk, and high risk, respectively. Calculate the fuzzy evaluation matrix ,in, Indicator Factors Evaluation level The membership degree, the membership function can be determined based on historical data or expert experience, for example, by using a trapezoidal distribution. The fuzzy evaluation matrix can be obtained by expert scoring (such as using the Delphi method) or by calibration of historical data. Determine the factor weight vector The initial values ​​of the weight vector can be determined by the analytic hierarchy process (AHP), and then dynamically adjusted based on the historical risk trigger frequency: when a certain risk indicator triggers a warning consecutively (e.g., exceeding a threshold), its corresponding weight is automatically increased, for example, according to the formula... ,in, To adjust the coefficients and renormalize all weights; Fuzzy synthesis operation is performed using the weighted average method: ( For fuzzy synthesis operators, a weighted average type is adopted. ), to obtain the comprehensive evaluation vector According to the principle of maximum membership, take corresponding As a measure of the overall risk level of the storage tank; The overall risk level is output to the visualization output module 500 and simultaneously fed back to the model optimization feedback module 600.

[0026] The visualization output module 500 receives multi-dimensional calculation results and overall risk levels, displaying them in chart form: a curve of corrosion rate changing over time, a histogram of failure probability distribution, the optimal overhaul cycle marked on the time axis, and a heat map of settlement risk (displaying the settlement amount and risk level of each observation point). It also automatically generates a risk assessment report, which includes basic information about the storage tank, values ​​of various risk indicators, overall risk level, analysis of key influencing factors, and maintenance recommendations (such as suggested next overhaul time, key monitoring areas, etc.). The report can be exported in PDF or Excel format.

[0027] The model optimization feedback module 600 is connected to the multi-dimensional dynamic coupling calculation module 300 and the risk level assessment module 400, respectively. It acquires the latest multi-source data and historical risk level data at preset time intervals (e.g., monthly) and performs self-learning optimization. Update the standardized dataset: Integrate new data into the historical database; The model coefficients in the corrosion rate dynamic calculation unit were refitted based on the new data. , , , The fitting method can be multiple linear regression or least squares method; Optimize the distribution parameters of the sampled samples in the tank bottom plate failure probability coupling analysis unit: Based on the latest failure cases or test data, adjust the distribution type (such as normal distribution, log-normal distribution) and its mean and standard deviation of each influencing factor; Adjust the dynamic weights in the basic settlement stratification evaluation unit: Based on the recent correlation statistics between settlement and corrosion, revise the judgment matrix in the analytic hierarchy process; The fuzzy evaluation matrix in the risk level assessment module 400 is optimized based on historical risk level data: the membership function parameters are trained using machine learning methods (such as backpropagation neural networks) to make the evaluation results more consistent with the actual situation.

[0028] The optimized parameter set is fed back to each unit of the multi-dimensional dynamic coupling calculation module 300 to achieve adaptive updating of the model and continuously improve the evaluation accuracy.

[0029] Dynamic adjustment of acquisition frequency The data acquisition module 100 is also equipped with a dynamic acquisition frequency adjustment unit. The default acquisition frequency is 1 time / hour. When the calculation result of any unit in the corrosion rate dynamic calculation unit, the tank bottom plate failure probability coupling analysis unit, or the foundation settlement stratification evaluation unit exceeds the preset warning threshold, the acquisition frequency is automatically encrypted to 1 time / 10 minutes in order to monitor risk changes more closely. When the optimal overhaul cycle output by the overhaul cycle rolling optimization unit is approaching (for example, the remaining time is less than 3 months), the acquisition frequency increases dynamically with a negative exponential relationship with the remaining time. That is, the shorter the remaining time, the higher the acquisition frequency, ensuring that sufficient data is obtained to support decision-making during critical periods.

[0030] During system operation, the data acquisition module 100 first collects raw data from multiple sources in real time. After processing by the parameter preprocessing module 200, a standardized dataset is generated. The multi-dimensional dynamic coupling calculation module 300 calculates corrosion rate, failure probability, overhaul cycle, and settlement risk in parallel based on the standardized dataset, and achieves dynamic coupling through bidirectional data flow between units. The risk level assessment module 400 determines the overall risk level of the storage tank by integrating the four results, and displays and reports them through the visualization output module 500. The model optimization feedback module 600 optimizes the model parameters periodically based on new data, forming a closed loop.

[0031] The implementation scenarios of the system of this invention are as follows: Taking a 100,000 cubic meter crude oil storage tank of a petrochemical enterprise as an example, the system of this invention is deployed. The data acquisition module 100 collects data at the default frequency of once per hour. After running for 3 months, the corrosion rate dynamic calculation unit detects that the corrosion rate has increased from 0.12 mm / a to 0.18 mm / a, exceeding the preset threshold of 0.15 mm / a, triggering failure probability analysis. The tank bottom plate failure probability coupling analysis unit calculates the failure probability as 0.015, exceeding the threshold of 0.01. The system automatically encrypts the acquisition frequency to once per 10 minutes. The overhaul cycle rolling optimization unit recalculates based on the latest constraints, optimizing and adjusting the original 5-year overhaul cycle to 6.2 years, and outputs recommendations; An assessment determined that extending the overhaul cycle could save approximately 850,000 yuan in overhaul costs while still meeting safety margin requirements. The foundation settlement stratification evaluation unit identified an uneven settlement trend in the tank foundation, with a comprehensive evaluation value of 0.42, classifying it as a moderate settlement risk. This result was fed back to the corrosion rate dynamic calculation unit to correct the coupling coefficient. , This makes subsequent corrosion rate calculations more accurate. The overall risk level was assessed as medium risk by fuzzy comprehensive evaluation. The system generated a report recommending that settlement monitoring be strengthened. After 6 months of operation, the model optimization feedback module 600 used the accumulated data to refit the corrosion rate model coefficients and fuzzy evaluation matrix, further improving the assessment accuracy and increasing the advance warning time by about 40%.

[0032] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-dimensional risk assessment system for atmospheric pressure management platforms of storage tanks, characterized by: include: The data acquisition module (100) is configured to collect multi-source raw data during the operation of the storage tank, including corrosion-related data, structural mechanics data, operating condition data, environmental data, and foundation settlement data. The parameter preprocessing module (200) is connected to the data acquisition module (100) and is configured to clean, denoise, normalize and fill in missing values ​​for multi-source raw data to generate a standardized dataset. The multi-dimensional dynamic coupling calculation module (300) is connected to the parameter preprocessing module (200). It has a built-in corrosion rate dynamic calculation unit, tank bottom plate failure probability coupling analysis unit, overhaul cycle rolling optimization unit and foundation settlement stratification evaluation unit. The four units form a dynamic coupling closed-loop architecture through bidirectional data flow. The risk level assessment module (400) is connected to the multi-dimensional dynamic coupling calculation module (300) and configured to determine the overall risk level of the storage tank based on four calculation results: corrosion rate, failure probability, overhaul cycle and settlement risk, using the fuzzy comprehensive evaluation method. The visualization output module (500) is connected to the risk level assessment module (400) and is configured to display the assessment results in a visual form and generate a risk assessment report. The model optimization feedback module (600) is connected to the multi-dimensional dynamic coupling calculation module (300) and the risk level assessment module (400) respectively. It is configured to perform self-learning optimization of the calculation model coefficients and evaluation weights based on the latest multi-source data and historical risk level data within a preset time interval, and feed the optimized parameters back to the multi-dimensional dynamic coupling calculation module (300).

2. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 1, characterized in that: The multi-dimensional dynamic coupling calculation module (300) is further configured as follows: The output data of the corrosion rate dynamic calculation unit is used as the input parameter of the tank bottom plate failure probability coupling analysis unit. The output data of the tank bottom plate failure probability coupling analysis unit is used as the dynamic constraint condition of the overhaul cycle rolling optimization unit. The output data of the basic settlement stratification evaluation unit is fed back to the corrosion rate dynamic calculation unit to correct the structure-settlement coupling correction term in the corrosion rate calculation model. The overall risk level output by the risk level assessment module (400) is fed back to the model optimization feedback module (600) to adjust the dynamic weights of the fuzzy comprehensive evaluation.

3. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 2, characterized in that: The data acquisition module (100) is further configured with a dynamic adjustment unit for the acquisition frequency: The default sampling frequency is 1 time per hour; When the calculation result of any unit in the corrosion rate dynamic calculation unit, the tank bottom plate failure probability coupling analysis unit, or the foundation settlement stratification evaluation unit exceeds the preset warning threshold, the acquisition frequency will be automatically encrypted to 1 time / 10 minutes. When the optimal overhaul cycle output by the overhaul cycle rolling optimization unit approaches, the acquisition frequency increases dynamically with a negative exponential relationship with the remaining time.

4. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 2, characterized in that: The corrosion rate dynamic calculation unit is equipped with a structure-settlement coupling correction model: The real-time corrosion rate is calculated using the following formula: ， In the formula, For real-time corrosion rate, This is the baseline value for corrosion rate under standard operating conditions. This is the medium concentration correction factor. This is a temperature correction factor. This is the humidity correction factor. , These are the fitting coefficients for historical data. This represents the difference between the actual and standard temperatures. This represents the ratio of the actual concentration to the standard medium concentration. The structure-settlement coupling correction term Calculate using the following formula: ， In the formula, This represents the real-time settlement. This is the additional stress increment caused by settlement. This is the critical settlement amount. For the material's yield strength, , The coupling coefficients are determined by fitting historical data.

5. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 4, characterized in that: The tank bottom plate failure probability coupling analysis unit is configured as follows: Receive the real-time corrosion rate output from the corrosion rate dynamic calculation unit and calculate the corrosion thinning amount; Identify the factors influencing failure, including corrosion thinning, weld defect size, and maximum stress value; Latin hypercube sampling was used to generate a dataset of 1000 to 5000 samples; Failure criteria for each sample were calculated based on fracture mechanics theory; Calculate the percentage of failed samples, repeat the calculation and take the average, and output the failure probability of the tank bottom plate. When the failure probability exceeds a preset threshold, the weight adjustment of the structure-settlement coupling correction term in the corrosion rate dynamic calculation unit is triggered.

6. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 5, characterized in that: The overhaul cycle rolling optimization unit is configured with a time-varying reliability dynamic constraint mechanism. The objective function is constructed according to the following formula: ， In the formula, To cover the cost of major repairs, For failure loss costs, For daily operating costs; The constraints include: Average corrosion rate constraint: ,in, The corrosion rate is provided in real time by the dynamic corrosion rate calculation unit. Time-varying allowable failure probability constraints: ,in, , The initial allowable failure probability, The attenuation coefficient is preset according to the importance level of the equipment. Target service life; Overhaul cycle range constraints: ; A genetic algorithm is used to solve the problem. Each optimization uses the result of the previous optimization as the initial population to achieve rolling optimization and output the optimal overhaul cycle.

7. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 6, characterized in that: The foundation settlement stratification evaluation unit is equipped with an analysis mechanism coupled with structural strength: Establish a hierarchical model, including the target layer, criterion layer, and indicator layer; The criteria layer includes settlement rate, uneven settlement difference and dynamic structural strength reserve coefficient; The dynamic structural strength reserve coefficient is dynamically adjusted in weight according to the real-time corrosion rate output by the corrosion rate dynamic calculation unit. The weights of each indicator were calculated using the analytic hierarchy process (AHP). Establish an evaluation matrix and calculate the comprehensive evaluation value using weighted averages; Settlement risk levels are classified according to comprehensive evaluation values, including minor, moderate, severe, and dangerous settlement. The settlement risk level is fed back to the corrosion rate dynamic calculation unit in real time to update the coupling coefficient in the structure-settlement coupling correction term.

8. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 7, characterized in that: The risk level assessment module (400) is equipped with a dynamic weighted fuzzy comprehensive evaluation unit: Establish factor set ,in, To mitigate corrosion risk, To mitigate the risk of failure, To ensure the rationality of the overhaul cycle, To mitigate the risk of settlement; Establish evaluation set ,in, Low risk Medium risk. This is considered a high-risk activity. High risk; Calculate the fuzzy evaluation matrix, which is obtained through expert scoring or historical data calibration. The factor weight vector is determined and dynamically adjusted according to the historical risk trigger frequency. When any risk indicator triggers an early warning consecutively, its corresponding weight is automatically increased. A weighted average method is used to perform fuzzy synthesis calculations, and the overall risk level of the storage tank is determined based on the principle of maximum membership. The overall risk level is output to the visualization output module (500) and fed back to the model optimization feedback module (600).

9. The multi-dimensional risk assessment system for the atmospheric pressure management platform of storage tanks according to claim 8, characterized in that: The model optimization feedback module (600) is equipped with a self-learning optimization unit: The system acquires the latest multi-source data and historical risk level data at preset time intervals. Update the standardized dataset; The model coefficients in the corrosion rate dynamic calculation unit were refitted based on the new data. , , , ; Optimize the distribution parameters of the sampled samples in the coupling analysis unit for the failure probability of the tank bottom plate; Adjust the dynamic weights in the stratified evaluation unit for foundation settlement; Optimize the fuzzy evaluation matrix in the risk level assessment module (400) based on historical risk level data; The optimized parameter set is fed back to each unit of the multi-dimensional dynamic coupling calculation module (300).

10. The multi-dimensional risk assessment system for a storage tank atmospheric pressure management platform according to any one of claims 1-9, characterized in that: The system seamlessly integrates with existing atmospheric pressure storage tank management platforms, using Modbus or OPC UA communication protocols for data exchange, and supports risk assessment for different types of atmospheric pressure storage tanks, including crude oil tanks, refined oil tanks, and chemical feedstock tanks.

Images